Reviews: Reviews

Sofirn SC21

Mon, 06 Sep 2021 14:58:00 +0000

Introduction

There is a question that gets asked a lot in the torch world: Which is the best torch?

Many would argue that there is no definitive answer to that, as it depends on the requirements and the usage scenario. The best answer to the question, as it is asked is this: The best torch is the one you have on you when you need it.

To carry a torch at all times is a thing most people do not think about and only after one starts doing so, does one realise how incredibly useful it can be. But to do so, it has to be small enough to be inconspicuous and powerful enough to be useful.

Another requirement for most users, except size and power, is low cost, so it can be affordable and not a big deal if it gets scratched or damaged in the line of duty.

Today's review is about a light that ticks all those boxes, and then some: The Sofirn SC21.

The Sofirn SC21 is a small, powerful and budget friendly light that also features a high CRI emitter and onboard charging. Intrigued? Read on to find out more!

Unboxing

The Sofirn SC21 comes in a generic brown box with a sticker that specifies the contents.

Inside the box we find the extensive manual and the accessories. The manual is in many languages and the accessories include a clip, a lanyard, 2 spare O-rings and a charging cable. The charging cable is USB type A to USB type C and its length is 104cm, including the plugs.

The light itself comes protected in a bubble wrap bag. Inside the bag, along with the light, there is an orange label, explaining that there is an insulator inside the light that prevents the battery from making contact so the light can be shipped safely. That insulator needs to be removed before the light can be used.

The Sofirn SC21 is made of aluminium and features a side button with an LED charge indicator. On the opposite side of the side button there is a rubber flap with the USB logo engraved on it.

The flap can be opened to reveal a USB type C charging port.

The tail is magnetic and the magnet is strong enough to hold the weight of the torch in any orientation. There are also 2 lanyard holes and 2 respective grooves so the lanyard does not compromise the ability of the light to tail stand.

The business end of the Sofirn SC21 features a glass lens protecting an orange peel reflector. The emitter used in the Sofirn SC21 is a Samsung LH351D 5000K 90 CRI LED. It is good to see a neutral white, high CRI emitter used in an EDC torch. I expect this till take a toll on brightness and run times, but it is a sacrifice I am willing to make for better light quality.

The torch unscrews around the middle to reveal the battery compartment, with the battery already installed. An insulator is covering the driver side, preventing the positive terminal of the battery from making contact with it.

The positive terminal of the battery makes contact with the driver PCB through a brass contact point.

The negative terminal makes contact with the body with a good quality spring. The spring is not very thick but it should be more than capable of transferring the power required without significant losses.

The battery that comes with the Sofirn SC21 is a Sofirn branded 16340 Li-Ion button top battery, rated at 800mAh.

The clip can be placed on the back side of the light, as shown below. It allows for lens down deep carry and is bidirectional, which some people like as it allows clipping the light on to a hat and using it as a headlamp, but others dislike as it is not as secure as a unidirectional clip.

The end of the clip drags on the head of the torch when the two parts are unscrewed / screwed so I expect that over time it will damage the finish, unless it is carefully lifted to avoid that.

Quality

The milling quality and anodization are very good, without any sharp edges or visible defects. Even under close inspection the finish and knurling look good.

Specifications

The specifications of the Sofirn SC21, as found on the Sofirn website, are listed in the table below.

The Samsung LH351D 5000K 90 CRI LED is a high light quality choice, rather than a high brightness / high efficiency one and I agree with it. The company claims a maximum output of 1000 Lumen, which is a bold claim. We will test that. The throw is rated at 135m.

User Interface

The user interface of the Sofirn SC21 is designed to please both those that prefer a stepped mode torch and those who like ramping.

Out of the box, the light comes in stepped mode. The stepped mode works as follows:

From OFF single click to turn ON.
Press and hold to cycle through the main stepped modes (Low – Medium – High).
From ON single click to turn OFF.

To change to the ramping mode (or back from the ramping mode to stepped mode), you need to do 4 fast clicks while the light is on.

The ramping mode gives you full flexibility to adjust the brightness steplessly from Moon to Turbo level and works as follows:

From OFF single click to turn ON.
Press and hold to change brightness steplessly (“ramp”).
Ramping changes its direction when the button is pressed again within 1.5 seconds.
The light flashes once when it reaches the lower or upper end of the ramp.
From ON single click to turn OFF.

In either stepped or ramping mode:

From OFF hold for 1 second to turn on at Moonlight level.
Double click to activate Turbo mode from OFF or ON. While in Turbo mode, single click to return to the previously used mode.
Triple click to activate Strobe mode from OFF or ON. While in Strobe mode, single click to return to the previously used standard mode, or press and hold to cycle through SOS - Beacon - Strobe.

The light features both electronic and mechanical lock out.

The electronic lock out works as follows:

From OFF, 4 fast clicks to activate lockout.
Another 4 fast clicks to deactivate lockout and turn the light on at the memorized level.
When the light is locked, the main LED blinks twice when the button is pressed to show the status of being locked.
While in lockout mode, hold the button to use Moonlight mode momentarily.

If you prefer a mechanical lock out, unscrewing the battery tube by 1/4 turn, will break the connection of the negative terminal of the battery to the driver.

The Sofirn SC21 also features mode memory, so it will turn on at the last used level (except Turbo).

Beam-shots

The beam pattern of the Sofirn SC21 is what can be expected from a Samsung LH351D emitter in a shallow, orange peel reflector. It provides a balanced beam with some throw and flood, perfect for EDC.

The hot spot is well defined and large and the spill is uniform without any artefacts, thanks to the OP reflector.

I tested the Sofirn SC21 outside, over a distance of 70m.

The following video shows a comparison of the Sofirn SC21 with the Olight Baton 3 and the Olight S1R Baton II. The Sofirn SC21 offers a more neutral and high CRI beam while the Olight torches have a cooler tint and low CRI but higher output.

Driver

The driver of the Sofirn SC21 provides constant current to the emitter on all modes. There is no PWM that my camera could detect on any output level. The driver also features Thermal Regulation, Reverse Polarity Protection and Low Voltage Protection.

Tint and Size Comparison

The tint of the Sofirn SC21 is neutral white, at 5000K.

In the comparison photo below, you can see the Sofirn SC21 in the middle, compared to the much cooler and greener tints of the Olight Baton 3 on the left and the Olight S1R Baton II on the right. The Samsung LH351D 5000K emitter used in the Sofirn SC21 is high CRI (90). The photo was taken with the white balance set to 5500K.

The length of the Sofirn SC21 is 73mm, which is 10mm longer than the two Olights. That is mostly due to the fact that it uses a reflector and a glass lens instead of the TIR optic used by the two Olights. This offers the advantage that the glass is much harder to scratch than the plastic TIR and does not burn like the plastic can if there is any debris on it, so I actually prefer it.

The USB C port also takes more space than the proprietary magnetic charging that Olight uses. As much as I like the magnetic charging system, there is something to be said for not having to carry around a proprietary cable.

Battery and Charging

The battery included with the Sofirn SC21 is a 16340, rated at 800mAh and I measured it at exactly 788mAh. The light has Low Voltage Protection and turns off when the battery voltage drops to 2.5V which is too low and will wear out the battery, so it should have been set higher. The battery's internal resistance was measured at 79mΩ. These measurements show that a high quality battery is actually included with this light.

The indicative LED on the switch of the Sofirn SC21 shows the level of the battery charge. Green means that the remaining charge is at between 100% and 70%, red that it is below 70% and flashing red that it is critical and the light will soon turn off. I would have preferred at least one intermediate indication between 70% and almost empty.

Charging the Sofirn SC21 is very easy. Just lift the rubber cover and insert the provided (or any) USB type C cable to charge the light. Both USB A to USB C and USB C to USB C cables can be used as well as any charger, including the ones that support PD. This is very convenient as you can charge the Sofirn SC21 with any USB C cable and charger you have at hand.

The indicative LED on the switch flashes red to indicate the light is charging. It turns green when the charging is completed, at 4.21V.

The company advertises that the charging of the Sofirn SC21 takes 1 hour, with a 5V charger, capable of providing 1A.

It actually took 1 hour, 28 minutes and 25 seconds to charge the included battery from 2.5V to 4.21V inside the Sofirn SC21. The maximum current drawn was 0.9541A, so a charger that can provide at least 1A is recommended. A charger is not provided with the light but you can use your phone charger.

A charging current of almost 1A for a 800mAh 16340 battery is rather high and despite being convenient as it charges the battery fast, it will take a toll on the battery longevity. That said, most lights do the same, including the 2 Olights we saw earlier.

Output & Runtimes

The Sofirn SC21 is rated at a maximum output of 1000 Lumen and 135m of throw.

I do not own a multi thousand dollar worth integrating sphere, just a logging Lumen meter and a home made integrating tube. The array is calibrated with 3 separate, professionally measured lights and gives me consistent results, but there is definitely room for error and deviations are to be expected.

The output of all modes as well as the respective run times are shown in the table below.

According to my measurements, Moonlight is 0.5 Lumen (instead of 1), Low is 17 Lumen (instead of 10), Medium is 93 Lumen (instead of 100), high is 323 Lumen (instead of 400) and Turbo is 848 Lumen at turn on (instead of 1000).

The maximum output (at turn on) of 848 Lumen is short of the advertised 1000 by about 15% but still very respectable for the size of the light and especially the fact that it uses a high CRI, neutral white emitter. Output at 30 seconds was 798 Lumen and at 1 minute the output was still 786 Lumen. At 2 minutes the output had decreased to 744 Lumen and then started to decline faster due to thermal regulation to reach reach 206 Lumen at 00:02:51, where it stabilized until the temperature dropped enough and the output rose to 307 Lumen (High) at 00:10:34. It then stayed at that level, with the temperature slowly rising, until at 00:27:54 thermal regulation decreased the output to 254 Lumen. The light remained at that output level until 01:01:45, when the temperature had dropped enough. It then tried to increase the brightness to 307 Lumen, which only happened momentarily as by then the battery did not have enough voltage to support that level of output. Therefore, the light stepped down to 91 Lumen (Medium) for about 6 minutes and then to 17 lumen (Low). It then sustained that output for as long as possible. Unfortunately, the light will not turn off when the battery can no longer sustain the Low output and the brightness just declines with the voltage. It would have been much better if the light just turned off when the battery dropped to that level as that way we would see only regulated output and the battery longevity would be better.

The following graph shows the first 10 minutes of the Turbo Full Runtime Graph in greater detail.

Turbo is good and impressive but is hardly the mode that is actually most used in a torch. Therefore, I decided to make a runtime graph for High.

The graph is self explanatory. The output is stable and regulated. It maintains High output for 58 minutes, then steps down to Medium for 6.5 minutes and then to Low for as long as the battery can sustain it. Again, we see there is no cut off when the battery can no longer sustain the Low output.

The following graph shows the first 10 minutes of the High Full Runtime Graph in greater detail.

The Sofirn SC21 is advertised to produce 4533cd and therefore have a throw of 135m. I measured 3488.16cd which means that the actual throw is 118m, about 12% less than advertised. This is to be expected as the maximum output measured was 848 Lumen instead of the 1000 Lumen advertised, which is about 15% less.

Comparison with Olight Baton 3 & Olight S1R Baton II

I was wondering how much of a disadvantage on brightness and efficiency does the high CRI emitter of the Sofirn SC21 introduce compared to lights that prioritize brightness and efficiency over tint and CRI. So, I decided to compare the above runtime graphs with those of the Olight Baton 3 and the Olight S1R Baton II.

It was clear from the tint comparison that the Sofirn SC21 offers a much more pleasant, less green and neutral tint than the two Olights and also much better colour rendition as it is 90 CRI instead of the 70 CRI Luminous SST40 emitters the Olights use.

As we can see in the comparative graph below, the Olight Baton 3 is by far the most efficient light, but also "cheats" a bit by setting its high output lower than the Olight S1R Baton II, to reserve power.

The Sofirn SC21 did surprisingly well, both in brightness and in efficiency considering the neutral tint and high CRI.

The following graph shows the first 10 minutes of the Turbo Full Runtime Comparison Graph in greater detail.

What is even more interesting is the much more realistic scenario of using the lights on High rather than Turbo. That way, the much hotter Samsung LH531D of the Sofirn SC21 does not need to step down due to thermal regulation and sits between the two Olights in brightness. It also outperforms the Olight S1R Baton II in run time, albeit while being less bright.

The following graph shows the first 10 minutes of the High Full Runtime Comparison Graph in greater detail.

There is no winner here, just 3 very good EDC lights, each with its advantages and disadvantages, according to the emitter that was selected and the programming of the driver. It does speak volumes though that the Sofirn SC21 is standing as equal amongst equals, with comparative advantages and disadvantages depending on preference and usage scenario, with 2 lights that cost twice what it does.

Current Draw

The Sofirn SC21 has a low parasitic drain that is below the ability of the clamp meter to measure. The Moonlight Mode only draws 10mA. The Low, Mid and High modes need 60mA, 197mA and 788mA respectively and Turbo required 2.83A.

Conclusion

The Sofirn SC21 is an small EDC torch that ticks many boxes. It uses a Li-Ion 16340 battery to power a 5000K, 90 CRI Samsung LH351D emitter and produce a maximum of 848 Lumen and 118m of throw (measured). This is indeed an excellent performance for a high CRI light with neutral tint and even though it falls short of the advertised 1000 Lumen and 135m, it is very respectable and more than enough for EDC purposes while the quality of the light is more than enough compensation for the reduced performance.

The driver is regulated and provides stable output with no PWM on any level. It incorporates Thermal Regulation, Reverse Polarity Protection and Low Voltage Protection. The only flaw in my opinion is that the low voltage protection only kicks in at 2.5V which is not good for the battery. It should turn off the light at around 3V.

The build quality and anodization are very good and the user interface is simple, intuitive and versatile, providing both stepped and ramping options.

The button is easy to press and incorporates an indicative LED which shows the battery level and also when the light is charging and when the charging is finished.

The Sofirn SC21 uses a glass lens and aluminium reflector combination which is much harder to scratch than a plastic TIR optic and does not melt if there is debris on the lens.

The tail cap is magnetic and the magnet is strong enough to hold the light at any orientation. The clip allows for lens down deep carry and is bidirectional so the Sofirn SC21 can be clipped to a hat to use as a headlamp.

The Sofirn SC21 can be purchased directly from the Sofirn Website and the cost at the moment this review is written is $23.99, including the battery or $20.99 without the battery. Shipping costs $3.99 and tax varies depending on the country of destination.

Let's list the Pros and Cons of the Sofirn SC21:

Pros

+ Value for money.

+ High CRI.

+ Neutral 5000K tint.

+ No PWM at any output level.

+ Glass lens and Aluminium reflector.

+ Aluminium Alloy construction.

+ Very good anodization and fit and finish.

+ USB type A to C and type C to C charging.

+ Fast battery charging, in less than 1.5h.

+ Low Voltage Protection.

+ Thermal Regulation.

+ Reverse Polarity Protection

+ Well balanced beam.

+ 16340 Li-Ion 800mAh (788mAh measured) battery included.

+ Battery level and charging LED indicator.

+ Simple and intuitive stepped and ramped UI.

+ IP68.

+ Bidirectional clip, which some users like as it can be clipped on to a hat to use as a head lamp.

+ Compatible with all button top 16340 batteries that can provide at least 3A.

Cons

- Low voltage protection turns off the light at 2.5V which is too low and can damage the battery and affect its longevity.

- Fast battery charging can affect battery longevity.

- Bidirectional clip, which some users dislike as it is not as sturdy and easy to use as unidirectional clips.

- The clip could scratch the finish over time.

- The battery level indicator LED could have more levels.

 TheLAB.GR

Thanks to Sofirn for providing the light for review

Polymeros Achaniotis 07/09/2021

Acebeam E70-AL

Sun, 22 Aug 2021 08:04:00 +0000

Introduction

This review is brought to you by a stroke of good luck. I was looking at the Acebeam E70-AL as the design and specifications had caught my eye, when I stumbled upon a giveaway the company was doing on BLF. Tyche ruled in my favour and upon receiving my prize, I decided that a review was in order.

Acebeam is well known amongst torch enthusiasts as a higher end company and both their products and their price tags reflect that. The Acebeam E70-AL may not be the most value for money option available but the specifications and quality are way beyond what you will find in budget friendly lights.

Are you interested to see what the Acebeam E70-AL has to offer? Let's start at the beginning.

Unboxing

The Acebeam E70-AL comes in a white box with a clear window that allows a view of the light. The box emphasizes the 5 year warranty and the high efficiency, constant current boost circuit, both attesting to the high quality of the Acebeam E70-AL.

The back side of the box does not offer any more information on specifications, but rather the company address, a few QR coded links and the usual certification logos.

The top and bottom have no print, so nothing to show. One of the longer sides lists the possible applications of the light while the other shows if the battery is included, the LED colour temperature and a warning regarding safe usage.

Regarding the LED choice, there is 6500K which is what my prize came with, 5000K and another option not depicted on the box, which has recently been added to the company's website, a 95+ CRI 4500K LED. The 6500K and 5000K are 6V LEDs while the 95+ CRI 4500K is a 12V LED and comes with a different driver.

The output of the 6500K emitter is rated at 4600 Lumen, the 5000K 10% less (according to an company representative) and the 95+ CRI 4500K emitter at 2500 Lumen.

The box features magnetic closure and opens like a book to reveal its contents and a thank you note from the company.

The accessories include a charging cable, 2 spare O-rings, a storage bag and a lanyard not pictured here as it came attached to the light (as seen in the photos above). The leaflets include the manual, a warranty card, safety instructions and a note explaining the battery is in the light and the insulation sheet needs to be removed before use.

I removed the lanyard to better show the light, which features a unique design.

The Acebeam E70-AL is made of aluminium and has flawless machining and anodization. The anodization of the main body is black and there is an inner tube which is anodized blue. The design is very harmonious and pleasing to the eye.

The bezel is crenulated and the head features alternating circular and oblong designs which work well together.

The 8 oblong grooves are meant to house 12x2mm tritium vials according to Acebeam, but are not deep enough to completely cover them, measuring only 1.60mm in depth. As the glass tritium vials are very fragile, protruding equals 100% chance of breaking. There are some hard to obtain 11.5x1.4mm and 10x1.5mm tritium vials that would fit nicely but even if you can find them, the colour choice is quite limited. I would suggest to Acebeam to either make the grooves deeper by 0.5mm or, if not possible, to not advertise them as fit to house 12x2mm tritium vials.

The body of the light features a helix design, with cut outs revealing the blue anodized inner tube and creating a beautiful visual effect. It also features 6 oblong grooves where it connects to the head that are meant to house 6x1.5mm tritium vials and are perfectly sized for it.

The tail part of the tube features tapped holes where the simple and functional clip comes pre-attached. There are 3 sections at the back end of the body. One with the clip installed, one with tapped holes, where the clip could be moved to - no idea why though - and one with bigger holes, where the lanyard came pre-installed.

Removing the lanyard is very easy but re-installing it requires thin tweezers, hence why the company probably opted to ship the light with the lanyard already installed.

In total, the Acebeam E70-AL features an intricate design which looks quite busy but still harmonious and pleasing to the eye. The machining of such a design is not an easy feat, especially with the level of detail and quality we see here.

The business end of the light comes with a protective plastic film, which must be removed before use.

Under the protective film, we can see the shallow orange peel reflector and the Cree XHP70.2 emitter. The combination of such a large LED with the shallow OP reflector guarantee a very floody beam.

The back end of the light is quite simple. It is where the switch for the operation of the light is located. I like the simple, recessed design of the switch which allows the light to tail stand and helps to protect from accidental activation.

The web address of the company is printed in white, which I find unnecessary and deducting from the aesthetics of the light. I do not mind the name of the light and the hot warning symbol printed at the front, but I would also prefer it if the CE and crossed out bin symbols were omitted.

The head and body unscrew to reveal that the battery ships inside the light, with the positive terminal insulated for safety. The threads are square cut, smooth and come nicely greased.

The battery is an Acebeam branded 21700 Li-Ion battery, rated at 5100mAh and it features a USB C port near the positive terminal.

Both battery contacts feature thick, good quality springs. These are probably adequate, but I would have still liked to see double springs or spring bypass wires in such a powerful light, to reduce the resistance even further. This would not have changed the brightness, as the Acebeam E70-AL features a boost driver, but could have helped a little with efficiency and run times.

As mentioned before, there is an inner tube, as the light features an electronic switch at the back and both the negative terminal of the battery and the switch need to be connected to the driver. On the driver end, the silver ring seen on the perimeter contacts the outer tube and connects to the battery negative and the two gold plated contacts near the perimeter connect to the inner tube and transfer the switch presses. The spring in the middle of the driver makes direct contact with the positive terminal of the battery.

Having springs on both ends makes the light resilient to bumps and drops, as the battery is suspended between two springs under pressure and is not likely to momentarily disconnect and turn the light off.

Quality

We have already seen the extraordinary quality the Acebeam E70-AL has, for a production light. Even at close inspection, both the machining and the anodization are perfect.

No edge is sharp, not a single spot less than perfectly anodized. This is indeed a top quality light.

Specifications

The specifications of the Acebeam E70-AL, as mentioned on the company's website, can be seen in the following table:

The Acebeam E70-AL is made of aerospace grade aluminium which has premium type 3 hard anodization. It features the very powerful Cree XHP70.2 LED which allows it to achieve an impressive maximum brightness of 4600 Lumen. The maximum throw distance is 240m, which means that the light is very floody, as expected from the configuration of the LED and the reflector.

The Acebeam E70-AL is quite resilient and sturdy, with an IP68 water proof rating and a 1m impact resistance

User Interface

The user interface of the Acebeam E70-AL is simple and intuitive.

Despite featuring a recessed button, a lock mode and the ability to hard lock the light by unscrewing the head 1/4 of a turn, a double click is required to turn the light on. This prevents accidental activation of unlocked lights and protects even the most careless of users from accidental activations, which would be disastrous with such a powerful light.

The user interface of the Acebeam E70-AL is as follows:

From Off
- Double click to turn on (with memory for the 4 main modes), click again to turn off.
- Click and hold from off to turn on at Moonlight mode. Keep holding until if flashes three times to lock the light. Click and hold again to unlock into Moonlight mode.
- Triple click for Strobe. Click to turn off or double click for Turbo.
From On
- Click to turn off.
- Hold to cycle through the main modes (Low, Mid1 ,Mid2, High).
- Double click for Turbo. Double click again to return to previous mode.
- Triple click for Strobe. Click to turn off or double click for Turbo. From Turbo, click to return to previous (before Strobe) mode. There is no way to go directly from Strobe to a previous, non Turbo mode.

The rated output and runtimes of all the modes can be seen in the table below.

Beam-shots

The beam pattern of the Acebeam E70-AL is exactly what can be expected from a large emitter combined with a shallow, orange peel reflector. The light is very floody, with a large hot spot and a lot of spill.

I tested the Acebeam E70-AL outside, over a distance of 70m.

The following video shows a comparison of the Acebeam E70-AL with the Emisar D4V2 Ti (with 4 Cree XP-L HI 5000K emitters) and the Fireflies E07 (with 7 Nichia 219b sw45k emitters).

Driver

The driver of the Acebeam E70-AL is a digitally regulated, constant current boost driver, which should be able to maintain constant brightness regardless of the battery voltage and also features smart temperature control, reverse polarity protection and low voltage protection. Very refreshing to see such a high quality driver.

There is no PWM in any of the modes.

Tint and Size Comparison

The tint of the Acebeam E70-AL is cool white at 6500K (5000K and 4500K with high CRI options are also available) and it is definitely above BBL and distinctly green. This is by no means a pleasing tint nor is the light high CRI. The intend was to maximize output and efficiency.

In the comparison photo below, you can see the Acebeam E70-AL in the middle, compared to the more neutral tint of the Emisar D4V2 Ti (with Cree XP-L HI 5000K emitters) on the left and the much rosier tint of the Fireflies E07 (with Nichia 219b sw45k emitters) on the right.

The above photo also offers a size comparison between the lights. The Acebeam E70-AL is not a light that can be easily carried in your trousers pocket, but it fits fine in a winter jacket or a backpack.

Battery and Charging

The battery included with the Acebeam E70-AL is a 21700, rated at 5100mAh and features onboard USB C charging.

The specifications of the Acebeam IMR21700NP-510A with Built-in USB C Port, as shown on the company's website, are listed in the following table:

I measured the capacity of the battery at exactly 4716mAh which is rather low. The battery's internal resistance was measured at 26mΩ. The light has Low Voltage Protection and turns off when the battery voltage drops to 2.7V.

Charging the Acebeam IMR21700NP-510A with Built-in USB C Port is very easy. Just connect the provided cable or any other USB C cable to the charging port and its other end to a charger. Both USB A to USB C and USB C to USB C cables can be used as well as any charger, including the ones that support PD. This is very convenient as you can charge the Acebeam IMR21700NP-510A with Built-in USB C Port with any USB C cable and charger you have at hand. The LED on the battery lights up red to indicate that it is charging and turns green when the charging is completed, at 4.14V.

The Acebeam IMR21700NP-510A with Built-in USB C Port took 6 hours, 22 minutes and 27 seconds to charge from 2.7V to 4.14V. The maximum current drawn was 0.899A, which is low for a 21700 battery and explains the very long charging time. A charging circuit which can support 2A charging would have been much better.

As the specifications indicate that the standard charging current is 1A, I can't help but wonder if it would be OK to charge the battery at 2A through a charger with large enough cradles to fit a protected button top 21700.

A very nice feature of the Acebeam IMR21700NP-510A with Built-in USB C Port is its ability to function as a power bank. This is facilitated by the USB A port on the included charging cable, but a USB C to USB C cable can also be used, even for devices that do not support power delivery.

When the Acebeam IMR21700NP-510A with Built-in USB C Port is functioning as a power bank, its indicative LED turns blue.

There is no information available regarding the current the battery can provide through its USB C port so I had to investigate. Anything above 2.7A and the circuit would immediately reset. I gradually reduced the current until I found it to be initially stable at 2A, which is a good output. Unfortunately, this only worked with a fully charged battery and after 40 minutes of drawing 2A, the circuit started resetting. At the end of the test, and after countless resets, the battery voltage was 3.14V.

I tempered my expectations and tried drawing 1A from the battery and this seems to be the maximum supported current for the power bank function, as the battery was able to maintain it with stability for 3 hours, before the circuit reset a few times and then turned off. I would have much preferred it if it had turned off without the resets, as I am not sure if they could damage any sensitive equipment being charged. The battery voltage at the end of the test was 2.99V.

The Acebeam IMR21700NP-510A with Built-in USB C Port is a battery packed with nice features such as onboard USB C charging and power bank functionality, along with a large 5100mAh capacity. Nevertheless, the actual capacity of 4715mAh and the slow charging speed leave something to be desired, especially considering the $23.90 price tag.

EDC Bulb

Along with the light and battery, my giveaway prize included a nice little accessory which adds a lot of value to the Acebeam IMR21700NP-510A with Built-in USB C Port. It is the Acebeam EDC Bulb.

The Acebeam EDC Bulb attaches to the top of the Acebeam IMR21700NP-510A with Built-in USB C Port and connects to the battery's USB C port to draw power.

The installation and operation are explained in the animation below.

The specifications of the Acebeam EDC Bulb as found on the company's website, are in the following table:

This tiny plastic attachment adds a lot of functionality to the Acebeam E70-AL and is a must buy, especially since it normally costs only $3.99 and is currently on offer for a mere $0.10. Unfortunately, it is also out of stock.

The modes that the Acebeam EDC Bulb supports, along with their respective run times, are listed in the table below:

The Acebeam EDC Bulb is so simple and at the same time so useful and nicely implemented! It really adds value to the Acebeam IMR21700NP-510A with Built-in USB C Port.

Current Draw

The Acebeam E70-AL has a low parasitic drain that is below the ability of the clamp meter to measure. The Moonlight mode only draws 20mA. The Low, Mid1, Mid2 and High modes need 83mA, 216mA, 748mA and 1.961A respectively and Turbo required 10.40A.

These values are rather low for the Lumen output and attest to the efficiency of the driver and emitter combination.

Output & Runtimes

The Acebeam E70-AL with the 6500K emitter is rated at maximum of 4600 Lumen and 240m of throw.

According to my measurements, the maximum output (at turn on) was 4494 Lumen, which is short of the advertised 4600 but still very close to the specs and definitely within the error margin of my equipment. ANSI output (at 30 seconds) was 3805 Lumen and the rapid output decline continued down to 1210 Lumen at 57 seconds. It then stayed stable at a very impressive for the size of the light 1218 Lumen for a further 1 hour and 54 minutes! After that, it stayed for 5 minutes at 153 Lumen and finally dropped to 53 lumen for a further 53 minutes, before it turned off.

The temperature regulation was very good and the body of the light was within the ability of a person to hold with bare hands. The test was done indoors, without any cooling or ventilation, so the results will be even better outside.

The following graph shows the first 10 minutes of the Full Runtime Graph in greater detail.

The stability and efficiency of the Acebeam E70-AL are truly impressive and make for a very usable and reliable light. In contrast to the unregulated drivers used in more budget oriented lights, the driver of the Acebeam E70-AL features a constant current output, providing a stable and constant light output and long run times.

Turbo output is short, as dictated by thermal regulation and the laws of physics, but it is there for the short bursts that it may be needed for. High output is incredibly steady and lasted for an impressive 1 hour and 55minutes, after the turbo activation, which is 10 minutes more than what the specifications say.

According to the specifications, the maximum throw of the Acebeam E70-AL is 240m and the maximum intensity 14400cd. I measured 210m and 11005.28cd.

I charged the battery of the Acebeam E70-AL with an external charger to 4.20V, to see if I can squeeze any more performance out of the light but the gain was minimal, increasing the maximum output from 4494 Lumen to 4561 Lumen, the maximum throw from 210m to 214m and the maximum intensity from 11005.28cd to 11437.92cd. Therefore, I think that the company's decision to terminate the charging at 4.14V and thus increase the longevity of the battery is correct.

Conclusion

The Acebeam E70-AL is a light that emphasizes quality.

The machining is intricate with no imperfections or sharp edges. The type 3 hard anodization is uniform and without flaws. The driver is fully regulated and provides constant current output without any PWM on any of the modes, while being very efficient. The Cree XHP70.2 emitter contributes to the high output and efficiency of the light.

The User Interface is simple and functional and extra care has been taken to prevent accidental activations.

The size is normal for a 21700 light and not too easy to fit into trousers pockets, although not impossible. I would prefer to carry the Acebeam E70-AL in my backpack or in the pocket of a winter jacket.

The optional Acebeam IMR21700NP-510A with Built-in USB C Port battery features USB C onboard charging and Power Bank functionality while the Acebeam EDC Bulb optional accessory adds a lot of usability and value.

The Acebeam E70-AL can be purchased directly from the company's website and the cost at the moment this review is written is $94.80, including the battery and shipping, worldwide. The light without the battery costs $74.90, but keep in mind that only button top protected 21700 batteries that can provide at least 10.40A of current will work.

Let's list the Pros and Cons of the Acebeam E70-AL:

Pros

+ Digitally regulated constant current boost driver which provides stable output and high efficiency with no PWM.

+ High Turbo output for the size of the light.

+ Long runtimes.

+ Impeccable and intricate machining.

+ Perfect type 3 hard anodization.

+ Aerospace grade Aluminium Alloy construction.

+ USB type C charging integrated in the battery.

+ Low Voltage Protection.

+ Quality springs on both sides of the battery.

+ Smart ITS Temperature Control.

+ Simple and intuitive UI.

+ 6 slots for 6x1.5mm tritium vials.

+ IP68.

+ The battery provides Power Bank functionality.

+ The Acebeam EDC Bulb optional accessory is very useful.

Cons

- The 8 slots on the head meant to house 12x2mm tritium vials are only 1.60mm deep and cannot protect the fragile glass vials so are not fit for purpose.

- The battery is expensive.

- The battery takes a long time to charge.

- Actual battery capacity is 4716mAh instead of 5100mAh.

- A holster should be included.

 TheLAB.GR

Thanks to Acebeam for doing the giveaway and Tyche for providing the light for review

Polymeros Achaniotis 24/08/2021

Wurkkos TS30

Mon, 02 Aug 2021 21:21:00 +0000

Introduction

With the multitude of companies that make torches nowadays, it is getting very hard to distinguish between the true value for money offerings and the overpriced, low quality ones. Price alone, certainly cannot be the determining factor as there are a lot of cheap lights out there that are not worth even a small fraction of their seemingly bargain of a cost.

Wurkkos is known as a good value for money brand and I thought it would be interesting to test one of their lights and see if they live up to their reputation. The model we will be reviewing is the Wurkkos TS30, which is one of their higher spec lights.

So without further delay, let's see what the Wurkkos TS30 has to offer and how it will do under our microscope.

Unboxing

The Wurkkos TS30 comes in a a nice cardboard box with the company logo and colours. The box appears not to be specific to the Wurkkos TS30 as the model is only indicated on the sticker at the bottom.

Opening the box, we are greeted with a very nice presentation. Everything is neatly arranged and protected within laser cut outs in the black protective foam. That is a really nice touch and stands out from some of the other budget offerings.

Inside the box, besides the light itself, we can find the manual, a tab instructing the owner to remove the insulator from the battery negative before using the light and the accessories.

The accessories consist of a lanyard, 2 spare O-rings, a battery spacer if anyone wants to use a 18650 battery in this 21700 light, a 1m USB A to USB C charging cable and a very aggressive strike bezel with 4 very long spikes.

The Wurkkos TS30 is made of black anodized aluminium and is a medium sized light. It is definitely not an EDC, pocket friendly little torch, but it will easily fit in a backpack or large jacket pocket.

It features a side switch surrounded by cooling fins, on the other side of which there is a rubber flap protecting the USB C charging port. The head is wider than the body, measuring at 49mm diameter, and features a stainless steel bezel with crenulations so you can see if the light is on when placed on its head.

The body has a pattern which offers some grip, without being very aggressive.

The flap is easier to open and close than some other lights I have used and seems quite secure.

The Wurkkos TS30 features a smooth triple reflector for its 3 Luminous SST40 emitters.

The tail cap has 2 lanyard holes and can allow the light to tail stand, somewhat precariously.

The battery tube can be unscrewed from both the head and the tail cap. The threads are smooth and the diameter is different on each side so the tube cannot be reversed. The Wurkkos branded 21700 battery comes inside the light and there is the mentioned black insulating disk blocking the negative contact, for safe shipping.

The head features a brass protrusion to make contact with the positive pole of the battery. The tail cap has a thick double spring, which looks more than capable of transferring a lot of current, with minimal losses.

The 21700 battery that comes with the Wurkkos TS30 is wrapped in a white sleeve that features the company logo. Despite the fact that the Wurkkos TS30 is advertised with a 4800mAh 21700 battery, the one in my sample claims to be 5000mAh. We are going to test that.

Strike Bezel

One of the accessories that come with the Wurkkos TS30 merits special mention. It is none other than the very aggressive strike bezel.

This is milled from a solid bar of stainless steel and cannot be too cheap to manufacture. Personally, I find it superfluous and it cannot be legally carried in the UK where I live and I guess in many other countries. But I get that some people may be into this kind of "tactical" stuff.

In my opinion, the cost of manufacturing this bezel could go into improving something else, as the torch already comes with a very nice bezel.

Exchanging the bezels is really simple. Just unscrew the one the light comes with and screw in the other.

The light looks... quite unique with the strike bezel installed.

Wurkkos advertises this bezel for self defence and breaking glass in an emergency.

Quality

The build quality, fit and finish of the Wurkkos TS30 is definitely beyond its price range. I was quite impressed by the fact that even the cooling fins are chamfered and the anodization is flawless everywhere, even on all the edges.

The blowout of the Wurkkos TS30 in the image below shows that everything is laid out as it should be. All materials seem of high quality and fit for purpose.

Specifications

The specifications of the Wurkkos TS30, as found on the company's website, are as follows:

As you can see, the Wurkkos TS30 uses 3 Luminous SST40 emitters to deliver its light output, which is rated at 5950 Lumen for the 6000K version, which we are testing and 5000 Lumen for the 5000K version. The maximum throw is rated at 480m.

The light uses the included 21700 battery but can also be used with a high drain 18650 battery, with the included sleeve. It is IPX-8 rated for water resistance.

The specifications state that the Wurkkos TS30 is 182 +/- 1g. I weighed my sample exactly at 178.7g. The battery weighs 74.1g, bringing the total to 252.8g. Not too bad for the power it packs.

User Interface

The user interface of the Wurkkos TS30 is quite simple and intuitive. The light features 2 modes, Stepped and Ramping. Out of the box, the light comes in Stepped Mode.

Stepped Mode:

From Off
- Click for on, click again for off.
- Click and hold from off to turn on at Eco (moonlight) mode. Keep holding until if flashes to change into Ramping Mode. Click again for off.
- Double click for Turbo. Click to turn off.
- Triple click for Strobe. Click to turn off.
- Quadruple click for lockout. Repeat to deactivate.
From On
- Click to turn off.
- Hold to cycle through the main modes (Low, Mid, High).
- Double click for Turbo. Click to return to previous mode.
- Triple click for Strobe. Click to return to previous mode.
- Quadruple click for lockout. Repeat to deactivate.

Ramping Mode:

From Off
- Click for on, click again for off.
- Click and hold from off to turn on at Eco (moonlight) mode. Keep holding until if flashes to change into Ramping Mode. Click again for off.
- Double click for Turbo. Click to turn off.
- Triple click for Strobe. Click to turn off.
- Quadruple click for lockout. Repeat to deactivate.

From On
- Click to turn off.
- Hold to ramp up or down. Ramping changes direction when the button is pressed again within 1,5 seconds. The light flashes to indicate reaching the lower or upper end of the ramp.
- Double click for Turbo. Click to return to previous setting.
- Triple click for Strobe. Click to return to previous setting.
- Quadruple click for lockout. Repeat to deactivate

The brightness, runtime and beam distance of each mode can be seen in the following table.

The Wurkkos TS30 features Mode Memory. This means that it memorizes the last brightness level used (except Strobe). Mode Memory cannot be deactivated.

While the light is in lock out mode, you can access the Eco mode momentarily by holding down the button.

If you prefer a mechanical lock, you can just unscrew the tail cap half a turn. This will disconnect the battery and is recommended for long storage.

Beam-shots

The beam pattern of the Wurkkos TS30 is exactly what can be expected from a triple light with smooth reflectors and 5x5mm emitters. Very close up the pattern shows a triangular shape but at any working distance it becomes perfectly round. There are several rings, typical with smooth reflectors and a well defined hot spot.

Installing the strike bezel effects the shape of the beam a lot.

I tested the Wurkkos TS30 outside, over a distance of 70m.

The following video shows a comparison of the Wurkkos TS30 with the Sofirn SP36 Pro (also with Luminous SST40 emitters) and the Fireflies E07 (with Nichia 219b sw45k emitters).

Driver

The driver of the Wurkkos TS30 is not regulated and uses PWM to dim the light, on all levels, except, of course, on full. The PWM is visible to the camera but not visible to the naked eye, on any level.

Even though I prefer constant current drivers, PWM is an efficient and cost effective way to achieve LED dimming and if it is done at a high enough frequency, as seems to be the case here, it is not a problem.

The Wurkkos TS30 driver also features Advanced Temperature Regulation (set at 60 degrees Celsius), Low Voltage Protection and Reverse Polarity Protection.

Tint and Size Comparison

The tint of the Wurkkos TS30 is cool white at 6000K (5000K option is also available) and even though it is above BBL, it is not significantly green. This is actually quite a nice tint for a 6000K Luminous SST40 emitter.

In the comparison photo below, you can see the Wurkkos TS30 in the middle, compared to the much greener tint of the Sofirn SP36 Pro (also with Luminous SST40) on the left and the much rosier tint of the Fireflies E07 (with Nichia 219b sw45k) on the right. The SST40 used in the Wurkkos TS30 are not high CRI.

The above photo also offers a size comparison between the lights.

Charging

The battery included with the Wurkkos TS30 is a 21700, rated at 5000mAh and I measured it at exactly 4922mAh. The light has Low Voltage Protection and turns off when the battery voltage drops to 2.7V. The battery's internal resistance was measured at 42mΩ. These measurements show that a high quality battery is actually included with this light.

The indicative LEDs on the switch of the Wurkkos TS30 show the level of the battery charge. Green means that the remaining charge is at least 30%, red that it is below 30% and flashing red that it is critical and the light will soon turn off.

Charging the Wurkkos TS30 is very easy. Just lift the rubber cover and insert the provided (or any) USB type C cable to charge the light. Both USB A to USB C and USB C to USB C cables can be used as well as any charger, including the ones that support PD. This is very convenient as you can charge the Wurkkos TS30 with any USB C cable and charger you have at hand.

The indicative LEDs on the switch flash red to indicate the light is charging. They turn green when the charging is completed, at 4.16V.

The Wurkkos TS30 took 3 hours, 13 minutes and 28 seconds to charge the included battery from 2.7V to 4.16V. The maximum current drawn was 1.7466A, so a charger that can provide at least 2A is recommended. A charger is not provided with the light but you can use your phone charger.

Current Draw

The Wurkkos TS30 has a low parasitic drain that is below the ability of the clamp meter to measure. The Eco (Moonlight) Mode only draws 4mA. The Low, Mid and High modes need 352mA, 1.24A and 6.63A respectively and Turbo required 16.99A.

Output & Runtimes

The Wurkkos TS30 6000K is rated at 5950 Lumen and 480m of throw.

According to my measurements, the maximum output (at turn on) was 5000 Lumen, which is short of the advertised 5950 but still very respectable for the size and cost of the light. ANSI output (at 30 seconds) was 4563 Lumen and at 1 minute the output had dropped to 4401 Lumen. At 2 minutes the output had decreased to 2943 Lumen and then declined rapidly to reach 747 Lumen at 00:02:44, where it stabilized and slowly decreased to 454 Lumen from 00:02:44 to 01:45:10.

Then a strange thing happened which I cannot fully explain. A slow oscillation in output started to occur, with its floor exactly where the declining curve of output would have been and its amplitude steadily rising. This means the output was at least equal and mostly higher than it would have been otherwise, even exceeding 1000 Lumen at some points. Then the gradual decline of the output resumed and the light turned off at 02:29:36, while it was still holding an output of 171 Lumen.

All in all, a very useful output profile, managing heat, battery capacity and output quite well.

The following graph shows the first 10 minutes of the Full Runtime Graph in greater detail.

I did discuss the strange oscillation with a couple of electronics experts but without a look at the driver components (the driver is glued in place and not removable) they could not offer more insight into the cause. I think it is happening around the time the battery voltage is near the Vf of the emitters, but why that could cause this behaviour, I cannot say.

If any electronics experts can offer an explanation, I would appreciate leaving it in the comments.

Conclusion

The Wurkkos TS30 is a value for money light that will not disappoint. Its aero grade aluminium body is well made and hard anodized and the fit and finish are beyond its price point. The attention to detail is also there, with chamfered cooling fins and impeccable anodization.

The size and weight are great for its output rating and the provided 21700 battery allows for ample runtime. The output is more than enough for most purposes and the beam profile is well balanced.

The tint of the Luminous SST40 LED is a cool, slightly above BBL but without being too green tint. The CRI is low, but this is countered by the high lumen output these emitters allow the light to achieve.

The driver uses PWM to dim the output, so PWM is present at all output levels, except on full. The frequency of the PWM is high enough to not be visible to the naked eye and did not tire me when using the light.

The Wurkkos TS30 can be purchased directly from the Wurkkos Website and the cost at the moment this review is written is $45.99, including shipping, worldwide. That is a lot of torch for the money!

Let's list the Pros and Cons of the Wurkkos TS30:

Pros

+ Value for money

+ High output and throw

+ Aerospace grade Aluminium Alloy construction

+ Impeccable anodization and fit and finish

+ USB type A to C and type C to C charging

+ Low Voltage Protection

+ Advanced Thermal Regulation

+ Well balanced beam

+ 21700 Li-Ion 5000mAh battery included

+ Low power and charging LED indicators

+ Simple and intuitive stepped and ramp UI

+ IP68

+ At least 2.5 hours of usable light per charge

+ Compatible with all high drain 21700 batteries

Cons

- The driver uses PWM to dim the emitter. A regulated, constant current driver could have been used instead of including the strike bezel.

 TheLAB.GR

Thanks to Wurkkos for providing the light for review

Polymeros Achaniotis 05/08/2021

Sofirn SP36 Pro

Wed, 28 Jul 2021 06:47:00 +0000

Introduction

It is not always easy to walk the line between quality and value for money but Sofirn seems to have mastered the skill well. About a month ago I reviewed their EDC style torch, Sofirn SC31 Pro, and it was so good for its price that it peaked my curiosity. It therefore prompted me to have a look at their bigger and brighter models.

I asked their representative and she obliged by sending me the Sofirn SP36 Pro, a 8000 lumen lighirnt with Andúril firmware, 4 LEDs and 3 batteries.

This is by no means an EDC light. Its size puts it in the Soda Can lights category, albeit on the thin side of the genre. The generous specs and competitive pricing make for a combination that is hard to beat.

But let's not get ahead of ourselves. We need to see the light first and what it can do.

Unboxing

The Sofirn SP36 Pro comes in an unassuming, generic brown cardboard box with a barcode sticker that provides information about the contents.

Inside the box, we find the light, protected in bubble wrap, a 1 meter long USB A to USB C charging cable, 2 spare O-rings and the manual.

The light comes with a little tag, secured around the body with a rubber band.

The tag explains that there is an insulator in the battery compartment, stopping the batteries from making contact and thus rendering the light inoperable for safe shipping. The insulator needs to be removed before the light can be used.

The Sofirn SP36 Pro is a black, cylindrical light, about 12,5cm long and with a diameter of about 5cm at its widest point, which is the front bezel. The body features knurling which allows for a secure grip without being too aggressive. The button is rubbery, textured and slightly raised, so it is easy to find by touch, but can also be pressed while in a bag or in a (very large) pocket. Thankfully, there is a lock out option to cover for that contingency. The area around the button has some heatsinking in the form of fins, while at the exact opposite side there is a rubber flap.

Lifting the flap, reveals the USB C charging port.

Beauty is in the eye of the beholder so I will not presume to tell you if the Sofirn SP36 Pro looks good or not, but I can definitely say that it is an elegant and unassuming design that seems to mean business.

That is even more true about the business end of the light, which features a quadruple smooth reflector with deep enough cups to ensure a healthy amount of throw. The emitters at the bottom of those cups are 4 Luminous SST40, 6500K LEDs.

The battery compartment can be accessed by unscrewing the light into two halves and contains 3 batteries. The front side has an insulator covering which must be removed to reveal the circular battery contacts.

The batteries are Sofirn branded, size 18650 and rated at 3000mAh.

The inner circular battery contact is for the positive contacts of the 3 batteries while the outer circular battery contact gets the negative via the metal body of the battery compartment. The metal body of the battery compartment gets the negative from the batteries through 3 thick double springs at its bottom. The thickness of the springs, the fact that they are double and the whole design seems capable of transferring the power of the batteries to the light's driver circuit with minimal losses. It is also easy to mechanically lock out the light by unscrewing the battery compartment by half a turn, thus breaking the contact between its non anodized lip and the outer circular contact of the head.

With the insulator out of the way, the Sofirn SP36 Pro is ready for action, which is indicated by the 2 green indicator LEDs on the switch.

Quality

The build quality of the Sofirn SP36 Pro is surprisingly good - for the price point and feature set - and on par with other Sofirn lights.

The fit and finish are excellent, the knurling is of good quality and the anodization is without any flaws, except a few points at the edge of the cooling fins. The edges of the fins are also somewhat sharp, not enough to cut, but if you run your finger against them they seem like they would benefit from a little refining.

Specifications

The specifications of the Sofirn SP36 Pro, as found on the company's website, are as follows:

The light features 4 SST40 emitters at 6500K (cool white - there is also a neutral white version at 5000K) which provide 8000 lumen of max output and 450m of max throw. It is made of aerospace grade aluminium and is 126,7mm long and 50mm wide (which I verified to be correct). The weight without the batteries is 300g (I measured 297g without the batteries and 437g with the batteries).

The Sofirn SP36 Pro is USB C rechargeable and rated IPX8.

User Interface

The user interface of the Sofirn SP36 Pro is a love or hate deal, as the light features the Andúril UI. Personally, I love it and deeply enjoy the fact that it is feature packed but still provides simple, quick and intuitive access to the basic functions. Others hate it and consider it too much work, as the manual is extensive and even the flow chart found at the company's website can be intimidating to look at.

There is another kind of flow chart, provided for Andúril by the coder, which I personally prefer. They are different approaches to the same user interface:

In my experience, after the initial shock, it is very easy to start using the light and have an occasional look at the flow chart to remind oneself of the more advanced functions.

An important thing to do when you take any Andúril light out of the box is to perform a temperature calibration, as the light will depend on it to perform proper thermal control and balance brightness with temperature. That is also the best time to set the temperature limit. I find 50C to be a good temperature limit and the tests done in this review were done on a calibrated light and with a 50C temperature limit.

Beam-shots

The beam pattern of the Sofirn SP36 Pro is a product of the LED type and size and the reflector it uses. With the 5x5mm SST40 emitters and smooth reflector, it is not surprising that the result is a somewhat tight hot spot that throws nicely, surrounded by a nice, usable spill. All in all, a very balanced and usable beam pattern.

There are various rings and coronas that are typical of smooth reflectors and of multi emitter lights and may annoy white wall hunters, but those are unavoidable in this configuration and do not cause any problems in real world use.

I tested the Sofirn SP36 Pro over a distance of 70m.

The following video shows a comparison of the Sofirn SP36 Pro with the Olight X7R, the Fireflies E07 (Nichia 219b sw45k) and the Lumintop X9L.

Driver

The driver of the Sofirn SP36 Pro uses PWM to dim the light, on all levels, except, of course, on full. The PWM is visible to the camera but not visible to the naked eye, on any level.

Tint

The tint of the Sofirn SP36 Pro is the cool, greenish (above BBL) tint that is expected of the Luminous SST40 emitter. It is very similar to the tint of the Olight X7R. Next to them, for reference, the Fireflies E07 with Nichia 219b sw45k emitters which are high CRI and very rosy (below BBL) and the Lumintop X9L which uses a Luminous SBT90.2 emitter. The SST40 used in the Sofirn SP36 Pro is not high CRI.

Charging

The batteries included with the Sofirn SP36 Pro are rated at 3000mAh and I measured them to be right around that number (2950mAh / 3060mAh / 3049mAh) The light has under voltage protection and turns off when the battery voltage drops to 2.8V. The batteries' internal resistance was measured at around 120mΩ.

Charging the Sofirn SP36 Pro is very easy. Just lift the rubber cover and insert the provided USB type C cable to charge the light.

The indicative LEDs on the switch turn red to indicate the light is charging. They turn green when the charging is completed, at 4.22V.

Charging the Sofirn SP36 Pro from 2.8V to 4.22V took 4 hours, 57 minutes and 17 seconds. The maximum current drawn was 1.8712A, so a charger that can provide at least 2A is recommended. A charger is not provided with the light but you can use your phone charger.

Current Draw

The Sofirn SP36 Pro has a very low parasitic drain that is below the ability of the clamp meter to measure. It shows between 0mA and 1mA, with the indicative LEDs on the switch, on (they can be turned off through the Andúril UI). The light also has a very low moonlight mode that only draws 3mA. The top of the ramp drew 5.93A and Turbo required 16.34A.

You can run the Sofirn SP36 Pro with all 3 batteries or with 2 or even 1 battery. Each of the batteries provided with the Sofirn SP36 Pro can give up to 10A (tested) so in order to get full brightness on Turbo you need to use at least 2 batteries in the light. All measurements were taken with all 3 batteries in the light.

Output & Runtimes

The Sofirn SP36 Pro is rated at 8000 lumen output and 450m of throw. I measured it both with batteries fresh off the charger and after they had rested.

With batteries fresh off charging, the maximum output (at turn on) was 7417 lumen, which is short of the advertised 8000 but still very respectable for the size of the light. ANSI output (at 30 seconds) was 6716 lumen and at 2 minutes the output had dropped to 1490 lumen. It then gradually increased to 1767 lumen until the 5 minute mark. Over the next 30 seconds the output dipped to only 300 lumen and gradually increased to over 500 lumen over the next 15 minutes, to stay at around that level until the 5 hours and 40 minutes mark.

With rested batteries, the maximum output (at turn on) was 6726 lumen, which is 9.3% less than with batteries fresh off the charger. ANSI output (at 30 seconds) was 6364 lumen, which is 5.2% less than with batteries fresh off the charger. At 2 minutes the output had dropped to 2327 lumen, which is 36% more than with batteries fresh off the charger. It seems the initial extra burst of energy that the batteries can give fresh off the charger has a big toll! The output gradually decreased to 2190 lumen until the 4 minutes and 5 seconds mark and over the next 20 seconds the output dipped to 518 lumen, where it remained stable, gradually increasing over the next hours, to reach a maximum of 931 lumen at 5:13:55. It then dropped gradually, to 222 lumen, at 5:40:10.

From these 2 scenarios it is obvious that even though the batteries fresh off the charger will give a bigger, more impressive boost, the advantage lasts about a minute and has consequences for the entire runtime of the light. Moreover, it is highly unlikely it will happen in real life anyway, as most people do not keep the light on charging all the time.

The temperature was very well controlled, as you can see in the runtime graphs. Turning on the light at Turbo, will cause a rapid rise in temperature, followed by a big dip in output, while the temperature stabilizes. This can be avoided if the light is not on Turbo, but used at a more moderate output level.

The light was temperature calibrated, according to the manual, and the temperature limit was set at 50C. The actual temperatures on the button and on the body of the light can be seen on the graphs and were kept well within the set limit.

All in all, the Sofirn SP36 Pro gave very usable light for about 5.5 hours, when turned on at Turbo, which is very respectable. Usage at more moderate levels, will of course, result in higher runtimes and lower, if any, output dips, due to temperature management.

The maximum intensity of the light, with rested batteries, was measured at 56784 cd, which translates to a throw of 476,59m, which is 5.9% more than the 450m advertised.

Conclusion

The Sofirn SP36 Pro is a value for money, Soda Can sized light that will not disappoint. Its aluminium body is well made and hard anodized and the fit and finish is beyond its price point. The mildly sharp edges of the fins are not a concern, but could have been smoother.

The size and weight are great for its output rating and the 3 provided 3000mAh 18650 batteries allow for ample runtime. The output is more than enough for most purposes and the beam profile is very balanced.

The tint of the Luminous SST40 LED is a cool, above BBL (greenish) tint and the CRI is low, but this is countered by the high lumen output this emitter allows the light to achieve.

The Sofirn SP36 Pro can be purchased from the Sofirn Website and the cost at the moment this review is written is $49.99, including shipping, worldwide. That is a lot of torch for the money!

Let's list the Pros and Cons of the Sofirn SP36 Pro:

Pros

+ Value for money

+ High output and throw

+ Aerospace grade Aluminium Alloy construction with good fit and finish

+ USB type C charging

+ Low Voltage Protection

+ Thermal regulation

+ Well balanced beam

+ 3x 18650 Li-Ion 3000mAh (actually measured) batteries included

+ Low power and charging LED indicator

+ Andúril UI

+ Lighted button with indicative LEDs

+ IP68

+ At least 5.5 hours of usable light per charge

+ Compatible with all button top 18650 batteries

Cons

- The fins could have smoother edges

- The driver uses PWM to dim the emitter

 TheLAB.GR

Thanks to Sofirn for providing the light for review

Polymeros Achaniotis 30/07/2021

Sofirn SC31 Pro

Wed, 23 Jun 2021 15:45:00 +0000

Introduction

There are many quality torches out there for the enthusiasts and the collectors and even more cheap and cheerful ones for people who just want to have some light in the dark. But the characteristic that the majority of buyers are after is the one that is actually the hardest to find: Value for money.

This is the area where, in my experience, Sofirn actually excels, as I have yet to find a brand that balances quality and price as well as they do. And while they are still more expensive than most cheap and cheerful (aka rubbish) torches on the market, they are not so by much, making their offering quite appealing to many users.

The model we will be looking at in this review is the Sofirn SC31 Pro.

The Sofirn SC31 Pro is an EDC style torch, so it is a pocketable size, packed with features and providing a very decent light output, at a very competitive price. Intrigued? Let us see exactly what it does and how it accomplishes it.

Unboxing

The Sofirn SC31 Pro comes in a colourful cardboard box, which is common for many Sofirn models and the exact model is indicated only on the sticker at the back of the box. As the box is common for many models, no features or specifications are indicated on it.

Inside the box is a clear plastic moulded case that holds the torch securely and protects it during shipping. In the recess, under the plastic case, we find the accessories as well as the manual.

The accessories include the charging cable, the manual and a zip lock bag with the rest of the accessories.

The charging cable is USB type A to USB type C and its length is 104cm, including the plugs.

The zip lock bag contains a lanyard and 2 spare O-rings.

The Sofirn SC31 Pro is made of aluminium with black anodization and has a side button and a pre-installed clip. The body has knurling, to make it grippy and the tailcap features a hole for the lanyard. The sides of the driver portion feature some small fins for heat dissipation. Everything is nicely finished and chamfered, except for the lanyard hole and the fins that are somewhat sharper, but not much.

On the opposite side of the button there is a rubber cover and under it, a USB type C charging port.

The tail cap has some knurling that helps to grip it and screw it on or off. The battery ships inside the light but an insulator prevents it from touching the negative spring so the light cannot be activated during shipping. The threads are square cut and come nicely greased. They are anodized, so even a slight unscrewing of the tail cap will break electrical contact and mechanically lock the light.

The included battery is a Sofirn branded button top 18650 with a capacity rating of 3000mAh.

The battery tube can unscrew from the torch head also. This allows it to be replaced with the optional Sofirn 18350 Short Tube, available on the Sofirn website for $1.49.

There are springs on both the tail cap and the head of the torch, ensuring compatibility with different length 18650 batteries and also the continuity of the circuit during drops and bumps. The springs are thick and of high quality. They are more than adequate for the current this torch draws.

The included tail cap has a lanyard hole (which could have smoother corners) and is not magnetic. An optional, Sofirn Magnetic Tailcap is available at the Sofirn website for $1.69.

The Sofirn SC31 Pro features a smooth reflector and a cool (6500K - in our sample) or neutral (5000K) Luminus SST-40 emitter.

Quality

The build quality of the Sofirn SC31 Pro is surprisingly good - for the price point and feature set - and on par with other Sofirn lights.

The fit and finish is excellent, the knurling is of good quality and the anodization is without any flaws.

There are a couple of somewhat sharp edges - specifically on the fins and the lanyard hole - and the 2 indicative LEDs on the switch (which can be turned off or set to blinking mode through the Andúril UI) are not quite aligned, which is a perfectionist's nightmare, but these are details that would raise the cost and can be considered acceptable at this price point.

Specifications

The specifications of the Sofirn SC31 Pro, as found on the company's website, are as follows:

The net weight is declared to be 59g. I weighed it at 61.7g without the battery and 107.7 with the battery. The length and width of the torch are not mentioned. I measured the length to be 11.5cm and the width 2.6cm.

The Sofirn SC31 Pro is rated at 2000 lumen light output and 200m of throw, which is an excellent performance for its size.

User Interface

The user interface of the Sofirn SC31 Pro is a love or hate deal, as the light features the Andúril UI. Personally, I love it and deeply enjoy the fact that it is feature packed but still provides simple, quick and intuitive access to the basic functions. Others hate it and consider it too much work, as the manual is extensive and even the flow chart can be intimidating to look at.

In my experience, after the initial shock, it is very easy to start using the light and have an occasional look at the flow chart to remind oneself of the more advanced functions.

Beam-shots

The beam pattern of the Sofirn SC31 Pro is a product of the LED type and size and the reflector it uses. With the 5x5mm SST-40 and a smooth reflector, it is not surprising that the result is a somewhat tight hot spot that throws nicely, surrounded by a nice, usable spill. All in all, a very balanced and usable EDC beam pattern.

There are various rings and coronas that are typical of smooth reflectors and may annoy white wall hunters, but those are unavoidable with smooth reflectors and the smooth reflector is what is giving the Sofirn SC31 Pro a very nice throw.

I tested the Sofirn SC31 Pro over a distance of 70m, which is more than adequate for an EDC style torch.

The following video shows a comparison of the Sofirn SC31 Pro with the very well known and much more expensive Olight Baton Pro. Both torches are rated at 2000 lumen and it is obvious that the Sofirn SC31 Pro has a tighter hot spot while the Olight Baton Pro has more spill. I will let each of you decide which one you prefer.

Driver

The driver of the Sofirn SC31 Pro uses PWM to dim the light, on all levels, except, of course, on full. The PWM is visible to the camera but not visible to the naked eye, on any level.

Tint

The tint of the Sofirn SC31 Pro is the cool, greenish (above BBL) tint that is expected of the Luminous SST-40 emitter. It is very similar to the tint of the Olight Baton Pro. Next to them, for reference, the Fireflies E07 with Nichia 219b sw45k emitters which are high CRI and very rosy (below BBL). The SST-40 used is not high CRI.

Charging

The battery included with the Sofirn SC31 Pro is rated at 3000mAh and I measured it to be exactly that! The light has under voltage protection and turns off when the battery voltage drops to 2.8V. The battery internal resistance was measured at 80mΩ.

Charging the Sofirn SC31 Pro is very easy. Just lift the rubber cover and insert the provided USB type C cable to charge the light.

The indicative LEDs on the switch turn red to indicate the light is charging. They turn green when the charging is completed, at 4.16V.

Charging the Sofirn SC31 Pro from 2.8V to 4.16V took 2 hours, 6 minutes and 53 seconds. The maximum current drawn was 1.8859A, so a charger that can provide at least 2A is recommended. A charger is not provided with the light but you can use your phone charger.

Current Draw

The Sofirn SC31 Pro has a very low parasitic drain of 1mA, with the indicative LEDs on the switch, on (they can be turned off). The light also has a very low moonlight mode that only draws 4mA. The maximum current draw, on Turbo, is 6.11A, which the provided battery is more than capable of.

Output & Runtimes

The Sofirn SC31 Pro is rated at 2000 lumen output and 200m of throw. I measured it at 1899 lumen at turn on (Turbo), 1785 lumen ANSI (30sec), while throw actually exceeded the specs by 10% and measured at 220m (12108cd).

The temperature was very well controlled, as you can see in the runtime graphs below. Turning on the light at Turbo, will cause a rapid rise in temperature, followed by a huge dip in output, while the temperature stabilizes. This can be avoided if the light is not on Turbo, but used at a more moderate output level.

The light was temperature calibrated, according to the manual, and the temperature limit was set at 50C. The actual temperatures on the button and on the body of the light did not exceed 51.5C and 38.2C, respectively.

All in all, the Sofirn SC31 Pro gave very usable light for about 2.5 hours, when turned on at Turbo, which is very respectable. Usage at more moderate levels, will of course, result in higher runtimes and lower, if any, output dips, due to temperature management.

The Full and 10' runtime graphs tell the story in great detail. All significant changes in output are marked.

Conclusion

The Sofirn SC31 Pro is a value for money, EDC torch that will not disappoint. Its aluminium body is well made and hard anodized and the fit and finish is beyond its price point. The mildly sharp edges of the fins and the lanyard hole are not a concern, but could have been smoother.

The size and weight are great for EDC and the provided 3000mAh 18650 battery allows for ample runtime. The output is more than enough for EDC purposes and the beam profile is very balanced, with enough spill and a lot of throw for its size.

The tint of the Luminous SST-40 LED is a cool, above BBL (greenish) tint and the CRI is low, but this is countered by the high lumen output this emitter allows the light to achieve.

The Sofirn SC31 Pro can be purchased from the Sofirn Website and the cost at the moment this review is written is $25.99, plus $2.99 shipping, worldwide. That is a lot of torch for the money!

Let's list the Pros and Cons of the Sofirn SC31 Pro:

Pros

+ Value for money

+ 6061 Aluminium Alloy construction with good fit and finish

+ Hard and flawless anodization

+ USB type C charging

+ Low Voltage Protection

+ Thermal regulation

+ Well balanced beam, good for EDC use

+ 18650 Li-Ion 3000mAh (actually measured) battery included

+ Low power and charging LED indicator

+ Andúril UI

+ IP68

+ At least 2.5 hours of usable light per charge

+ Compatible with all 18650 batteries

+ Removable, pre-installed clip

+ Magnetic tailcap ($1.69), short (18350) tube ($1.49) and diffuser ($1.69) available as optional accessories

Cons

- The fins and the lanyard hole could have smoother edges

- The driver uses PWM to dim the emitter

- Indicative LEDs on the switch are not aligned

 TheLAB.GR

Thanks to Sofirn for providing the light for review

Polymeros Achaniotis 25/06/2021

Brinyte PT18pro Oathkeeper

Mon, 01 Feb 2021 07:37:00 +0000

Introduction

I am not really into tactical torches, as I prefer the EDC type, with many modes and complex user interfaces that the tactical ones usually lack. And to be honest, I was always wondering why can't tactical torches incorporate both the quick, simple, tactical operation and the versatility of more than 1 mode.

Apparently, Brinyte thought the same when designing the Brinyte PT18pro Oathkeeper, which is a tactical torch with simple and quick tactical operation, without sacrificing multiple modes.

In fact, it looks like Brinyte made no compromises when designing this torch and included everything, so the question becomes, how well did they implement all the features and does the torch live up to its ambitious specs? Time to find out.

Unboxing

I received my review sample of the Brinyte PT18pro Oathkeeper in a plain brown box, which is the practice of Brinyte with samples, to minimize the shipping costs. The retail version comes in a nice box.

The light comes with a full accessory pack, which includes a holster, a charging cable, a couple of spare O-rings, a lanyard and the operation manual.

The operation manual comes in a unique format, written on 4 cards, held together with a small clip. It includes all the information needed to understand and operate the torch.

The holster is very well built. It has a Velcro closure, 2 side pockets that can carry extra batteries and a plastic belt loop with a 360 degree swivel.

The Brinyte PT18pro Oathkeeper comes in two colours, black and desert tan. Our sample is the latter and I find it to be a beautiful and practical design. It is definitely too large to carry in a pocket, but that is what the provided holster if for.

It features a large head, a side switch with LED indicator, a removable pocket clip (which is quite superfluous for a torch of this size), a tactical ring and what I will name a "claw ring", with an aggressive claw.

On the opposite side of the side switch, there is a pad for the magnetic charging cable. The tail cap features 2 large, rubber buttons, for the tactical operation.

The big head features a relatively deep and wide reflector, which is almost smooth, promising a beam with a lot of throw but also some useful spill, perfect for a tactical scenario.

The rubber tail cap buttons are easy to use, even with thick gloves. The tail cap also includes a hole to attach a lanyard.

The battery ships in the light itself, with an insulator tab preventing the negative pole of the battery from making contact. This has to be removed before the light can be operated. The battery is a Brinyte branded, button top 18650 Li-Ion battery, rated at 3100mAh.

I think the size of the light warrants the use of the more energy dense 21700 battery type and I hope Brinyte will make that change in the next version. That would provide better run times and the possibility for higher output as well.

After removing the tail cap, the claw ring can also be removed and, after that, the tactical ring, which features a lanyard hole. The clip just snaps on, and can be removed as well.

There are 2 O-rings on the tail part of the light, one for the water proofing of the tail cap and the other to stabilize the tactical ring.

The claw ring is very helpful at gripping the torch and getting it out of the holster both quickly and securely and also provides a very secure grip while operating the light. The aggressive claw is good looking and possibly useful in tactical situations but it will most certainly be illegal to carry in many countries, so Brinyte made it easily removable and the light can be used fully without it.

The part of the claw ring that goes over the tail cap has 2 internal ridges that have to be aligned with corresponding grooves on the tail cap, for proper installation.

The bezel of the light is also removable and the light can be used with or without it. It is not very aggressive and it is useful as it allows you to see if the light is on when it is head standing, so I see no reason to take it off. Maybe Brinyte is planning to release a more aggressive bezel, as an accessory or one made of stainless steel.

The light features thick, good quality springs on both sides of the battery, which should ensure uninterrupted use if the torch is bumped or dropped and also reduce the electrical resistance, maximizing the output.

The light fits in the holster comfortably and there are cut outs for the claw ring and for the part of the tail cap that is protruding from it.

The cut out for the claw ring is much too large, which is strange for a purpose built holster. I wish they had made it a closer fit.

The protruding tail cap, allows for tactical operation of the torch, even when it is in the holster and the cut out at the bottom of the holster allows the light to shine out of it. It seems like the lower part of the holster does not fit the head of the torch well, so the round hole is deformed quite a bit when the torch is in the holster.

Even though the holster has very nice features and good quality, it should be refined, to fit the claw ring and the head of the torch more closely.

Quality

The quality of the Brinyte PT18pro Oathkeeper is that of a premium production torch. The fit and finish is perfect and the anodizing is flawless.

My sample came with some paint chipping on the secondary tail switch housing and 2 small nicks on the back of the tail cap. As this is a review sample that may have been handled before, I cannot say if this is a common occurrence but I expect it is not. I am fairly certain that if a retail torch was delivered like this, Brinyte would replace the tail cap, under warranty.

Specifications

The main features of the Brinyte PT18pro Oathkeeper, as found on the Brinyte website can be seen below:

Features

Brinyte patented tactical ring design, easy to grab at top speed
Compatible with one 18650 battery or two CR123A batteries
Fast charging function
Intelligent power indicator
Regulated power supply maintains constant brightness
Overcharge, over-discharge and overheat protection
Reverse polarity protection prevents damage from improper battery installation
Alloy aluminium reflector with professional optical analysis
Aircraft-grade high strength aluminium AL-6061-T6 with premium Type III military hard-anodized anti-abrasive finish

There is not much one could ask that is not in there.

The technical parameters of the light, as found on the company's website, are as follows:

Technical Parameters

ANSI/NEMA FL1	Turbo	High	Middle	Low	Strobe	SOS
Output	2000+ Lumens	450+ Lumens	60+ Lumens	10+ Lumens	2000+ Lumens	60+ Lumens
Runtime	1min + 90mins	150mins	930mins	1870mins	/	/
Beam Distance	360+ m / 393.70+ yds
Intensity	32000+ cd
Impact Resistant	1m / 1.09yds
Protection Proof	IP68
Working Voltage	3.0 – 6.0V
Dimension	164mm (Length) x 25.4mm (Body Dia) x 41mm (Head Dia)
Net Weight	170g / 0.37lb (excluding battery)

These are some ambitious numbers, for the size and type of this torch. We will certainly put them to the test.

User Interface

The user interface of the Brinyte PT18pro Oathkeeper combines the directness and simplicity that defines tactical operation with the option to have a more complex interface, controlled by the side switch, with more modes and usability.

There is a very clear diagram in the manual that explains the operation of the light, in detail:

Beam-shots

The Brinyte PT18pro Oathkeeper produces a beam with a well defined hot spot that promises good throw. The hot spot fades into a greenish corona that whitens out towards the edge of the spill. There are numerous rings in the beam, which is normal for a smooth reflector.

I tested the Brinyte PT18pro Oathkeeper over a distance of 70m, which is what I would expect a tactical scenario usage would be.

The following video shows a comparison of the Brinyte PT18pro Oathkeeper with 2 very well known contenders, the 1800lm Olight M2R Pro Warrior and the 2200lm Olight Warrior Pro. I think the Brinyte PT18pro Oathkeeper is holding its own very well against the two very strong contenders. I will let each of you decide which one you prefer.

Driver

The driver of the Brinyte PT18pro Oathkeeper provided a steady beam, without apparent PWM on all 4 output levels of the light.

It also features constant brightness levels (as we will verify in the runtime test) as well as overcharge, over-discharge, overheat and reverse polarity protections. That seems like a complete set of features and a very well designed driver

Tint

In the following photos you can see the tint of the Brinyte PT18pro Oathkeeper in comparison to the Olight Warrior Pro and the Olight M2R Pro Warrior.

The pictures were taken with a manually set 5500K white balance. As you can see, the Brinyte PT18pro Oathkeeper has a cool white tint, while both Olight torches feature warmer tints.

Charging

The Brinyte branded, button top, 18650 Li-Ion battery that comes with the Brinyte PT18pro Oathkeeper is rated at 3100Mah. I measured the actual capacity at 3003Mah, using the Opus BT-C3100 smart charger. The internal resistance of the battery was measured at 92mΩ.

The charge level of the battery is indicated on the side switch, as depicted below.

The charging cable has a blue light on it that makes it easy to find in the dark and illuminates enough to help you locate the charging pad on the torch. As soon as you get the charging cable close enough to the charging pad, it snaps on magnetically.

The indictor LED on the side button will turn orange momentarily...

...and then red, as the charging begins. When the charging is completed, the indicator LED will turn to green.

If the indicator light remains orange, that indicates something is wrong. It could be dirt / debris on the charging surfaces or a malfunction.

The under voltage protection will turn the torch off, as soon as the battery voltage drops below 2.8V. The charging from that level until full, took 2 hours, 5 minutes and 15 seconds. The diagram of the charging can be seen below. The charging terminated when the battery reached 4.19V.

1 minute after the light is turned off, the indicator LED on the side switch will turn on, to make it easy to find the light in the dark. The brightness level is perfect for the purpose of helping to locate the light in the dark without being distracting and is comparable to tritium vials.

The power consumption of this function is negligible and it will take many months before it has any serious impact on the battery level. It can be turned on or off by holding the side switch down for 10 seconds.

Current Draw

So let us measure the power consumption of the Brinyte PT18pro Oathkeeper when it is off and also in all output modes.

The clamp meter was calibrated first.

The power consumption when the light is off is just 6mA.

The torch consumes 119mA on low, which is rated to produce 10 lumens.

Medium is rated at 60 lumens and consumes 206mA.

High is rated at 450 lumens and consumes 937mA.

Finally, turbo mode is rated at 2000 lumens and draws 6.5A.

Here is a table with all the measured values and outputs.

Output & Runtimes

I measured the output and runtime of the Brinyte PT18pro Oathkeeper in my home made integrating tube. A high quality electronic thermometer with 2 probes was also use to monitor the temperature. One of the probes was placed near the side switch and the other on the battery tube.

The Brinyte PT18pro Oathkeeper outperformed its own specs in both brightness and duration. The temperature regulation worked flawlessly and kept the temperatures very low, not exceeding 35.2 Celsius.

I think the temperature setting is quite conservative and it could definitely allow for higher temperatures, which would in turn allow for longer turbo runtimes.

The only measurement where the Brinyte PT18pro Oathkeeper fell slightly short of its specs was the light intensity. The Specs say it has an intensity of 32000cd while I measured 31567cd. This means that the throw is 360m in the specs while I measured it to be 355m. This is very close and can be attributed to differences in environmental conditions, which influence the measurements.

Conclusion

The Brinyte PT18pro Oathkeeper is a tactical torch, with dual tactical tail switches, offering Turbo and Strobe modes at the touch of a button. It also features a side switch, with a battery level and charge indicator LED, that allows for 4 output levels, Strobe and SOS functions. That is an excellent combination of tactical and EDC functionality, with no compromises in either.

It features a deep (for the torch size) and almost smooth reflector which provides a well balanced beam with excellent throw and useful spill. The tint is a cool white.

The quality of the machining is very good as is the anodization and the light is IP68 water proof rated. The feel of the switches is very tactile and the tail switches are easy to locate and use, even with thick gloves.

The light comes with a full set of accessories, including a holster, a lanyard, spare O-rings and a pocket clip. It has both a tactical ring and a claw ring, which is easily removable, as it is probably illegal to carry in some countries.

The magnetic charging is very easy to use and the provided battery is a Brinyte branded button top 18650, rated at 3100mAh, which I measured to be 3003mAh. I believe that for this size light, Brinyte could have used a 21700 battery, which would provide even better run times and output. That said, the light already exceeds its specs and the output and run times are very good.

The driver has a full set of every protection possible and provides a stable and PWM free output.

There is not much that can be improved on the Brinyte PT18pro Oathkeeper. The only possible upgrades would be the use of a 21700 battery and a better fitting holster. Well done Brinyte.

The cost of the Brinyte PT18pro Oathkeeper comes to $135, which is justified for the build quality and multitude of features.

Let's list the Pros and Cons of the Brinyte PT18pro Oathkeeper:

Pros

+ Combination of tactical operation dual tail switches and multiple-mode side switch

+ High quality fit and finish

+ Flawless anodization

+ Magnetic charging

+ Constant current driver with full set of protections and no visible PWM

+ Well balanced beam, good for tactical use

+ Removable claw ring and tactical ring

+ Power and charging LED indicator

+ IP68

+ Battery, holster, lanyard and spare O-rings included

+ Compatible with all 18650 batteries

+ Two CR123A batteries can be used if needed

Cons

- A 21700 battery design would be better, for this size torch

- The included holster should have a better fit around the head of the light and a smaller tactical ring cut out.

 TheLAB.GR

Thanks to Brinyte for providing the light for review

Polymeros Achaniotis 01/02/2021

Brinyte T28 Artemis Hunting Flashlight Review

Tue, 22 Sep 2020 06:48:00 +0000

Introduction

Despite living in an age of rapid scientific and technological advancement, there is actually very little innovation going on. Companies focus more on marketing than research and development and "new" products seem more like copies of each other than developed and designed with the needs of the consumer in mind.

As a reviewer I enjoy seeing products with true innovation, products that were developed with some actual thought in the design process. Such is the Brinyte T28 Artemis that Brinyte sent me to review.

The Brinyte T28 Artemis is a hunting light and its design and function are specifically tailored for the job. With an interface that is intuitive and easy to use -with or without gloves- and a triple light source (white, red, and green) without the need to carry extra filters, the Brinyte T28 Artemis is the most complete hunting light anyone could wish for.

Does that mean it is perfect? Well, nothing is and I do not expect this light to be the exception. But let's delve into the review and find out if Brinyte did justice to Artemis, the ancient Greek goddess of hunting.

Unboxing

I received the review sample in a plain cardboard box, but that is not what the consumer gets, as the Brinyte T28 Artemis normally comes in a nice retail package.

Inside the box there is a comprehensive instruction manual, laid out on 3 cards, which are held together with a nice hook and lanyard.

There are also some spare o-rings and a USB type-A to type-C charging cable, which is long enough to comfortably set the battery on a table while charging it and not have it hanging from the socket, as might be the case with the cables some other manufacturers provide.

As this is a light that can be mounted on a riffle with the appropriate mount (not included), a remote switch is included in the box. The remote switch replaces the tail cap of the light, has a spiral cable and includes a button to power on or off and separate buttons for increasing or decreasing brightness.

The back of the switch has an adhesive-backed Velcro patch, so the switch can be mounted and removed freely.

The cable has a stress relief, to avoid damage from movement. The conducting spring is not particularly thick, but it should be sufficient, since the battery is only subjected to modest power draw.

Last but not least, the star of the show, the light itself. A lanyard comes pre-attached to the tail of the light. The lanyard is of nice quality but I would prefer it to be adjustable and have a sturdier tying end.

The aesthetics of the light are nice and seem to be permeated by a faint military air, in a modern way. The anodization is flawless and there are no sharp edges anywhere.

The light can tail stand, but I would not trust it as the head is a lot wider than the tail. Headstands are a lot safer, and the crenellated bezel ensures you can see if the light is on.

The battery ships safely inside the light casing, with one of its terminals insulated by a tab.

The battery is a 21700 Brinyte branded and claims a capacity of 5000mAh. It has a USB C port near the positive terminal and can be directly charged with the provided cable and a charger.

The Brinyte T28 Artemis features a spring on the driver end as well as the tail cap. This ensures the light will not turn off if bumped, as pressure is applied on both ends of the battery. The springs are similar to the one on the remote switch, nice quality but not thick.

This is how the light looks with the remote switch installed.

Quality

There is not much to say about the quality of the Brinyte T28 Artemis. The fit and finish is as good as I have seen in any production light. The anodization is flawless and the controls are responsive and sturdy.

The threads are nicely cut and even though thicker springs are always nice to have, if the light is not drawing much current they do not make an important difference.

The lanyard could (and should) be adjustable and with a thicker tying end, but that is not a major complaint as it is replaceable.

Specifications

The features and technical parameters of the Brinyte T28 Artemis, as found on the manufacturer's website, are the following:

FEATURING

· Powered by one 21700 rechargeable Li-ion battery

· Brinyte patented tri-color lighting sources switch design

· Rotating head to zoom in and out the beam

· Tail stepless dimmer switch design, easy to meet any brightness requirements

· Threaded lens for more effective concentrating effect

· Regulated power supply maintains constant brightness

· Overcharge, over-discharge and overheat protection

· Reverse polarity protection prevents damage from improper battery installation

· Aircraft-grade high strength aluminum AL-6061-T6 with premium Type III military hard-anodized anti-abrasive finish

TECHNICAL PARAMETERS

ANSI/NEMA FL1	Turbo(White)	Turbo(Red)	Turbo(Green)
Output(Flood/Spot)	650+/360+Lumens	140+/55+Lumens	130+/70+Lumens
Max Runtime	135mins	220mins	270mins
Beam Distance	525+m/572+yds	255+m/277+yds	340+m/370+yds
Intensity	69600+cd	16480+cd	29160+cd
Impact Resistant	1m/1.09yds
Protection Proof	IP66
Working Voltage	3.0-4.2V
Dimension	182~196mm(Length) x 25.4mm(Body Dia) x 54mm(Head Dia)
Net Weight	200g/7.05oz (excluding battery)

Unfortunately, I do not have the equipment to measure the brightness of the light to compare to the claims. The weight is a bit more than claimed though, at 216.2g. The battery weight is 73.6g.

The only point where I will have to strongly disagree with the specifications claimed is the "regulated power supply which maintains constant brightness". That is not the actual case as the driver is not a buck/boost driver and the brightness decreased as the voltage of the battery diminished. It would have been much better if the light actually featured a driver that maintained constant brightness until the voltage protection kicked in.

User Interface

The user interface is where the Brinyte T28 Artemis truly shines - pun intended. The light turns on with a 1.5 second depress of the tail switch. This ensures that accidental activation is unlikely. Nevertheless, it is not impossible, as it could happen in a backpack, so I would like to have seen a more secure lockout option. A momentary depress of the same switch will turn the light off.

The brightness level is controlled by twisting the same switch, like a volume dial. It controls the brightness in a step-less way, from 2% to 100%. Step-less rotary control is my preferred user interface in any light, as it is the most intuitive way to accurately dial in the exact amount of light needed, something that can be important in hunting.

A green LED next to the button confirms the light is working normally, which is a nice touch. Its brightness is low enough so it is not distracting.

The button / dial is large enough to use with thick gloves, which is essential for hunting and verifies that a lot of thought went into design of the Brinyte T28 Artemis.

The remote switch also has on / off and dimmer functions. On / off is controlled by a momentary depression of the central switch. + and - control the brightness.

The sturdy, military styled selector at the front is the means to select the LED used. The Brinyte T28 Artemis has 3 LEDs inside, and this selector actually moves the one needed to the centre, behind the lens. That is a very clever design, as other hunting lights have to either rely on add-on coloured filters or separate reflectors / optics for each colour LED, limiting the optic size available to each LED and / or increasing the size of the light.

The unique design of the Brinyte T28 Artemis that moves the required LED in place, allowing for the use of a single, large optic, is optimal.

But not only is the Brinyte T28 Artemis a step-less rotary control light with 3 separate LEDs, it is also a zoomie. So, twisting the head will zoom in the light, focusing the beam and increasing the throw distance.

The controls of the Brinyte T28 Artemis combine to create the most comprehensive, intuitive, easy to use and complete user interface I have seen to date.

Beam-shots

The Brinyte T28 Artemis uses a Fresnel lens. This will provide the nice, uniform beam of a conventional spherical lens, at a fraction of the size and weight.

Behind the lens, are 3 LEDs, a white in the middle, flanked by a red and a green LED. The relevant selector with mechanically move the selected LED at the centre, behind the lens and connect it to the driver.

In that way the Brinyte T28 Artemis can produce white, red or green light, using the full size of the lens.

The beam in full flood is not very floody. As the light is meant for hunting, it will be used from middle to long distances, so it makes complete sense that it is calibrated for those distances.

In full zoom, the beam is very concentrated.

Here are the same, on a flat white surface. In full flood the Fresnel lens creates a very uniform beam, with a defining circle around it. The LED itself can be just spotted in the centre.

In full zoom, some artefacts are visible, but that does not show in the real world, at long distances, where the full zoom would be used.

Here is a short video, using the Brinyte T28 Artemis outside.

Driver

The driver of the Brinyte T28 Artemis uses PWM to control the brightness, even at full brightness. That is to be expected and the PWM was not visible to my eyes.

The shortcoming of the driver that needs to be addressed is the fact that it is not regulated, meaning the brightness of the light is dependant on the battery voltage. As the battery charge is used up and the voltage drops, the light becomes dimmer.

Tint

The tint of the white LED in the Brinyte T28 Artemis seems to be an above the BBL, cool CCT. Here it is compared to some other lights. From left to right we have the Brinyte T28 Artemis, the Olight S2R Baton II, the Sunwayman V20C and the Fireflies E07 Ti/Cu.

The beam-shots are with the Brinyte T28 Artemis in full flood, in mid zoom and in full zoom.

And here is how the colours compare to the white lights.

Charging and Runtimes

To charge the Brinyte T28 Artemis, you have to take out the battery and plug it into your own charger, using the provided cable. A red indicator light in the centre of the positive terminal will signal the charging. It will turn blue when the charging is completed.

The battery is a Brinyte branded 21700 lithium-ion with a claimed capacity of 5000mAh, which I actually measured to be 5040mAh, so right on the money.

Voltage protection kicked in when the battery voltage dropped to 2.68V. Charging it from that level, took 3 hours, 56 minutes and 25 seconds. The charging terminated when the battery reached 4.19V.

The maximum current drawn from the charger was 1.4248A, so your charger must be capable to provide at least that. If not, the battery will still be charged, but charging will take longer.

The runtimes specified and measured are in the table below. The ones measured are by far greater than the specified values. It is apparent that the specified runtimes refer to where the light produced is bright enough to be usable in hunting. As the driver is not actually regulated, the brightness will drop with the voltage of the battery, so the light, after a point, will not be bright enough to be usable. I hope Brinyte will address the issue and make a new version with a buck / boost driver, as this is my only grievance with this light.

The light may have gotten dimmer as the voltage of the battery dropped but there was no visible step down due to thermal regulation. That is to be expected from a 650lm light of this size. The highest temperature measured during the run-time tests was 45.6C.

Current Draw

The current drawn from the battery of the Brinyte T28 Artemis was measured in all configurations using a digital clamp meter. The calibration for both the tail and the remote switches was done first.

Then the parasitic drain was measured, which is very low.

Here are the measurements for the lowest and highest settings, using the white LED, for both switches.

Here are the measurements for the lowest and highest settings, using the red LED, for both switches.

Here are the measurements for the lowest and highest settings, using the green LED, for both switches.

The table below contains all the measurements, in Amps.

Conclusion

The Brinyte T28 Artemis is a very well designed and executed hunting flashlight. It features white, red and green LEDs, a step-less rotary control and a zoom function with a Fresnel lens. Its IP rating of IP66 ensures protection from dust and powerful water jets, which is as much as anyone can expect from a zoomie and good enough for all hunting conditions.

The light is solidly built, with good fit and finish and a flawless anodization. I wish the logo on the tube would align with something on the head of the light, but that is actually surprising expensive to ensure and definitely not worth the increase in price it would cause.

The user interface is the most complete, comprehensive and easy it could be and designed to be used with thick gloves, if needed.

My only grievance is that the driver is not actually regulated and brightness is directly dependant on battery voltage. It would be so much more useful and nice if the light could actually maintain constant brightness, regardless of the battery level.

The cost of the Brinyte T28 Artemis comes to $145.50 which is justified for the build quality and multitude of features. It would be even more so, with a better, fully regulated driver.

Let's list the Pros and Cons of the Brinyte T28 Artemis:

Pros

+ Extremely intuitive and complete user interface

+ High quality fit and finish

+ Flawless anodization

+ 3 separate LEDs for white, red and green, moved into place at the centre behind the lens by a mechanical selector

+ Fresnel lens

+ Step-less rotary control

+ Zoom function

+ Remote switch for power and brightness control included

+ IP66

+ 21700 USB C rechargeable battery with 5040mAh actual capacity

Cons

- Despite claims for a regulated driver that provides constant brightness, the brightness was actually dependent on battery voltage

- Accidental activation is unlikely, but not impossible

- The included lanyard should be adjustable and have a sturdier tying end

 TheLAB.GR

Thanks to Brinyte for providing the light for review

Polymeros Achaniotis 24/09/2020

OCZ Vector 180 SSD Review (in English)

Wed, 25 Feb 2015 14:31:00 +0000

Introduction

Today, there is nothing common about our torture subject. No, ladies and gentlemen. Today we have a drive of noble ancestry which will have to work hard to fill the shoes of its predecessor.

OCZ, in the course of renewing its consumer SSD line, is replacing the successful Vector 150 with the new OCZ Vector 180.

So, what is it with all the naked cables, the lightning and general mayhem in the image? Just a marketing stunt?

No, definitely not! They are related to a new function of the OCZ Vector 180 which sets it apart from its predecessor. And what would that be, you might ask. Yeah, right, like I would tell you everything in the introduction!

So, let us begin with the torture, because as the saying goes, when the going is tough, good SSDs get going! Or something to that effect.

And first of all, Anatomy 101!

From the packaging to the Silicon - What you get

The packaging of the OCZ Vector 180 is the standard OCZ, illustration paper box, protected by nylon, which we removed so you can see the box better in the picture.

- The "box"; Are you drunk? There's two of them!

All right, all right, I am not (very) drunk! There are indeed two boxes as OCZ sent us two different capacities of the OCZ Vector 180 to test for you. So, just for you, the readers of TheLab.gr, today's review will include the generous capacity of 480GB and the even more generous capacity of 960GB!

Going on with the packaging, the front is minimalistic and elegant, with a picture of the drive and its model, while a sticker on the lower right declares the capacity.

The back is more exuberant and includes information about the basic specifications of the drive and the performance of the various capacities. It also includes its certifications, two stickers with the model, the serial number and bar-codes as well as the claim, in 12 languages, that the OCZ Vector 180 can replace advantageously conventional Hard Disk Drives.

Excellent packaging, you might say, but lets move on to the juicy stuff! OK, at once!

Opening the packaging, we find the drive protected in a case of hard see-through plastic, as it is the norm of OCZ and with it an assortment of accessories worthy of its noble ancestry. That includes a 2,5" to 3,5" adapter, screws, a comprehensive installation manual in many languages and two leaflets, one of which includes instructions to download and install Acronis True Image 2013 as well as the required activation code.

You thought I forgot the second leaflet, right?

WEll, here it is, from both sides! This leaflet gets special attention as it is about the warranty of the OCZ Vector 180 which starts with this leaflet prompting you in 11 languages to not return the drive to the retailer, in case of malfunction, but to contact OCZ directly. You will find out why in the next page of this review.

Let us set aside for now the nice accessories and the papers and cut to the chase! Minimalistic and beautiful are two words befitting the OCZ Vector 180, with the top being common for all capacities.

The sticker on the bottom has information as the model, the product code, the serial number, certifications, the country where the drive was made(Taiwan), the firmware it comes with, as well as the voltage and current it requires.

The four screws in the corners are the ones we obviously have to remove to gain access to the interior of the drive. One of them is covered by a sticker warning of warranty loss if it is removed. There goes the sticker, there goes the warranty with it and with a simple Philips screwdriver we remove the four screws.

Four more screws keep the PCB of the OCZ Vector 180 firmly attached to the top of the chassis and a thermal pad connects the controller of the drive to the cover we just removed, ensuring cooling for the controller.

We remove these screws as well and gain access to the other side of the PCBs.

Please take note that in all pictures of both drives, on the left we have the OCZ Vector 180 480GB and on the right the OCZ Vector 180 960GB.

Under the thermal pad, on both drives, lies the well known Indilinx Barefoot 3 M00, marked IDX500M00-BC. It is the exact same controller we find in the στον Vector 150 and AMD R7, which is made by OCZ and it runs at 397MHz.

The cache memory for the controller is a chip by Micron on each side of the PCB, so we get two chips for the OCZ Vector 180 460GB and another two for the OCZ Vector 180 960GB.

These chips even though they come from different batches, as declared by the first number on them, are of the same type, specifications and capacity, as declared by the second number on them. They are the Micron D9QBJ which are DDR3 1600Mhz CL11 chips with a capacity of 4Mbit, that is to say 512ΜΒ each. This means that each drive sporting two of these chips, has 1GB of cache memory, the exact same of the 480GB Vector 150.

Where the two drives differ, is the capacity of their NAND chips. As you have seen, each drive has 16 NAND chips, 8 on each side of the PCB. The difference is that the OCZ Vector 180 480GB has Toshiba TH58TEG8DDKBA8C A19 256Gbit NAND chips while the OCZ Vector 180 960GB has Toshiba TH58TFG9DFKBA8K A19 512GBit NAND chips.

Let us do some math, so you can confirm that I am completely sober.

On the 480GB drive, we have: 16 x 256Gbit = 16 x 32GB = 512GB total. The drive offers 480GB to the user, so the missing 32GB, one of the 16 NAND chips ή or 6,25% of total capacity is reserved for the over provisioning of the drive.

On the 960GB drive, we have: 16 x 512Gbit = 16 x 64GB = 1024GB total. The drive offers 960GB to the user, so the missing 64GB, one of the 16 NAND chips ή or 6,25% of total capacity is reserved for the over provisioning of the drive.

And last, we see the part of the PCB containing a big orange SMD part. It is a capacitor, with a large, for PCB standards, capacitance, which is related to the new function of the OCZ Vector 180. Are you thinking lightning and general mayhem? Be patient! You will soon know!

The observant among you will have noticed that on the right PCB, the one from the OCZ Vector 180 960GB there is a Voltage Step Down chip and a coil missing. It is because this capacity of the drive does not need them.

After mentioning the excellent quality we observed , let us proceed to the next page to find out what all these parts can do!

Specifications - What is this thing capable of?

What could anyone do to improve on the already excellent specifications of the OCZ Vector 150? Certainly not much as it is already pushing the limits of the SATA 6GB/s protocol. Then what are the improvements that justify replacing it with a new model?

As you can see in the following table, the OCZ Vector 180 brings a new capacity not available with its predecessor, 960GB. It also replaces the Toshiba 19nm NAND chips with the newer Toshiba A19 NAND chips which, albeit slightly slower, they are cheaper raising expectations for more competitive end products. The 5 year warranty remains but it is upgraded to ShieldPlus, which we will elaborate on in the next page. Finally, there is an added function, called Power Failure Management Plus!

So, that is what all the lightnings and mayhem was about. Now, what it is and what it does exactly, you will see in the next page.

Let us start the specifications of the OCZ Vector 180 with a table showing the performance of its various available capacities. Note that the best performing model is the 480GB which is the largest capacity using 256GBit NAND chips, while the 960GB model, using 512GBit NAND chips, comes second.

In this table we see the reliability features of the OCZ Vector 180. And first of all the encryption features which are 256-bit AES-compliant and support only ATA passwords while the industry is turning to the safer and easier to use TCG Opal encryption. Be that as it may, desktop users will be fine with it.

Continuing down the table, we see the ECC error correction of the OCZ Vector 180, the support of S.M.A.R.T. and the fact that the drive is rated for 50GB of writes per day for 5 years.

Here we can see the physical characteristics and parts of the OCZ Vector 180, which we saw in detail in the previous page.

Here we see the power consumption of the [OCZ Vector 180 and the high tolerances it has to environmental factors, as all SSDs do.

It is noteworthy that the OCZ Vector 180 sadly does not feature low power states (slumber and DevSleep), so its power consumption in idle state is comparatively higher than the competition.

In the next table, we have the MSRPs for the various capacities of the OCZ Vector 180. Notice that the recent aggressive pricing policy of OCZ is also applied to the OCZ Vector 180. The price per GB is highly tempting for such a premium drive, especially as capacities rise.

Following is the placement of the OCZ Vector 180 in the OCZ consumer SSD product line. You can also see the most important specifications of the various models and their cost.

And last, a table comparing the specifications and cost of the OCZ Vector 180 to that of some of its major competitors. The price of the OCZ Vector 180 is the MSRP, as it is not out in the shops yet, while the prices of the competing products are from newegg.com on 2/2/2015.

Let us now proceed to the next page to answer any questions that may have risen regarding the special features of the OCZ Vector 180.

Power Failure Management Plus & ShieldPlus Warranty - What the others do not have

Let us see at last what exactly is this Power Failure Management Plus (PFM+). As you can see in the following diagram, thanks to special routines in the firmware but also to that big orange capacitor we saw, the OCZ Vector 180 can protect itself and your data from power abnormalities or even total power loss.

We have to clarify that, as mentioned in the diagram, the protection covers the resting data and not the in flight data.

An excellent feature, that is truly worth the trouble of creating a new model. But it is reasonable to ask, could such a feature, requiring constant and periodic checks, have an impact on performance?

The answer is yes. And no.

As you can see on the screenshot of the AIDA 64 Average Write Access, there is one and only one but fairly large latency spike during the testing of the OCZ Vector 180 960GB. A similar latency spike, but on a smaller scale, is found of the test of the OCZ Vector 180 480GB. This latency spike is not particularly noticeable, but it is measurable and as you will see in the tests we will subject the two OCZ Vector 180s, it can in some cases cause delays.

This is the price the OCZ Vector 180 has to pay in order to provide the important extra safety of the PFM+. Let us also not forget that the drive is still using its first version of firmware and it is possible that the effect could diminish in the future with a firmware update.

Let us now address the other point where the OCZ Vector 180 has the edge over its predecessor which is no other than the warranty.

OCZ is launching a new kind of warranty in the SSD market which promises to make users' lives easier. According to this new warranty which OCZ calls ShieldPlus, you no longer need any proof of purchase for the RMA procedure as the serial number of the product is enough. But this is not even the best part. In case of the product malfunctioning and after you contact OCZ and start the RMA procedure, a new SSD is forwarded to the consumer by the company. After the consumer receives the new drive he sends the malfunctioning one back to the company at OCZ's expense.

All we can say is a big well done to OCZ for the warranty of the OCZ Vector 180 and that we wish other companies would follow the same path. Such a warranty cannot but point out the great confidence OCZ has on the quality of its product.

Speaking of which, OCZ has finally released the official figures on drive failures in recent years and we can see that from 2010 to 2014 there was an incredible increase in drive reliability which seems to be due to the cooperation of OCZ with Toshiba and the usage of the latter's NAND chips.

So, these are the extra features of the OCZ Vector 180! Let us now go and see the Software OCZ offers with it.

Software - Complete and Renewed

Along with the OCZ Vector 180, OCZ is releasing their new software, SSD Guru.
The software includes all the functions you will need to perform on you SSD Drive. We will display and explain its capabilities.

When you first run the software you will see the screen above and in the horizontal menu the first tab "Overview" will be selected. On the upper left of the tab, there is a drop down menu where you can select any OCZ drive available in your system. Non-OCZ Drives will not appear.

Choosing a drive, we can see its picture below as well as the Model and Serial Number.

Below that, we can see the four sub-tabs with the first, "Dashboard", already chosen.

On the right there is a bigger windows with five modules. In the module "Capacity" we can see the capacity of the drive as well as details on how much of it is free and how much is used. In the drop down menu we can choose if the statistics apply to the whole of the drive or just one of its partitions.

In the module "Interface" we can see the interface protocol, which idealy is SATA III (6GB/s) and if the AHCI is enabled. The color of the dots change from green to orange or red acordingly to the severity of the non-ideal situation, which generally happens on various parts of all the modules and is explained by clicking the round "?" button in each module.

The round "?" button in each module explains their functions in detail as well as the various color codes they may employ.

The module "Health" shows the Health of the drive and the color denotes if the drive is in good health (green) , needs to be monitored for future replacement (orange) or needs to be immediately replaced (red).

The module "Updates" shows the firmware of the drive and reports if it is up to date.

If the drive has a temperature sensor, as in the case of the OCZ Vector 180, there is one more module, "Temperature" which provides a graphical display of the drive's temperature.

In the sub-tab "SSD Details" we can see details for the chosen SSD drive.

In the sub-tab "System Details" we can see details for our whole system.

In the sub-tab "SMART" we can see in a clear and easy to understand way, some of the S.M.A.R.T. details of the drive.

We continue with the other tabs.

The second tab of SSD Guru is "Tuner" and has only one sub-tab, "SSD Tuner".

This includes two modules, "Trim", where you can give a manual Trim command, and "Over Provisioning" where you can assign more Over Provisioning space. And we say "more" as all OCZ drives already have 6,25% of their space assigned to Over Provisioning, which you cannot remove from this module but you can add more.

The third tab of SSD Guru is"Maintenance" and has only one sub-tab, "Tools".

This includes two modules, "Firmware Update" where from the drop down menu you can choose to upgrade the firmware through the internet, if there is a newer firmware for the chosen drive, or from a file on our local disk, and the module "Secure Erase" where we can give a secure erase command to restore the drive to its original factory state.

The fourth tab of SSD Guru is "Settings" and includes only one sub-tab, "Settings".

This includes three modules. The first, "Logging" provides the ability to activate and deactivate the logs of SSD Guru. The second module, "Monitoring", includes to options to load the SSD Guru with Windows as well as to keep it running in the system tray when you close the GUI. The last module in the tab is "Proxy server" which gives us the ability to assign a proxy server for the communication of the SSD Guru with the internet.

The fifth and last tab of the SSD Guru is "Help" and includes only one sub-tab, "Help".

This includes two modules. The first is "Contact OCZ" and includes links to the support of OCZ as well as its Forums. The second module is "Save support package" where we can create a file with the required data to send to the support of OCZ. This is very convenient, especially to the less experienced users.

On the whole, we can say that SSD Guru can simply and easily perform all the functions needed by an SSD User. The only thing missing, is an interactive guide to improve Windows setting for use with an SSD Drive.

Test System and Methodology - Let's see, what are we using and how

The CPU is the Intel G3258 Anniversary Edition, with unlocked multiplier which allowed us to easily overclock it to 4,5 GHz. Keeping in mind that the SSD test with high IOPS require hick clocks but run on only one thread, this particular CPU is perfect for our tests.

The RAM is at 8GB which more than covers our needs for the tests, The PSU is the V650 by Cooler Master and the system is housed by the Cooler Master Test Bench V1.0.

Our warmest thanks toAsrock and Cooler Master for the parts they provided for the construction of our test system!

The operating system on which we run our tests is Windows 7 64bit and it includes, aside of all the drivers, the basic programs that come with the motherboard so we can achieve a more realistic and close to fact test environment. The driver of the SATA controller used was the Intel 13.10.1058, obviously in AHCI mode.

The tests which the two OCZ Vector 180s will undergo are divided into 4 categories:

1: Synthetic benchmarks on an empty drive, where the tested SSD is a secondary drive to the test system, empty and relaxed. These are the perfect conditions you will never see in real usage.

2: Synthetic benchmarks on a system drive, where the operating system is installed in the tested SSD. Here things come a little closer to reality, but synthetic benchmarks are still synthetic.

3: Steady State Benchmarks where we measure the performance of the SSD while it writes random data for a prolonged period of time so it reaches its worst possible performance. Here things are the opposite of before and thus very unfavorable and unlikely to appear during the usual home usage. But they are the rule in database and file servers.

4: Real Life Tests! Here, using the special suite created for us by the programmer Stelios Sifakis (whom we thank!), we perform some of the most usual tasks and time them. In these tests we place the most gravity for the score as they show the real experience a user will have.

The SSDs included in our comparison are the AMD R7 240GB, the Corsair Force LX 256GB, the Kingston SSDNow KC300 240G, the Kingston SSDNow V310 960GB, the OCZ ARC 100 240GB, the Samsung 840 EVO 250GB, the top performing Samsung 850 Pro 256GB and of course, our leading actors, the OCZ Vector 180 960GB and OCZ Vector 180 480GB.

At this point we would like to note the following. Due to the special nature of the OCZ Vector 180 and because of the PFM+, in some runs of certain tests that happened to coincide with a latency spike, we got abnormally low readings. These readings were rejected and not taken into account in the mean values, because to properly incorporate them to the statistical analysis we would have to run each test thousands of times, which is not possible. So you should know that the mean values you will see in the charts do not include the effects of the latency spikes, except if noted otherwise.

Stay tuned. After the tests there is commentary and score!

Synthetic Benchmarks in an Empty Drive - We start easy

We start with the HD Tune Pro 5.50 which measures the reading and writing IOPS in unpartitioned SSDs. Lets see how the OCZ Vector 180 fares against its competitors:

In the read tests the OCZ Vector 180 is the best in both tested capacities, while the 480GB model seems to have even more IOPS than the 960GB one.

The write tests gave similar results.

We continue with the Average Access tests of AIDA 64 4.50. With the AIDA 64 Average Read Access and Average Write Access, executed in a drive without partitions we see the response time of the SSD to various requests. In other words, we measure its latency. Here, lower is better.

In the Average Read Access test, the OCZ Vector 180 performs admirably and in among the fastest drives.

In the Average Write Access test the performance is also great, but we also see the one and only one latency spike which is larger and more noticeable in the 960GB model.

Following is the AS SSD Benchmark, version 1.7.4739.38088.

In the read tests OCZ Vector 180 is exceptional, right on the heels of the top performing Samsung 850 Pro 256GB.

But in the write tests it takes the lead, even by a small margin, as both OCZ Vector 180 models showed exemplary performance.

In the copy tests, which are a mixture of read and write, the performance of the OCZ Vector 180 was also among the best.

Following are the tests of the CrystalDiskMark 3.03b x64.

In this test we also measure excellent performance for the OCZ Vector 180 as it is only slower than the two Samsung drives, while...

on the write tests it ties with the Samsung 850 Pro 256GB and the AMD R7 240GB for the first place.

Moving on to the score from the Anvils Storage Utilities 1.1.0.

In this test the OCZ Vector 180 is also slower than the two Samsung drives, mainly due to differences in performance in the Read tests.

We now come to the suites of Futuremark whom we older guys remember as MadOnion. These suites try to simulate real working conditions. Lets start with the PCMark 05.

Here OCZ Vector 180 holds its own in all tests but the Virus Scan which is the known Achilles heel of all OCZ drives we have tested to date.

Going on to the more contemporary PCMark7.

Nothing but good results for the OCZ Vector 180 which is second only to the Samsung 850 Pro 256GB. We also see a difference between the two capacities of the OCZ Vector 180 in the "Starting Apps" test, where the OCZ Vector 180 480GB is faster than the 960GB version.

And we conclude this series of tests with the most recent PCMark8, where the results are in seconds and lower is better.

Here we have nothing special to notice, except the good performance of the OCZ Vector 180 480GB in the "Photoshop Heavy" test.

Good results for the new nobility of OCZ. Let us proceed to the next page and take things up a notch.

Synthetic Benchmarks on a System Drive - Things are getting interesting

We spice things up a bit, running the same benchmarks but this time the operating system is installed on the SSD tested.

Once again we start with the HDTune Pro 5.50 but this time obviously on a partitioned drive and we only run the read test as the write test is destructive and would have erased all data on the drive.

The existence of the operating system on the drive, limited the performance of all tested drives, but in doing so, it also gave the OCZ Vector 180 a bigger lead over its competitors. This drive seems to be having excellent response times.

We continue with the Random Read Access of AIDA64 4.50, on a partitioned drive now. The Random Write Access, being destructive, is omitted.

And we confirm our previous findings as the OCZ Vector 180 has unbelievably low latencies!

Going on to the AS SSD Benchmark version 1.7.4739.38088, where there is nothing to tell, as the results are on par with the ones we saw in the previous page where the OCZ Vector 180 did not have an operating system installed on it.

Following are the tests of the CrystalDiskMark 3.03b x64, where the same thing happens.

Continuing with the Anvils Storage Utilities 1.1.0, on which we see no notable difference from the test without the O.S.

Following are the tests of Futuremark.

In all the tests of PCMark05, PCMark7 and PCMark8 the results are similar to the ones on the empty drives. If you don't believe us, here they are:

And we finish with BootTimer which measures the time from the loading of the windows kernel until the desktop is usable.

Both models perform similar to each other and the AMD R7 240GB, slightly behind the Samsung 850 Pro 256GB and the Corsair Force LX 256GB.

Similar results to the previous page for the OCZ Vector 180 that is steadily in the higher ranks of the synthetic benchmark results.

Let us now hit the tough road!

IOMeter Steady State Tests - We press it, can it take it?

The next test is purely sadistic. We make the SSD write random 4K packages for 12 hours straight without resting and measure its IOPS every five minutes. After some time and as the garbage collection and trim features do not have time to work, the SSD reaches a steady state where every write has to preceded by a deletion of the NAND cell and the SSD operates at its worst performance.

This case is very unlikely to occur in light, casual usage but it is the norm in database and file servers. You should consider the following results according to the heaviness of your workload .

OCZ takes pride in the performance of its SSDs in this particular test and the OCZ Vector 180 is no exception. The Steady State IOPS of the OCZ Vector 180 are among the best and near them are only other OCZ models.

Moreover, due to the large capacities of our samples, they can maintain their top performance for longer before the curve starts to drop, something you can clearly see on the left part of the chart.

observing the following diagram, where we see the last two hours of the Steady State Test, we see that the OCZ Vector 180 480GB holds at slightly above 22000 IOPS in Steady State, just shy of its 23000 IOPS specification and the AMD R7 240GB which exceeded its 20000 IOPS specification.

The OCZ Vector 180 960GB on the other hand falls to 18200 IOPS, significantly lower than its specification which is 20000 IOPS and even less than the entry level ARC 100 240GB which exceeded its 18000 IOPS specification.

It is the first time we see an SSD by OCZ not living up to their Steady State IOPS Specifications and we do not know if it is due to the newer and cheaper but also slightly slower NAND chips, the PFM+ feature or that they are still on their first firmware, but in any case the Steady State IOPS of these drives are still far better than the competition. We remind you that the OCZ Vector 150 480GB has a 26000 IOPS specification in Steady State.

Following is the mixed workload test. SSDs in a system rarely do 100% reads or 100%% writes like in the synthetic benchmarks. In the following test we start with three five-minute tests of 100% random reads and keep the average. We continue with three five-minute tests of 95% random reads and 5% random writes and keep the average and so forth as indicated in the graph until we get to 100% random writes. Note that this particular test is run on a drive that has already reached steady state and tries to simulate heavy multitasking.

Here, as was expected, the OCZ Vector 180 480GB marginally bests the AMD R7 240GB while the OCZ Vector 180 960GB fares similarly to the OCZ ARC 100 240GB.

Finishing these very hard tests, we have to stress that it is a very rare occasion to find such extreme conditions in normal everyday use. Their value is to show us the other side of the synthetic benchmark coin and let us try to find the truth somewhere in the middle.

But instead of guessing we have a better suggestion. The suite of Real World Tests is coming to clear the field, with true tasks in windows and exact timing.

Real Life Tests - How is it in real life?

We start with the windows startup time test. One might ask, didn't we already do that with BootTimer? As we said, BootTimer measures the time from the loading of the windows kernel until the desktop is usable. But windows itself records in the event viewer the total time of the windows start up, until all services are loaded. As reason indicates, this time is significantly longer.

In Starting Windows the two OCZ Vector 180 achieved the second and third best times, after the Kingston KC300 240GB that had managed a truly unbelievable time.

We continue with real file copying.

We copy 32000 4K files, 500 MP3s, 1000 21MP JPGs, a 7,52GB ISO 7.52GB and a 10GB High Definition MKV movie. The respective times are shown in the diagram below with the OCZ Vector 180 having slightly better times than the AMD R7 240GB in the first three tests but slightly worse in the bug file copying tests. There, it seems that the latency spike issue is effecting the results. On the whole, the file copying times are exceptional.

Next, we will store a folder containing the setup of the Office 2013 in a rar file (using the "store" setting), where the OCZ Vector 180 has the best time of all the drives and...

...then we will decompress a - compressed - rar containing the setup folder of Office 2013, making, again, the best time.

We continue with a virus scan of the system32 folder of windows using the Avira Antivirus, where the drives by OCZ traditionally do not do well. We see that the OCZ Vector 180 has less of a problem, but it still isn't as fast as some of the competitors.

Next we will open a 324MP TIF file with Photoshop CS6 save it as BMP and close Photoshop. In this test the OCZ Vector 180 480GB and the OCZ Vector 180 960GB share third place, right behind the drives from Samsung, but with negligible difference.

Then we open Word 2013, load a huge DOCX with images links etc, save it as DOC and close Word.

Here the OCZ Vector 180 960GB achieved the best time with the OCZ Vector 180 480GB slightly behind, but the differences between all drives are small to negligible.

Next, we install MS Office 2013, where the latency spikes again make their presence known and are unavoidable as a single run lasts more than three minutes. As a result we see the not so great performance of the two OCZ Vector 180 in this test, especially the 960GB model. We have to mention that we in fact saw even slower times in some runs of the particular test, which we could not incorporate in the mean value we use in the chart, for statistical reasons.

Thus, and with the significant influence of this last test, the total time of the Real Life Tests is as follows:

Not at all bad, as the OCZ Vector 180 480GB got the second place, behind the unbeatable Samsung 850 Pro 256GB, while the OCZ Vector 180 960GB got the fourth place, due to the larger latency spikes.

It is certain that the OCZ Vector 180 is not an easy drive to review. How does one evaluate the all the positives and some negatives we saw?

Lets proceed to the last page to sum it all up.

Conclusions and Score - To the point

Life is full of surprises and delicate balances. The new nobility of OCZ surprised and troubled us and now we have to find the balance.

It is a drive with exceptional build quality, a renowned controller, a lot of cache, and tested, very good NAND chips.

It also incorporates the innovative, for a consumer SSD, PFM+ feature that protects the drive and your stored data from a catastrophic power loss but also has its price in the form of latency spikes.

In the same time, it offers the wonderful new warranty by OCZ, the ShieldPlus, according to which any drive replacement that is needed during the warranty period will be done at your home, with simple and quick procedures.

As far as Features go, the OCZ Vector 180 loses the top mark of 10 by just a little because its controller does not support TCG/Opal encryption and the low power state DevSleep.

In Performance, we will be even more harsh and deduct points for the subtle but sometimes noticeable latency spikes and for the fact that we could not achieve the specified Steady State IOPS. Here the OCZ Vector 180 960GB loses more points than the OCZ Vector 180 480GB as the larger capacity model has larger and more noticeable latency spikes.

Regarding the price, we do not have a specific street price, as the drive has not hit the stores yet, but if you take a look at the MSRPs in page three of this review (all those who skipped to the end, can now go back), they are just great for a drive of this class. And as the capacity rises, the price per GB gets even better. Thus, in price, the OCZ Vector 180 960GB covers the ground it lost to the OCZ Vector 180 480GB in performance.

Thus, we will have a common score for both OCZ Vector 180 capacities we tested.

Lets sum up the advantages and disadvantages of the OCZ Vector 180:

Advantages
+ Very good results in all Benchmarks
+ Very good results in the Steady State Tests
+ Very good results in the Real Life Tests
+ Power Failure Management Plus (PFM+) feature
+ 5 year ShieldPlus warranty
+ Specifications for 50GB / day for 5 years
+ Support for TRIM and Garbage Collection
+ Support for AES 256 encryption
+ Complete and functional Software

Μειονεκτήματα
- Latency Spikes
- Lower than specified Steady State IOPS
- No support for TCG/Opal
- No support for DevSleep

Based on all of the above, the final score of the OCZ Vector 180 is:

TheLab.gr

Our thanks to OCZ for supplying the review samples

Polymeros Achaniotis 24/03/2015

AMD R7 240GB SSD Review

Wed, 22 Oct 2014 08:10:00 +0000

The brand name AMD is known to everyone who knows anything about computers. It automatically brings to mind CPUs and Graphics Cards. But today, this name belongs to an SSD, which with profuse courage climbed on to the torture bench in the dungeons of TheLab.gr, in a dark chamber with chains hanging from the walls, filled with strange instruments, right next to the Reviewers' Office.

Courage, said we? Audacity, is the correct term! Granted, AMD is a big company, but how can their first attempt in the difficult SSD field have such grit? There is a mystery here which awakens my inner Sherlock Holmes! We promise to get to the bottom of this case and solve the mystery of the AMD R7 240GB.

Come Watson, let us proceed to the next page!
[break=From the packaging to the Silicon - What you get]
AMD R7 240GB comes in a paper box, dominated by red and black. It is enclosed in plastic wrap which ensures you will be the first to open the paper box.

Without the plastic wrap, thing become clearer and you can, if you click on the pictures and see the HD version. read details about the product.

Starting from the front of the package, we can see the logo of AMD, the SSD itself, the fact that this drive belongs to the R7 series and that it has a capacity of 240GB. At this point we have to comment that the R7 series by AMD comprises Low-Range to Mid-Range products. So, it stands to reason that this is what we should expect from the AMD R7 240GB. And we wonder, will there be an AMD R9 SSD?

On the back, on the left side we can see some of the characteristics of the reviewed SSD, which we will go into later, but for now lets keep that it has a SATA III 6GB/s interface, 7mm height to fit inside ultrabooks, supports TRIM and comes with a 4 year warranty. On the right we can see in 12 languages, an effort fo convince us that SSDs are better than HDDs.

There are also some stickers with bar codes and serial numbers and, of course, on the bottom left the logo of the company, OCZ.

Wait a minute Watson! OCZ Storage Solutions, A Toshiba Group Company! That's the secret of the AMD R7 240GB! It is in fact an SSD by the experienced OCZ! The plot thickens!

Inside the paper packaging, the AMD R7 240GB is accompanied by its generous accessories. We can see the drive itself in a clear plastic protective case, a 3.5" adapter and three leaflets.

The first leaflet concerns the warranty of the AMD R7 240GB, urging us in case of malfunction to not return the product at the retailer where we purchased it but to visit their technical support and RMA gate at the written web address.

At this point we will go into some details about the warranty. OCZ is launching a new kind of warranty in the SSD market which promises to make users' lives easier. According to this new warranty which OCZ calls ShieldPlus, you no longer need any proof of purchase for the RMA procedure as the serial number of the product is enough. But this is not even the best part. In case of the product malfunctioning and after you contact OCZ and start the RMA procedure, a new SSD is forwarded to the consumer by the company. After the consumer receives the new drive he sends the malfunctioning one back to the company at OCZ's expense. This warranty is valid for all of Europe and so far only covers the OCZ ARC 100 and the AMD R7.

All we can say is a big well done to OCZ for the warranty of the AMD R7 240GB and that we wish other companies would follow the same path.

In the next picture we can see the installation manual of the AMD R7 240GB.

In the last leaflet we find the instructions to download and install Acronis True Image HD, perhaps the most well known drive cloning software, which is bundled with the AMD R7 240GB.

We are done with the papers and we come to the plastic protective case of the AMD R7 240GB, which also contains the screws that will be needed to install the drive, either as it is or with the use of the 3.5" adapter.

Here is the SSD itself, front and back. Its dual nature asserts itself here as well, as it did on the packaging. So, on the front it is labeled as an AMD, Radeon Solid State Drive,while at the back it is described as an OCZ Radeon R7 Sata III Solid State Drive 240GB.

The sticker on the back side carries information such as the model, the product code, the serial number, the drive's certifications, the country where the drive was manufactured, the Firmware version of the drive, as well as its voltage and amperage.

The four screws in the corners are obviously the ones we must remove to gain access to the interior of the drive. One of them is covered by a sticker bearing the warning that the drive's exceptional warranty will be void if it is removed. There goes the sticker, and so does the warranty and with a Philips screwdriver, there go the four screws.

Four more scres hold the pcb of the AMD R7 240GB on the casing and a thermal pad brings the controller in contact with the lid we just removed, which doubles as the controllers cooler. A sticker on the NAND chips bearing a code, gives the impression that it concerns the batch of the NANDs used in the particular drive.

Away with all that and in the picture of on the right we can see the essence of the AMD R7 240GB.

Here we see a close-up of the controller of the AMD R7 240GB which is none other than the known Indilinx Barefoot 3 M00 with the code IDX500M00-BC. It is exactly the same controller we find inside the Vector 150, the flagship of OCZ, which is clocked at 397MHz.

In the picture of the right, we have one of the totally 16 NAND chips on the drive. They are made by Toshiba and have a capacity of 16GB each, bringing up the total capacity of the AMD R7 240GB to 256GB. As the SSD only offers 240 of them to the user, we must assume that the remaining 16GB are used for overprovisioning. The NAND chips used in the AMD R7 240GB are the latest 64Gbit A19nm synchronous MLC NAND by Toshiba and are coded TH58TEG7DDKBA4C. Theoretically they will be a bit slower than the ones used in the Vector and Vertex models due to the smaller lithography, but they will also be cheaper.

On each side of the PCB there is a chip by Micron which provides the cache memory for the controller of the SSD. Specifically they are 2 chips coded D9PSH, 256MB each, which give a total of 512MB of cache to the Indilinx M00. The rest of their specifications are that they are DDR3 chips clocked at 800 MHz (DDR3 data rate = 1600Mhz) and CL=11. For more information you can visit Micron's web page.

In the picture on the right you can see the other side of the PCB which includes 8 NAND chips, 1 cache chip and on the lower left, the power circuit of the drive.

At this point we are done with the anatomical analysis of the AMD R7 240GB. Join us on the next page to see what all these we have seen can offer us.
[break=Specifications - What is this thing capable of?]
In the following table we can see the nominal performance of the AMD R7 at the capacities it comes in. We observe that the Read Speed is not influenced by the capacity while the Write Speed is somewhat lower only in the 120GB model. The Read IOPS are displayed on the table and rise with the capacity while the Write IOPS are at 90.000 and the same for all capacities. The Steady State IOPS are dependent on capacity and measure 12000 IOPS for the 120GB model, 20000 IOPS for the 240GB model and 23000 IOPS for the 480GB model. Pretty impressive numbers, especially the ones for Steady State, which are exceptional and far above the competition.

Here we can see the physical characteristics and parts of the AMD R7 240GB, which we saw in detail in the previous page.

In this table we see the reliability features of the AMD R7 240GB. And first of all the encryption features which are 256-bit AES-compliant and support only ATA passwords while the industry is turning to the safer and easier to use TCG Opal encryption. Be that as it may, desktop users will be fine with it.

Continuing down the table, we see the ECC error correction of the AMD R7 240GB, the support of S.M.A.R.T. and the fact that the AMD R7 240GB is rated for 30GB of writes per day for 4 years.

Here we see the power consumption of the AMD R7 240GB and the high tolerances it has to environmental factors, as all SSDs do.

It is noteworthy that the AMD R7 240GB sadly does not feature low power states (slumber and DevSleep), so its power consumption in idle state is comparatively higher than the competition.

In the next page we will find the analysis of the software that accompanies the AMD R7 240GB.

[break=Software - Spartan but funtional]
The AMD R7 240GB does not include in its package any software but you can go to OCZ's web page and download the Radeon Toolbox.

The software is an exact copy of the known OCZ Toolbox, but with different colors to match the AMD color scheme and has all the basic functions one needs to perform on an SSD. We display and explain its functions below.

Running the software opens the screen above and in the white box it shows you all the OCZ SSDs in your system. In our case we can see the OCZ ARC 100 which is our OS drive and the Radeon R7. Choosing any of them from the list we see at its right the exact model, the capacity, the serial number, the firmware and the WWN (World Wide Name).

We must note that if no drive by OCZ is found in the system so you can choose it on the list, the rest of the tabs in the program will not open. So, having chosen one drive and pressing Tools, you get the following window...

...in which you can download a new firmware, if there is one, and upload it to the drive or send a manual TRIM command.

The third tab is about security...

...where we can secure erase our drive and return it to factory state, for security reasons so that one is not able to recover our data (or so they would have us think) but also to return the SSD to a clean state, where all NAND cells are clean and ready to write data. This procedure, contrary to what one might thing, is completed rather fast.

The last tab is the Details tab...

...where you can get details about your SSD and its S.M.A.R.T. status.

Choosing the security button on the top...

...just gives the option to perform the firmware downloads through a proxy serves and help...

...guess what, yes!, gives information about our choices in the program.

On the whole we can say that OCZ's software, although unimpressive, performs all the basic functions it should, while the only missing thing we would like would be to include a guide to improving the settings of windows to run on an SSD drive.
[break=Our Test system and methodology - Let's see, what are we using and how?]

Our test system is based on the Asrock Z97X Killer. This particular motherboard sports SATA III, SATA Express and Μ.2. interfaces making it able to support all SSDs in the market, currently and in the near future. Thus it provides the means to make direct comparisons on the same system.

The CPU is the Intel G3258 Anniversary Edition, with unlocked multiplier which allowed us to easily overclock it to 4,5 GHz. Keeping in mind that the SSD test with high IOPS require hick clocks but run on only one thread, this particular CPU is perfect for our tests.

The RAM is at 8GB which more than covers our needs for the tests, The PSU is the V650 by Cooler Master and the system is housed by the Cooler Master Test Bench V1.0.

Our warmest thanks toAsrock and Cooler Master for the parts they provided for the construction of our test system!

The operating system on which we run our tests is Windows 7 64bit and it includes, aside of all the drivers, the basic programs that come with the motherboard so we can achieve a more realistic and close to fact test environment. The driver of the SATA controller used was the Intel 13.10.1058, obviously in AHCI mode.

The tests which the AMD R7 240GB will undergo are divided into 4 categories:

1: Synthetic benchmarks on an empty drive, where the tested SSD is a secondary drive to the test system, empty and relaxed. These are the perfect conditions you will never see in real usage.

2: Synthetic benchmarks on a system drive, where the operating system is installed in the tested SSD. Here things come a little closer to reality, but synthetic benchmarks are still synthetic.

3: Steady State Benchmarks where we measure the performance of the SSD while it writes random data for a prolonged period of time so it reaches its worst possible performance. Here things are the opposite of before and thus very unfavorable and unlikely to appear during the usual home usage. But they are the rule in database and file servers.

4: Real Life Tests! Here, using the special suite created for us by the programmer Stelios Sifakis (whom we thank!), we perform some of the most usual tasks and time them. In these tests we place the most gravity for the score as they show the real experience a user will have.

The SSDs included in our comparison are the Samsung 840 EVO 250GB, the Kingston SSDNow KC300 240GB, the OCZ ARC 100 240GB, the fastest until now Samsung 850 Pro 256GB and of course our leading actor, the AMD R7 240GB.

Stay tuned. After the tests there is commentary and score!
[break=Synthetic Benchmarks on an empty drive - We start easy]
The bench is set and the AMD R7 240GB lay down bravely to take the torture.

The HD Tune Pro 5.50 measures the reading and writing IOPS in unpartitioned SSDs. Lets see how the AMD R7 240GB fares against its competitors:

In the Read Tests of the HD Tune Pro 5.50 the AMD R7 240GB takes an easy victory, starting our tests on the right foot...

...while in the Write Tests, where differences are smaller, we would have to say that it tied the Samsung 850 Pro 256GB for first place. Not bad at all!

We continue with the Average Access tests of AIDA 64 4.50. With the AIDA 64 Average Read Access and Average Write Access, executed in a drive without partitions we see the response time of the SSD to various requests. In other words, we measure its latency. Here, lower is better.

In the Average Read Access, all the SSDs tie for first place, except the Kingston SSDNow KC300 240 GB which has a significantly higher latency.

In the Average Write Test, the AMD R7 240GB keeps its average latency at 2ms and gets another victory!

Following is the AS SSD Benchmark, version 1.7.4739.38088.

In the read test the AMD R7 240GB stays on the heels of the Samsung drives, except the 4K Read Test, where it lags behind.

In the Write Tests it fares better and we would say that on the whole it shares first place with the Samsung 850 Pro 256GB.

In the copy tests the AMD R7 240GB comes third, with its best performance in the ISO Test and its worst in the Game Test.

Following are the tests of the CrystalDiskMark 3.03b x64.

In the read tests of the CrystalDiskMark the AMD R7 240GB does not fare well.

On the contrary, in the Write Tests it comes first in every single test!

Moving on to the score from the Anvils Storage Utilities 1.1.0...

...where the AMD R7 240GB gets the third place, behind the drives by Samsung.

We now come to the suites of Futuremark whom we older guys remember as MadOnion. These suites try to simulate real working conditions. Lets start with the PCMark 05.

Here the AMD R7 240GB does not do so well, especially in the Virus Scan Test.

Going on to the more contemporary PCMark7.

In this test the AMD R7 240GB keeps close to the competition, except in the Starting Apps test where if falls significantly behind.

And we conclude this series of tests with the most recent PCMark8, where the results are in seconds and lower is better.

Here, excluding a weakness of the Kingston SSDNow KC300 in the Photoshop tests, the differences are insignificant.

Thus, we conclude the synthetic benchmarks on an empty drive with the impressions the AMD R7 240GB being generally positive. We have to notice that the Performance in some of the tests is exceptional while in others it is unexplainably low. Perhaps a better optimized firmware could solve issues such as the low performance in Virus Scan.
[break=Synthetic Benchmarks on a system drive - Things are getting interesting]
We are making things a little harder now, running the same benchmarks but this time the operating system is installed on the SSD tested.

Once again we start with the HDTune Pro 5.50 but this time obviously on a partitioned drive and we only run the read test as the write test is destructive and would have erased all data on the drive.

With the OS installed in the drive the situation in HD Tune Pro 5.50 changed with the AMD R7 240GB getting fourth place, behind the SSDs by Samsung but also the theoretically lower OCZ model, the ARC 100 240GB.

We continue with the Random Read Access of AIDA64 4.50, on a partitioned drive now. The Random Write Access, being destructive, is omitted.

Here the AMD R7 240GB has the same performance as the Samsung 850 Pro 256GB and lags behind the Samsung 840 EVO 256GB and OCZ ARC 100 240GB.

Going on to the AS SSD Benchmark version 1.7.4739.38088.

In the Read Tests AMD R7 240GB comes third while...

...in the Write Tests it keeps the first place it had in the empty drive tests.

In the Copy Tests, with a first and a second position, we would say that on the whole it comes second only to the Samsung 850 Pro 256GB.

Following are the tests of the CrystalDiskMark 3.03b x64.

The presence of the OS does not seem to effect this test and so we again see the AMD R7 240GB finishing in the third place.

In the Write Tests the AMD R7 240GB rivals the Samsung 850 Pro 256GB everywhere but the 4K QD32 test, where instead of getting a performance boost compared to the 4Κ Test as is expected, we see a performance decrease. The same thing is true for the OCZ ARC 100 240GB and the Kingston KC300 240GB. Seems like a firmware issue.

Continuing with the Anvils Storage Utilities 1.1.0...

...where the results are like to the ones on empty SSDs. The AMD R7 240GB keeps the third place.

Following are the tests of Futuremark. In all the tests of PCMark05, PCMark7 and PCMark8 the results are similar to the ones on the empty drives. If you don't believe us, here they are:

And we finish with BootTimer which measures the time from the loading of the windows kernel until the desktop is usable.

Here the AMD R7 240GB came second, after the lightning fast Samsung 850 Pro 256GB, coming less than half a second behind it and leaving the rest more than half a second behind itself.

This concludes the synthetic benchmarks on a system drive. The AMD R7 240GB was decent, with some obvious weaknesses in certain tests.

Let's hit the tough road!
[break=IOMeter - We press it, can it take it?]
The next test is purely sadistic. We make the SSD write random 4K packages for 12 hours straight without resting and measure its IOPS every five minutes. After some time and as the garbage collection and trim features do not have time to work, the SSD reaches a steady state where every write has to preceded by a deletion of the NAND cell and the SSD operates at its worst performance.

This case is very unlikely to occur in light, casual usage but it is the norm in database and file servers. You should consider the following results according to the heaviness of your workload .

OCZ takes pride in the performance of its SSDs in this particular test, both of its higher models and, as we can see here, the AMD R7 240GB that sweeps the floor with its competition. Any comparison is unnecessary.

Following is the mixed workload test. SSDs in a system rarely do 100% reads or 100%% writes like in the synthetic benchmarks. In the following test we start with three five-minute tests of 100% random reads and keep the average. We continue with three five-minute tests of 95% random reads and 5% random writes and keep the average and so forth as indicated in the graph until we get to 100% random writes. Note that this particular test is run on a drive that has already reached steady state and tries to simulate heavy multitasking.

As expected, the AMD R7 240GB storms this test as well.

Finishing these very hard tests, we have to stress that it is a very rare occasion to find such extreme conditions in normal everyday use. Their value is to show us the other side of the synthetic benchmark coin and let us try to find the truth somewhere in the middle.

But instead of guessing we have a better suggestion. The suite of Real World Tests is coming to clear the field, with true tasks in windows and exact timing.
[break=Real Life Tests - How is it in real life?]
We start with the windows startup time test. One might ask, didn't we already do that with BootTimer? As we said, BootTimer measures the time from the loading of the windows kernel until the desktop is usable. But windows itself records in the event viewer the total time of the windows start up, until all services are loaded. As reason indicates, this time is significantly longer and in particular...

...AMD R7 240GB is packed very close together with the rest of the SSDs in the test, except the Kingston KC300 240GB which is faster by 1.3 seconds.

We continue with real file copying.

We copy 32000 4K files, 500 MP3s, 1000 21MP JPGs, a 7,52GB ISO 7.52GB and a 10GB High Definition MKV movie. The respective times are shown in the diagram above with the AMD R7 240GB getting second place, behind the Samsung 850 Pro 256GB.

Next, we will store a folder containing the setup of the Office 2013 in a rar file (using the "store" setting)...

...where the AMD R7 240GB is the fastest and then we will decompress a - compressed - rar containing the setup folder of Office 2013.

Here the best time is achieved by the Samsung 840 EVO 250GB with the AMD R7 240GB following very closely.

We continue with a virus scan of the system32 folder of windows using the Avira Antivirus, where...

...the AMD R7 240GB displays once more its weakness in Virus Scanning.

Next we will open a 324MP TIF file with Photoshop CS6 save it as BMP and close Photoshop.

In this test the [AMD R7 240GB did not do very well and got the fourth place.

Then we open Word 2013, load a huge DOCX with images links etc, save it as DOC and close Word.

Fourth place once more for the AMD R7 240GB.

Next, we install MS Office 2013...

...where the best time is made by the Kingston SSDNow KC300 240GB with the Samsung 850 Pro 256GB just behind and the AMD R7 240GB getting third place.

Thus end the Real Life Tests. The last chart shows the total time of these tests where...

The AMD R7 240GB had the second best performance, behind the Samsung 850 Pro 256GB. Their difference is 14,703s which translates to a little less than 3,37%. Not at all bad if you keep in mind that the Samsung 850 Pro 256GB is the fastest SSD we have ever tested.
[break=Conclusions and score - To the point]
My dear Watson, the time has come to reach a conclusion. We have in our hands an SSD which AMD puts in the R7 series, which is to say, low to mid range. But it offers anything but low to mid range performance. Coming close to the fastest consumer SSD at a distance of only 3,37% and with the promise it will keep that performance even in the toughest of conditions, it is surely worthy of our attention.

Is it worth investing in the AMD R7 240GB? The answer is, it depends. For users who use their computers lightly and don't put a lot of stress on their SSDs, the choices in the market are plenty, including the faster Samsung 850 Pro 256GB. For users who do heavy multitasking and constant writes that stress the SSD, the AMD R7 240GB is very enticing. For most people who are somewhere in the middle, let each decide where to put the weight in choosing their SSD.

Of course, we have to take into consideration that the AMD R7 240GB comes with the best warranty in terms of conditions we have ever seen, not only on an SSD but on any product!

And so we get to the score:

a) In Features, we take away a few points because the AMD R7 240GB does not have low power states (slumber and DevSleep) and just a tad more for the missing TCG Opal encryption.

In Price the AMD R7 240GB is not available in the Greek market at this moment but the suggested price is 144€, including V.A.T.. The cheapest price for the Samsung 850 Pro 256GB which was the only SSD in our possession which performed better than the AMD R7 240GB is 175€, including V.A.T., in stores with availability and including shipping. That comes to a price difference of 31€, or about 21,53% while the difference in real life tests performance - which are the ones we care about the most - was 3,37%. Even if we include stores with no availability and exclude shipping the cost of the 3,37% faster Samsung 850 Pro 256GB comes to 158,68€, which is 10,19% more expensive. Some will argue that the Samsung 850 Pro comes with a 10 year warranty, which is important, but the 4 years offered by the AMD R7 240GB are not for a regular warranty, but for the new and improved Shieldplus, with great advantages. To make a long story short, we can say that according to what it offers and compared with the competition, the AMD R7 240GB achieves a great value for money. It will be marked accordingly.

c) Usability does not have any real significance in SSDs. They all now have a 7mm height which allows for installation everywhere from a desktop to an ultrabook and, let's face it, their installation has a worst case scenario of screwing four screws. For these reasons, the Usability category will not be taken into account in the SSD Scores.

d) In the Performance department, we have a lot to discuss. It depends how one would like to approach the subject. In the synthetic benchmarks the AMD R7 240GB was mediocre. Considering that synthetic benchmarks are not indicative of performance in real use, they make a very small part of the overall score. On the contrary in Real Life Tests the AMD R7 240GB did very well and came second overall, beating strong competition. Moreover, in our sadistic tests, it scored excellent showing that it is capable of maintaining its high performance over time no matter how hard you treat it. For all of the above, its performance score will be quite high.

Advantages

Exceptional performance in the steady state test.
Exceptional performance in the mixed workload test.
Steady performance over time.
Good performance in the Real Life Tests./li>
It includes a 3.5" adapterli>
It includes the very well known cloning software Acronis True Image HD.
The best warranty we have ever seen.
Spartan but complete software.
Multi-language installation manual.

Disadvantages

Mediocre scores in synthetic benchmarks.
Absence of a windows optimization tab in the Radeon Toolbox.

Based on the above, the total score of the AMD R7 240GB is:

We thank OCZ and AMD for providing the test sample

Polymeros Achaniotis 22/10/2014

Samsung 850 Pro 256GB Review

Mon, 06 Oct 2014 08:16:00 +0000

Some things in life are milestones, defining the standards in their categories. Samsung's SSD line is recognised as the absolute combination of reliability and performance and only their spicy price afforded the competition some breathing room. After the very successful Samsung 830 and the even better Samsung 840 Pro, the announcement of the Samsung 850 Pro justifiably raised great expectations!

SSD technology is rapidly advancing and every once in a while we have a new generation with bigger capacities, higher speeds, lower prices. At the same time, limitations of the SATA 6GB/s protocol have for some time now limited SSD performance and many companies are already turning to other protocols in order to unleash the potential of their SSD technologies. The Samsung 850 Pro insists on using the ageing protocol, which is completely logical, as the majority of users do not yet possess a computer with something better.

The question is, will the Samsung 850 Pro be worthy of the tradition its predecessors created and strech the limits of SATA 6GB/s just a bit further? Make a cup of coffee and after the first sip, follow us to the next page!

[break=From the packaging to the silicone - What you get]
The sample sent to us by Samsung is a review sample and not a retail part and comes in a simple brown box, without accessories. Thus, we will proceed straight to the photos of the Samsung 850 Pro itself.

Samsung continues the tradition of the minimalistic look of their SSDs, to the degree that the only identifying feature on their top side is the color of the destinctive square. In our case, the colour you see signifies the model 850 Pro.

The capacity and other details about the drive, are written on a sticker on the down side. As you can see on it, our sample is the 256GB variety and its model number is MZ-7KE256.

There are also all the classic numbers, codes and certifications, but no user is interested at that, and so we will not bore you with it from the second page of the review. But we will ask you to remember that there is a PSID number, which we will explain soon.

Next, you see the data and power plugs which we photographed for no particular reason, but every reviwes does it and we couldn't do less!

And in the photograph on the right, wonder and frustration expressed in one screw. Well, not in one, in three to be precise. Because there are two more of these accursed screws under the sticker we saw before, forming a triangle. Not the Bermuda triangle, but destructive nonetheless. Why do we say so?

Samsung chose to seal, lock and bolt the cell of the 850 Pro with these screws, called pentalobe torx. Screwdrivers for these are very rare in the Greek market and we were not able to get hold of one in the limited time we had available. So, carefully and with many pains we managed to unscrew two of the three screws, but the third was screwd decisively and with malice. After wearing its head and our screwdriver, we enlisted the help of a power drill and the volt opened, without casualties, except that our sample now locks with two instead of three screws.

And after our little adventure, we finally gained access to the inner of the drive. But, lo! The tiniest of PCBs! What is going on here? WHere are the 256GB?

In the center of the left photograph, we see predominant a Samsung constructed, big (as compared to the PCB) chip coded S4LN045X01-8030. It is the controller of the Samsung 850 Pro which is a triple core ARM Cortex R4 and exactly the same that was used in the Samsung 840 EVO, eventhough Samsung has not divulged if it has kept the 400 MHz timing or they have increased it for the 850 Pro. As usual, it supports TRIM and Garbage Collection. It also supports AES 256-bit hardware-based encryption as well asTCG/Opal V2.0 and Encrypted Drive (IEEE1667). And for all those who, very justifiably, find boring to read about all that, the bottom line is that the controller of the Samsung 850 Pro supports all concurent encryption protocols, so if you are in the mood of hiding your data, this can be done easily, reliably and withoud sacrificing performance, as the whole process is done on a hardware level. The PSID (Physical Security ID) we asked you to remember is required to unlock the drive (but not the data it holds) if you forget the code you used to lock it. Finally, the controller is protected from losing any data due to overheating by a technology called Dynamic Thermal Guard.

But the real power of the Samsung 850 Pro resides in the photograph on the right, where we see one of the NAND chips, constructed by Samsung itself, utilized by the drive. These chips are an innovation of the company, used for the first time in the Samsung 850 Pro and they are a revolutionary technological advance of NAND technology. Specifically, they are the first 3D NAND chips, called 3D V-NAND and they are the reason the PCB is so small, as only four of them can provide the 256GB capacity required.

We will not tire you with specifics of the 3D V-NAND technology, but we will quote its main advantages: These chips have 32 layers, instead of 1 found in the traditional NAND chips, which results in much larger capacity per chip. They are built using a 40nm process, allowing for increased reliability and larger life span than the usual NAND chips which have now reached 20, 19, even 15nm.

On this side of the PCB, we have two 3D V-NAND chips coded K9PRGY8S7M. From this code, with quite a bit of searching, we derive that this chip has 8 dyes. 3D V-NAND chips have a capacity of 86Gbit per dye. So, the two chips on this side have a total capacity of 2 Χ 8 Χ 86Gbit = 1376Gbit.

The last chip on this side of the PCB is the cache memory chip of the Samsung 850 Pro, constructed by - none other than - Samsung, coded K4P4G324EQ-FGC2, which is an LPDDR2 DRAM chip with a capacity of 512ΜΒ.

The other side of the PCB has another couple of 3D V-NAND chips, and change.

The "change" is a chip coded A0M4FA for which we were unable to find any data, but given its position it looks like the power controller of the board.

The two 3D V-NAND chips of this side are coded K9HQGY8S5M and have 4 dyes each. Thus, this side contributes 2 Χ 4 Χ 86GBit = 688GBit to the capacity of the Samsung 850 Pro 265GB.

All in all, on both sides of the PCΒ we have 1376Gbit + 688Gbit = 2064Gbit or 2064 / 8 = 258GB.

258GB? HAve we made a mistake? Isn't everything in computers based on the powers of 2? Shouldn't it have been 256GB? Well, no, it should not have to be. This is the total capacity of the drive, eventhough Samsung leaves the traditional 256GB for the user and keeps the other 2GB as a small overprovisioning, which of course can be increased using the provided software, as in previous models.

We hope we didn't tire you with all the technicalities, but the Samsung 850 Pro 256GB is an SSD packing a lot of technology and some things needed to be explained.

We should also mention that the retail version does not include anything more than the SSD itself and an installation manual. No bracket to install the SSD in a 3.5" cage or even screws. The software provided is only available for download at the company's web page.

Let us now proceed to the next page.
[break=Specifications - What is this thing capable of?]
Some of the specifications of the Samsung 850 Pro 256GB were already analysed in the previous page. But you can see them all together, along with a lot more in the following table.

As you can see, the Samsung 850 Pro 256GB is a consumer SSD with standard dimensions and height which allows for installation everywhere, even in ultrabooks. It has a SATA 6Gb/s interface but is compatible with older protocols, sacrificing a large portion of its performance.

All capacities are available in the Greek market at this moment and specifically the 128GB model (103,55€), the 256GB model (164,77€), the 512GB model (321,14€) and the 1TB model (574,69€).

Sequential read speed is at the limit of the SATA 6Gb/s protocol and is rated at 550MB/s for all capacities while sequential write speed is at 520MB/s for all capacities except 128GB which drops to 470MB/s.

Random Access IOPS τυχαίας are impressive, especially the ones with Queue Depth = 1 and rated at 10.000 IOPS for reading and 36.000 IOPS for writing, for all capacities, while with QD=32 they climb to 100.000 and 90.000 respectively.

What is even more impressive is the great reliability Samsung claims, with an MTBF of two million hours and a total write capacity of 150TB! Besides, the product is covered by a 10-year warranty!

The rest are the classic features found in SSDs, with the great indurance to hardships and small consumption and the support of S.M.A.R.T. etc, etc...

The small table contains an analysis of the 3D V-NAND of the drive we will be testing, as detailed in the previous page, but because we are tidy we give it to you in table form also:

We hope we have not tired you a lot with the technical stuff, which we promise are over for this review. Come now with us to see the software offered by Samsung.

[break=Software - Nothing is left out]
The software provided by Samsung for its SSDs is trully complete. It includes two programs you can download at Samsung's web site.

Let's start with the management software called Samsung Magician.

The first tab is divided into three parts. In the first we can see general information about the drive, like the serial number, firmware, total capacity, health status, total writes, used and free space and drive letters allocated to the drive. We also have the possibility, pressing S.M.A.R.T., to view the relative information, and to register our drive, in which case the green tab indicating that the drive is Genuine appears, if of course we do have a genuine Samsung drive. Not that we have seen or heard of counterfeits, but obviously Samsung has taken all measures to ensure that there are none.

In the second part we see the mode the SATA interface of the drive is in, which should ideally be AHCI. The next box shows that our drive is connected through SATA 6Gb/s (SATA 3) which is desirable for the best performance.We also see the optimization of the OS which can be achieved with the help of Samsung Magician, as we will see.

The third part of the first tab we have the option to run a performance test.

The second tab of the Samsung Magician is dedicated to a more comprehensive Performance Benchmark. In the screenshot you can see the results of our sample.

The third tab also has only ony function, named Performance Optimization. It is actually the ability invoke a manual TRIM command and speed up your SSD in case the automatic TRIM function and Garbage Collection have not done a proper job cleaning up the NAND cells from previous writes.

The fourth tab automatically connects to the company's web page and provides firmware updates, which it downloads and executes if you choose to do so, without requiring special knowledge or effort from the user. The upgrade is done inside windows and after a restart, everything is ready!

The fifth tab contains the OS optimization assistant, for use with SSDs. It gives options for maximum performance, maximum capacity, maximum reliability and advanced options for power users to choose whatever fits you. Excellent help from Samsung to users of all levels!

The sixth tab is about Over Provisioning.Samsung, in contrast to many of its competitors who assign specific over-provisioning, lets the users decide the size of the capacity they want to sacrifice for increased performance and longevity of the drive, while guiding and proposing throught their software. Simply exceptional!

The seventh tab provides the ability to perform a Secure Erase of the drive, to bring it to an out-of-the-factory state. This does two things. First, it (theoretically) ensures that the data we had on the drive will not be recoverable and second that all NAND cells will be clean and ready to write, exactly like new, so the SSD will have its greatest performance. In case the drive contains the OS, Secure Erase cannot be performed from within windows and for that purpose the Samsung Magician offers the option to create a Bootable USB or Bootable CD/DVD to perform the Secure Erase.

The eightth tab gives the option to ebable the RAPID 2.0 technology. This technology uses a part of the RAM, up to 4 GB and in any case not more than 25% of the total ram, to improve the performance of the SSD. The use of this technology offers a huge increase in performance on the synthetic benchmarks, which does not translate to a similar increase in every day use. Maybe when software developers write with SSDs instead of HDDs in mind, we will see some real use for that technology, but for the time being, the difference is only in benchmarks. Note that if there are more than one Samsung SSDs in the system, only one can make use of the RAPID 2.0 technology.

The ninth tab contains the Data Security features of the drive and the option to activate them.

Thus ends the brief description of Samsung Magician.

Let us now proceed to the second piece of software Samsung offers to the users of its SSD, which is the Samsung Data Migration Software. As indicated by the name, this software will help you transfer your OS from any drive to your new Samsung SSD. And it will do so simply, fast and from within Windows. Of course, it will only work if you have a Samsung SSD as the target drive.

In the following screenshots you can see the simple procedure of transfering our OS from the current SSD to the Samsung 850 Pro 256GB.

So simple, so functional!

Trully impressive software, enjoyable, simple, functional.

The time we have all been waiting is nearing, the time of the tests. But before that, let's see how we are going to perform them.
[break=Our test system and methodology - Let's see, what are we using and how?]

Our test system is based on the Asrock Z97X Killer. This particular motherboard sports SATA III, SATA Express and Μ.2. interfaces making it able to support all SSDs in the market, currently and in the near future. Thus it provides the means to make direct comparisons on the same system.

The CPU is the Intel G3258 Anniversary Edition, with unlocked multiplier which allowed us to easily overclock it to 4,5 GHz. Keeping in mind that the SSD test with high IOPS require hick clocks but run on only one thread, this particular CPU is perfect for our tests.

The RAM is at 8GB which more than covers our needs for the tests, The PSU is the V650 by Cooler Master and the system is housed by the Cooler Master Test Bench V1.0.

Our warmest thanks to Asrock and Cooler Master for the parts they provided for the construction of our test system!

The operating system on which we run our tests is Windows 7 64bit and it includes, aside of all the drivers, the basic programs that come with the motherboard so we can achieve a more realistic and close to fact test environment. The driver of the SATA controller used was the Intel 13.10.1058, obviously in AHCI mode.

The tests which the Samsung 850 Pro 256GB will undergo are divided into 4 categories:

1: Synthetic benchmarks on an empty drive, where the tested SSD is a secondary drive to the test system, empty and relaxed. These are the perfect conditions you will never see in real usage.

2: Synthetic benchmarks on a system drive, where the operating system is installed in the tested SSD. Here things come a little closer to reality, but synthetic benchmarks are still synthetic.

3: Steady State Benchmarks where we measure the performance of the SSD while it writes random data for a prolonged period of time so it reaches its worst possible performance. Here things are the opposite of before and thus very unfavorable and unlikely to appear during the usual home usage. But they are the rule in database and file servers.

4: Real Life Tests! Here, using the special suite created for us by the programmer Stelios Sifakis (whom we thank!), we perform some of the most usual tasks and time them. In these tests we place the most gravity for the score as they show the real experience a user will have.

The SSDs included in our comparison are the Samsung 840 EVO 250GB, the Kingston SSDNow KC300 240GB, the OCZ ARC 100 240GB and of course our leading actor, the Samsung 850 Pro 256GB

Stay tuned. After the tests there is commentary and score!
[break=Synthetic Benchmarks in an empty drive - We start easy]

The HD Tune Pro 5.50 measures the reading and writing IOPS in unpartitioned SSDs. Lets see how the Samsung 850 Pro 256GB fares against its competitors:

In the reading test of the HD Tune Pro 5.50 the Samsung 850 Pro 256GB is on the heels of the OCZ ARC 100 240GB which is the fastest drive in this test...

...while in the write tests it only loses the 4K test by a small margin but wins all the rest. On the whole, we would have to say that in this test the Samsung 850 Pro 256GB comes second to the OCZ ARC 100 240GB.

We continue with the Average Access tests of AIDA 64 4.50. With the AIDA 64 Average Read Access and Average Write Access, executed in a drive without partitions we see the response time of the SSD to various requests. In other words, we measure its latency. Here, lower is better.

In the Average Read Access test, all the tested SSDs come to a tie, except the Kingston SSDNow KC300 240 GB which has a significantly higher latency.

In the Average Write Test, the OCZ ARC 100 takes the lead with only 0,03ms average latency while the Samsung 850 Pro 256GB follows.

Following is the AS SSD Benchmark, version 1.7.4739.38088.

In the read test the Samsung 850 Pro 256GB is fastest and is only second to the Samsung 840 EVO in the 4K reads.

In the write tests it is generally the fastest, without standing out but keeping the top place in all the tests.

In the copy tests the Samsung 840 EVO 250 GB is the only one that competes and sometimes exceeds the speed of the Samsung 850 Pro 256GB, but on the whole, the reviewed drive is again the winner.

Following are the tests of the CrystalDiskMark 3.03b x64.

In the read tests of the CrystalDiskMark the Samsung 850 Pro 256GB shares victories with the Samsung 840 EVO 250GB.

On the other hand, in the write tests its main competitor is the OCZ ARC 100 240GB, which manages to marginally win the 4K write test. On the whole, another win for our contestant!

Moving on to the score from the Anvils Storage Utilities 1.1.0...

...where the Samsung 850 Pro 256GB makes an easy victory.

We now come to the suites of Futuremark whom we older guys remember as MadOnion. These suites try to simulate real working conditions. Lets start with the PCMark 05.

Here the Kingston SSDNow KC300 240GB takes the lead with the Samsung 850 Pro 256GB in second place.

Going on to the more contemporary PCMark7.

In this test the Samsung 850 Pro 256GB makes a clean victory as it wins, even by a small margin, all the tests.

And we conclude this series of tests with the most recent PCMark8, where the results are in seconds and lower is better.

Here we also see a slight prevalence of the Samsung 850 Pro 256GB but generally, excluding a weekness of the Kingston SSDNow KC300 in the Photoshop tests, the differences are insignificant.

Thus, we conclude the synthetic benchmarks on an empty drive with the impressions the Samsung 850 Pro 256GB left us being quite positive.
[break=Synthetic Benchmarks on a system drive - Things are getting interesting]
We are making things a little harder now, running the same benchmarks but this time the operating system is installed on the SSD tested.

Once again we start with the HDTune Pro 5.50 but this time obviously on a partitioned drive and we only run the read test as the write test is destructive and would have erased all data on the drive.

With the OS installed in the drive the situation in HD Tune Pro 5.50 changed with the Samsung 840 EVO being the fastest and the Samsung 850 Pro 256GB coming second in most of the tests.

We continue with the Random Read Access of AIDA64 4.50, on a partitioned drive now. The Random Write Access, being destructive, is omitted.

Here the OCZ ARC 100 240GB ties the Samsung 840 EVO 250GB for first place while the Samsung 850 Pro 256GB follows.

Going on to the AS SSD Benchmark version 1.7.4739.38088.

In the read tests the field is similar to the empty drive tests as in...

...the writes test...

...and the copy tests. The Samsung 850 Pro 256GB is the overall winner.

Following are the tests of the CrystalDiskMark 3.03b x64.

The presence of the OS does not seem to effect this test and so we again see the Samsung 850 Pro 256GB conquering the read tests...

...as well as the write tests.

Continuing with the Anvils Storage Utilities 1.1.0...

...where the results are like to the ones on empty SSDs. The Samsung 850 Pro 256GB is again victorious.

Following are the tests of Futuremark. In all the tests of PCMark05, PCMark7 and PCMark8 the results are similar to the ones on the empty drives. If you don't believe us, here they are:

And we finish with BootTimer which measures the time from the loading of the windows kernel until the desktop is usable.

Here the Samsung 850 Pro 256GB is setting new standards, lowering the best time by almost a whole second.

This concludes the synthetic benchmarks on a system drive. The Samsung 850 Pro 256GB is the winner in most and definitely the overall winner.

Let's hit the tough road!
[break=IOMeter - We press it, can it take it?]
The next test is purely sadistic. We make the poor SSD write random 4K packages for 12 hours straight without resting and measure its IOPS every five minutes. After some time and as the garbage collection and trim features do not have time to work, the SSD reaches a steady state where every write has to preceded by a deletion of the NAND cell and the SSD operates at its worst performance.

This case is very unlikely to occur in light, casual usage but it is the norm in database and file servers. You should consider the following results according to the heaviness of your workload .

OCZ takes pride in the performance of its SSDs in this particular test, both of its higher models and, as we can see here, the ARC 100 that sweeps the floor with its competition. Any comparison is unnecessary.

Nevertheless, we have to note that the Samsung 850 Pro 256GB appears greatly improved compared to other non-OCZ SSDs ans keeps, even in this exhausting test, a presentable performance.

Following is the mixed workload test. SSDs in a system rarely do 100% reads or 100%% writes like in the synthetic benchmarks. In the following test we start with three five-minute tests of 100% random reads and keep the average. We continue with three five-minute tests of 95% random reads and 5% random writes and keep the average and so forth as indicated in the graph until we get to 100% random writes. Note that this particular test is run on a drive that has already reached steady state and tries to simulate heavy multitasking.

As expected, the OCZ ARC 100 leads this test as well but the Samsung 850 Pro 256GB keeps a good performance for a longer time and wins the first two tests. In any case, it seems to be a lot more durable than the Samsung 840 EVO 250GB which takes the last place in this test.

Finishing these very hard tests, where the OCZ ARC 100 shows its teeth, we have to stress that it is a very rare occasion to find such extreme conditions in normal everyday use. Their value is to show us the other side of the synthetic benchmark coin and let us try to find the truth somewhere in the middle.

But instead of guessing we have a better suggestion. The suite of Real World Tests is coming to clear the field, with true tasks in windows and exact timing.
[break=Real Life Tests - How is it in real life?]
We start with the windows startup time test. One might ask, didn't we already do that with BootTimer? As we said, BootTimer measures the time from the loading of the windows kernel until the desktop is usable. But windows itself records in the event viewer the total time of the windows start up, until all services are loaded. As reason indicates, this time is significantly longer and in particular...

...The Kingston SSDNow KC300 240GB takes the first place, while the Samsung 850 Pro 256GB comes second.

We continue with real file copying.

We copy 32000 4K files, 500 MP3s, 1000 21MP JPGs, a 7,52GB ISO 7.52GB and a 10GB High Definition MKV movie. The respective times are shown in the diagram above with the Samsung 850 Pro 256GB taking by storm all the file copying tests.

Next, we will store a folder containing the setup of the Office 2013 in a rar file (using the "store" setting)...

... and then we will decompress a - compressed - rar containing the setup folder of Office 2013.

In these tests the best times are made by the Samsung 840 EVO 250GB with the Samsung 850 Pro 256GB following really close.

We continue with a virus scan of the system32 folder of windows using the Avira Antivirus, where...

...the Kingston SSDNow KC300 240GB wins by a fourth of a second the Samsung 850 Pro 256GB.

Next we will open a 324MP TIF file with Photoshop CS6 save it as BMP and close Photoshop.

Here the Samsung 850 Pro 256GB made the best total time.

Then we open Word 2013, load a huge DOCX with images links etc, save it as DOC and close Word.

The Samsung 850 Pro 256GB is first once more, wven by a little bit.

Next, we install MS Office 2013...

...where the best time is made by the Kingston SSDNow KC300 240GB with the Samsung 850 Pro 256GB just behind and the rest watching the action from quite far.

Thus end the Real Life Tests. The last chart shows the total time of these tests where...

The Samsung 850 Pro 256GB has the best performance by far!
[break=Conclusions and score - To the point]
Once again, it is judgement hour, but we would not say it is hard this time. In fact things are quite clear and follow in the tradition set by the predecessors of the Samsung 850 Pro

Lets see how we calculated the score:

The Samsung 850 Pro 256GB has the best overall performance in the synthetic benchmarks, either on an empty drive or an OS drive. It maintains presentable results in the outrageously difficult IOMeter tests and wins by far where it really matters, in the everyday, true applications! In the performance section, we will give it 5 out of 5!

Regarding the specifications, the drive is trully complete, lacking nothing, as you can see in the specifications page (3). 5 out of 5!

As for the price, all these don't come cheap, with the Samsung 850 Pro 256GB costing 164,77€ at this moment, in the Greek market. Justifiably expensive, as it is the fastest SSD we have ever tested and with the most complete specifications, but we would rather it was a bit closer to the prices of the competition. 4 out of 5.

Usability does not have any real significance in SSDs. They all now have a 7mm height which allows for installation everywhere from a desktop to an ultrabook and, let's face it, their installation has a worst case scenario of screwing four screws. For these reasons, the Usability category will not be taken into account in the SSD Scores.

Lets summarize the advantages and disadvantages of the Samsung 850 Pro 256GB.

Advantages

+ Exceptional performance
+ Full set of specifications
+ Supports AES-256 encryption and is compatible with TCG/Opal 2.0 and IEEE-1667 as well as the Windows 8 eDrive
+ 10 years or 150ΤΒ of total writes warranty
+ Excellent Software including among others Data Migration and RAPID 2.0 funtion

Μειονεκτήματα

- High price
- Mediocre performance in the IOMeter tests

Based on the above, the total score of the Samsung 850 Pro 256GB is:

Out thanks to Samsung for providing the test sample

Πολύμερος Αχανιώτης 6/10/2014

OCZ ARC 100 240GB SSD Review - The long distance runner

Mon, 01 Sep 2014 16:39:00 +0000

The story of OCZ with SSDs is a story with all the elements of success. Ambition, conquest, risk, fall, resurrection. If it also had sex, we would be talking about Hollywood's new blockbuster. Anyway, the story is still interesting and the renewed OCZ, now flying on the wings of the powerful Toshiba, does not seem to want to change its tune.

Thus, OCZ keeps coming out with products sporting cutting edge performance in some areas, like the number of IOPS at steady state, not caring if the competition outstrips them in other areas. The enterprise oriented Vector 150 and the enthusiast Vertex 450 and Vertex 460 that OCZ has produced under the roof of Toshiba are all examples that confirm it.

OCZ now completes their SSD line with an entry level - mainstream SSD, the OCZ ARC 100. With genes from its older siblings and some concessions, the ARC 100 will try to bridge the gap between performance and price. Whether it will succeed will become apparent in the following pages.

From the packaging to the silicon - What you get

As you can see in the photos, the OCZ ARC 100 comes in a classic paper package in blue and white which will be a great hit in Paris this year. On the lower right of the front of the package there is a sticker indicating the capacity of the SSD, in our case 240GB. Incidentally, the OCZ ARC 100 comes in 120, 240 and 480 GB varieties.

On the back, in the light blue box we can see some of the features of the tested SSD, which we will go into later, but for now lets keep that the product is SATA III, has a 7mm height to fit into ultrabooks, supports TRIM and is covered by a three year warranty. On the right we see in 12 languages the attempt of OCZ to convince us that SSDs are better than conventional Hard Disk Drives.

Inside the paper packaging we find the SSD in transparent plastic which in combination with the packaging seems able to protect it from normal mishaps during transport. Included is an installation manual with many pages detailing in many languages the installation procedure of the SSD and the warranty terms for which -stay with us - we have some very interesting things to tell you. Other accessories such as a 3,5" adapter or some software, shine by absence. That is not necessarily bad, as long as it is mirrored in the price.

At last, the product itself, whose upper surface is covered in a full body sticker in the colors of the packaging, while the lower surface sports a white sticker with information such as the model, product number, serial number, certifications, country of origin (Taiwan), Firmware with which the SSD comes from the factory and voltage as well as current of operation.

Did you notice anything strange? Yes, the classic sticker that covers some screw and voids the warranty if removed is missing. As a matter of fact, even if it was there, which screw would it cover? Can you even get inside this SSD?

There are the screws! One on each corner of the upper surface, under the big sticker. OCZ decided to avoid defacing the ARC 100 with special warranty stickers and visible screws and hid them under the big sticker of the upper surface. In any case the Warranty terms warn not to mess with the sticker to get access to the screws, as that will void your warranty.

So be it, for you we lifted the sticker carefully and unscrewed the four screws that open the casing of the ARC 100. Then we unscrewed another four that fix the pcb to the lower metal cover and...

...here it is, almost naked covering only the essentials on the left while on the right we uncovered even that. We are talking about the thermal pad that covers the controller of the ARC 100 and brings it into contact with the metal casing to disperse the heat it produces. There is also a sticker with a code number on some two of the NANDs which gives the impression that it concerns the batch of the NANDs used on the drive. Gone!

Here we see a close-up of the controller of the ARC 100 which is none other than the known Indilinx Barefoot 3 M10 with the code IDX500M10-BC, clocked at 352MHz. It is exactly the same controller we find inside the Vector 150, the flagship of OCZ, but there it is clocked higher, at 397MHz.

In the picture of the right, we have one of the totally 16 NAND chips on the ARC 100. They are made by Toshiba and have a capacity of 16GB each, bringing up the total capacity of the ARC 100 to 256GB. As the SSD only offers 240 of them to the user, we must assume that the remaining 16GB are used for overprovisioning. The NAND chips used in the ARC 100 are the latest 64Gbit A19nm synchronous MLC NAND by Toshiba and are coded TH58TEG7DDKBA4C. Theoretically they will be a bit slower than the ones used in the Vector and Vertex models due to the smaller lithography, but they will also be cheaper.

On each side of the PCB there is a chip by Micron which provides the cache memory for the controller of the SSD. Specifically they are 2 chips coded D9PSH, 256MB each, which give a total of 512MB of cache to the Indilinx M10. The rest of their specifications are that they are DDR3 chips clocked at 800 MHz (DDR3 data rate = 1600Mhz) and CL=11. For more information you can visit Micron's web page.

In the picture on the right you can see the other side of the PCB which includes eight NAND chips, 1 cache chip and on the lower left, the power circuit of the drive.

Here is the power circuit with its coils and its capacitors and in the middle the chip coded TPS652510 by Texas Instruments which includes three 3 DC-DC converters and the requiredpower MOSFETs, while it can handle input of 4,5 up to 16V. For more information about it, you can visit Texas Instruments' web page.

In the last picture we see a piece of paper which is included in the package to warn the buyers to not return the product to their retailer but to contact OCZ directly in case of a problem. This is done in 11 languages, 5 on this side and 6 on the other. But why?

At this point we will go back to the warranty, as we promised. OCZ, starting with the ARC 100, is launching a new kind of warranty in the SSD market which promises to make users' lives easier. According to this new warranty which OCZ calls ShieldPlus, you no longer need any proof of purchase for the RMA procedure as the serial number of the product is enough. But this is not even the best part. In case of the product malfunctioning and after you contact OCZ and start the RMA procedure, a new SSD is forwarded to the consumer by the company. After the consumer receives the new drive he sends the malfunctioning one back to the company at OCZ's expense. This warranty is valid for all of Europe and so far only covers the ARC 100.

All we can say is a big well done to OCZ for the warranty of the ARC 100 and that we wish other companies will follow the same path. We cannot but think that such a warranty must come from the great confidence OCZ has about the quality of its product.

Specifications - What is this thing capable of?

In this table we can see the theoretical performance of the ARC 100 at its various capacities. We notice small differences in reading speed and somewhat larger in writing speed. Max IOPS are common to all capacities while max IOPS at steady state are dependent on capacity and are 12000 IOPS for the 120GB model, 18000 IOPS for the 240GB model and 20000 IOPS for the 480GB model. We cannot say we are impressed by the numbers, except for the steady state ones which are exceptional and far superior than the competition's.

Here we can see the physical characteristics and parts of the ARC 100, which we saw in detail in the previous page. Its exact dimensions are also provided.

In this table we see the reliability features of the ARC 100. And first of all the encryption features which are 256-bit AES-compliant and support only ATA passwords while the industry is turning to the safer and easier to use TCG Opal encryption. Be that as it may, desktop users will be fine with it.

Continuing down the table, we see the ECC error correction of the ARC 100, the support of S.M.A.R.T. and the fact that the ARC 100 is made for 20GB of writes per day for 3 years.

Here we see the power consumption of the ARC 100 and the high tolerances it has to environmental factors, as all SSDs do.

It is noteworthy that the ARC 100 sadly does not feature low power states (slumber and DevSleep), so its power consumption in idle state is comparatively higher than the competition.

In the last table we see the comparative features of the ARC 100 and its main competitors. While it does not win any prizes in sequential reads and writes, it features the highest steady state IOPS and a very high MTBF.
Software - Spartan but functional
The OCZ ARC 100 does not include in its package any software but you can go to and download the OCZ Toolbox.
The software has all the basic functions one needs to perform on an SSD. We display and explain its functions below.

Running the software opens the screen above and in the white box it shows you all the OCZ SSDs in your system. In our case, the OCZ ARC 100. Choosing any of them from the list we see at its right the exact model, the capacity, the serial number, the firmware and the WWN (World Wide Name).

We must note that if no drive by OCZ is found in the system so you can choose it on the list, the rest of the tabs in the program will not open. So, having chosen one drive and pressing Tools, you get the following window...

...in which you can download a new firmware, if there is one, and upload it to the drive or send a manual TRIM command.

The third tab is about security...

...where you can get details about your SSD and its S.M.A.R.T. status.

Choosing the security button on the top...

...just gives the option to perform the firmware downloads through a proxy serves and help...

Our test system and methodology - Let's see, what are we using and how?

Our test system is based on the Asrock Z97X Killer. This particular motherboard sports SATA III, SATA Express and Μ.2. interfaces making it able to support all SSDs in the market, currently and in the near future. Thus it provides the means to make direct comparisons on the same system.

The CPU is the Intel G3258 Anniversary Edition, with unlocked multiplier which allowed us to easily overclock it to 4,5 GHz. Keeping in mind that the SSD test with high IOPS require hick clocks but run on only one thread, this particular CPU is perfect for our tests.

The RAM is at 8GB which more than covers our needs for the tests, The PSU is the V650 by Cooler Master and the system is housed by the Cooler Master Test Bench V1.0.

Our warmest thanks to Asrock and Cooler Master for the parts they provided for the construction of our test system!

The operating system on which we run our tests is Windows 7 64bit and it includes, aside of all the drivers, the basic programs that come with the motherboard so we can achieve a more realistic and close to fact test environment. The driver of the SATA controller used was the Intel 13.10.1058, obviously in AHCI mode.

The tests which the OCZ ARC 100 will undergo are divided into 4 categories:

1: Synthetic benchmarks on an empty drive, where the tested SSD is a secondary drive to the test system, empty and relaxed. These are the perfect conditions you will never see in real usage.

2: Synthetic benchmarks on a system drive, where the operating system is installed in the tested SSD. Here things come a little closer to reality, but synthetic benchmarks are still synthetic.

3: Steady State Benchmarks where we measure the performance of the SSD while it writes random data for a prolonged period of time so it reaches its worst possible performance. Here things are the opposite of before and thus very unfavorable and unlikely to appear during the usual home usage. But they are the rule in database and file servers.

4: Real Life Tests! Here, using the special suite created for us by the programmer Stelios Sifakis (whom we thank!), we perform some of the most usual tasks and time them. In these tests we place the most gravity for the score as they show the real experience a user will have.

The SSDs included in our comparison are the very well known in this category Samsung 840 EVO 250GB, the Kingston SSDNow KC300 240GB and of course our leading actor the OCZ ARC 100 240GB.

Stay tuned. After the tests there is commentary and score!

Synthetic Benchmarks in an empty drive - We start easy

The HD Tune Pro 5.50 measures the reading and writing speed in unpartitioned SSDs. Lets see how the OCZ ARC 100 fares beside two of its competitors:

In the reading test of the HD Tune Pro 5.50 the OCZ ARC 100 is by far the fastest, especially in the 512B 4ΚΒ reads, and starts the testing on the right foot...

...while in the write test, despite keeping its high performance, the competition rises and things seem more balanced, wit the OCZ ARC 100 keeping the lead in the 4K writes.

We continue with the Average Access tests of AIDA 64 4.50. With the AIDA 64 Average Read Access and Average Write Access, executed in a drive without partitions we see the responce time of the SSD to various requests. In other words, we measure its latency. Here, lower is better.

In the Average Read Access test, the OCZ ARC 100 ties with the Samsung 840 EVO while the Kingston SSDNow KC300 shows a significantly higher latency.

In the Average Write Test, the OCZ ARC 100 sweeps the floor with only 0,03ms of average latency. The 0,11ms of max latency we see are due to only 1-2 spikes during the test.

Following is the AS SSD Benchmark, version 1.7.4739.38088.

In the read test the OCZ ARC 100 is somewhere in the middle surpassing the Samsung 840 EVO only in the 4K64Thrd reads.

In the write tests it fares even better, winning in the 4K writes also and lagging behind only in the not so important sequential writes.

In the copy tests the Samsung 840 EVO is the undisputable winner with the OCZ ARC 100 marginally beating the Kingston SSDNow KC300.

Following are the tests of the CrystalDiskMark 3.03b x64.

In the read tests of CrystalDiskMark the OCZ ARC 100 does not do well and lags behind its two competitors.

The write tests are a different story where it sweeps the competition and only loses at the sequential writes which are the least important.

Moving on to the score from the Anvils Storage Utilities 1.1.0...

...where the OCZ ARK 100 finishes in the middle. with the Samsung 840 EVO in the lead and the Kingston SSDNow KC300 bringing up the rear.

WE now come to the suites of Futuremark whom we older guys remember as MadOnion. These suites try to simulate real working conditions. Lets start with the PCMark 05.

Here the OCZ ARC 100 does not fare well at all, finishing almost always last, while the Kingston SSDNow KC300 was the king of the test.

Going on to the more contemporary PCMark7.

In this test, excluding the "Starting Apps" category where the Kingston SSDNow KC300 shines and the OCZ ARC 100 does not fare well, all 3 SSD were very close.

And we conclude this series of tests with the most recent PCMark8, where the results are in seconds and lower is better.

Here we see a slight predominance of the Samsung 840 EVO but generally and excluding the weakness of the Kingston SSDNow KC300 in the Photoshop tests, the differences are marginal.

Thus, we conclude the synthetic benchmarks on an empty drive with the impressions the ARC 100 left us being on the positive side. Despite the fact the the synthetic benchmarks are not its strong point, the ARC 100 leaded the scores in many tests and where it lagged behind it was not by much.

Synthetic Benchmarks on a system drive - Things are getting interesting

We are making things a little harder now, running the same benchmarks but this time the operating system is installed on the SSD tested.

Once again we start with the HDTune Pro 5.50 but this time obviously on a partitioned drive and we only run the read test as the write test is destructive and would have erased all data on the drive.

With the operating system on the drive the situation in HD Tune Pro 5.50 changes drastically with the Samsung 840 EVO dominating and the OCZ ARC 100 lagging behind while the Kingston SSDNow KC300 shows a very low performance.

We continue with the Random Read Access of AIDA64 4.50, on a partitioned drive now. The Random Write Access, being destructive, is omitted.

Here the OCZ ARC 100 ties with the Samsung 840 EVO while its higher max latency is due to only one spike during tested and considered unimportant.

Going on to the AS SSD Benchmark version 1.7.4739.38088.

In the read tests the field is similar to the empty drive tests with only difference the rise of the Samsung 840 EVO in the 4K64Thrd test. Thus, the OCZ ARC 100 lags behind the Samsung 840 EVO in this test.

In the write tests we also get similar results to the empty drive tests with the OCZ ARC 100 lagging behind in the sequential writes but storming the 4K64Thrd writes test.

Finally, in the copy tests the Samsung 840 EVO leads the ISO copy test but the OCZ ARC 100 wins the other two, taking the victory of this test at the points.

Following are the tests of the CrystalDiskMark 3.03b x64.

The reading tests remain similar to the ones on the empty drives with only difference the significant fall of the OCZ ARC 100 and Kingston SSDNow KC300 performance in the 4KQD32 test. The Samsung 840 EVO in the indisputable winner of the test.

In the write tests the OCZ ARC 100 keeps the lead in the 512Κ test, ties the Samsung 840 EVO in the 4Κ and loses by a great margin in the 4KQD32 test. The Kingston SSDNow KC300 comes last in all the tests.

Continuing with the Anvils Storage Utilities 1.1.0...

...where the results were similar to the ones of the empty drives. The OCZ ARC 100 lags behind the Samsung 840 EVO and leads the Kingston SSDNow KC300.

Following are the tests of Futuremark. In all the tests of PCMark05, PCMark7 and PCMark8 the results are similar to the ones on the empty drives. If you don't believe us, here they are:

And we finish with BootTimer which measures the time from the loading of the windows kernel until the desktop is usable.

Here best is the SSD which achieves the lowest time with the OCZ ARC 100 winning marginally over the Kingston SSDNow KC300 and by quite a bit more the Samsung 840 EVO.

This concludes the synthetic benchmarks on a system drive. The OCZ ARC 100 lost some ground in some of the tests but remained generally high. As a total impression we would place it a little below the Samsung 840 EVO and quite a bit above the Kingston SSDNow KC300.

Let's hit the tough road!

IOMeter - We press it, can it take it?

The next test is purely sadistic. We make the poor SSD write random 4K packages for 12 hours straight without resting and measure its IOPS every five minutes. After some time and as the garbage collection and trim features do not have time to work, the SSD reaches a steady state where every write has to preceded by a deletion of the NAND cell and the SSD operates at its worst performance.

This case is very unlikely to occur in light, casual usage but it is the norm in database and file servers. You should consider the following results according to the heaviness of your workload .

OCZ takes pride in the performance of its SSDs in this particular test, both of its higher models and, as we can see here, the ARC 100 that sweeps the floor with its competition. Any comparison is unnecessary.

Following is the mixed workload test. SSDs in a system rarely do 100% reads or 100%% writes like in the synthetic benchmarks. In the following test we start with three five-minute tests of 100% random reads and keep the average. We continue with three five-minute tests of 95% random reads and 5% random writes and keep the average and so forth as indicated in the graph until we get to 100% random writes. Note that this particular test is run on a drive that has already reached steady state and tries to simulate heavy multitasking.

As expected, the OCZ ARC 100 storms this test as well. Once we get past the 100% random read test where the Samsung 840 EVO has the best performance and include even a few writes, the Samsung 840 EVO plummets while the ARC 100 flies. The Kingston SSDNow KC300 has a very respectable performance.

Finishing these very hard tests, where the OCZ ARC 100 shows its teeth, we have to stress that it is a very rare occasion to find such extreme conditions in normal everyday use. Their value is to show us the other side of the synthetic benchmark coin and let us try to find the truth somewhere in the middle.

But instead of guessing we have a better suggestion. The suite of Real World Tests is coming to clear the field, with true tasks in windows and exact timing.

Real Life Tests - How is it in real life?

...the Kingston SSDNow KC300 takes the lead with more than a second's difference, while the OZC ARC 100 is just a tenth of a second faster than the last Samsung 840 EVO.

We continue with real file copying.

We copy 32000 4K files, 500 MP3s, 1000 21MP JPGs, a 7,52GB ISO 7.52GB and a 10GB High Definition MKV movie. The respective times are shown in the diagram above with the OCZ ARC 100 winning by far the copy tests of the two large files.

Next, we will store a folder containing the setup of the Office 2013 in a rar file (using the "store" setting)...

... and then we will decompress a - compressed - rar containing the setup folder of Office 2013.

Here the best times are made by the Samsung 840 EVO with the OCZ ARC 100 following close and the Kingston SSDNow KC300 slightly behind.

We continue with a virus scan of the system32 folder of windows using the Avira Antivirus, where...

...the Kingston SSDNow KC300 wins by a third of a second the Samsung 840 EVO while the OCZ ARC 100 lags behind be quite some time. Obviously virus scanning with Avira is not its strong point.

Next we will open a 324MP TIF file with Photoshop CS6 save it as BMP and close Photoshop.

All SSDs fare the same here with the OCZ ARC 100 taking a little more time to open Photoshop.

Then we open Word 2013, load a huge DOCX with images links etc, save it as DOC and close Word.

Here the OCZ ARC 100, despite the slight delay in reads, makes faster writes than the other SSDs achieving the best overall time, even if it is by a bit.

Next, we install MS Office 2013...

...were the best time by far is achieved by the Kingston SSDNow KC300 with the OCZ ARC 100 marginally losing to the Samsung 840 EVO.

Thus end the Real Life Tests. The last chart shows the total time of these tests where...

The OCZ ARC 100 has the better performance , owing largely to the great times it achieved in copying large files.

Conclusions and score - To the point

The hour of judgement is here. The OCZ ARC 100 is an SSD with personality. Where the competition shows its performance in outrageously favorable conditions, the OCZ ARC 100 shows what it can do in outrageously unfavorable ones. Should that mean something to us? Of course it should!

Out of the box, all SSDs are fast! Many are faster than the OCZ ARC 100. But with usage and the heavier the more so, performance drops and sometimes it does so way too much. There the OCZ ARC 100 differs, sporting the best performance in its category.

Is it worth it to invest your money on the OCZ ARC 100? The answer is, it depends. For light users who never stress their drives, another SSD with higher performance under favorable conditions and a lower price might be better. For users with heavy multitasking and frequent writes that stress the SSD, the OCZ ARC 100 is the only way to go in the mainstream category. For most users who are somewhere in the middle, let each decide where to put the weight in choosing their SSD.

Of course, we have to take into consideration that the OCZ ARC 100 comes with the best warranty in terms of conditions we have ever seen, not only on an SSD but on any product!

And so we get to the score:

a) In Features, we take away a few points because the OCZ ARC 100 does not have low power states (slumber and DevSleep) and just a tad more for the missing TCG Opal encryption.

In Price the official price in Germany is 110,50 euro, including 19% V.A.T. which is quite competitive for what the drive offers

c) Usability does not have any real significance in SSDs. They all now have a 7mm height which allows for installation everywhere from a desktop to an ultrabook and, let's face it, their installation has a worst case scenario of screwing four screws. For these reasons, the Usability category will not be taken into account in the SSD Scores.

d) In the Performance department, we have a lot to discuss. It depends how one would like to approach the subject. In the synthetic benchmarks the OCZ ARC 100 did not shine but we would say it faired about average. Considering that synthetic benchmarks are a long throw from real usage, they make a very small part of the overall score. On the contrary in Real Life Tests the OCZ ARC 100 finished first, outstripping 2 strong competitors. Moreover, in our sadistic tests, it scored excellent showing that it is capable of maintaining its high performance over time no matter how hard you treat it. For all of the above, its performance score will be quite high.

Advantages

Exceptional performance in the steady state test.
Exceptional performance in the mixed workload test.
Steady performance over time.
Good performance in the Real Life Tests.
The best warranty we have ever seen.
Spartan but complete software.
Multi-language installation manual.

Disadvantages

Mediocre scores in synthetic benchmarks.
Absence of a 3,5" adapter.
Absence of migration software.
Absence of a windows optimization tab in the OCZ Toolbox.

Based on the above, the total score of the OCZ ARC 100 is:

We thank OCZ warmly for providing the test sample

For TheLab.gr Polymeros Achaniotis 1/9/2014

[English Version] X2 Saturn 5.1 gaming headset: Where did that come from?

Thu, 13 Mar 2014 19:57:00 +0000

Recently, we reviewed X2's Solarblast headset, which promised to deliver a virtual 7.1 audio system experience. This time we test another product of X2, under the commercial name Saturn 5.1, composing of a real 5.1 audio system, with dedicated speakers for every auditory direction and an embedded subwoofer

Many believe that headsets of this category fail to deliver their promises, because humans are made to listen in stereo. Therefore, these believers support the idea that stereo headphones are more than adequate to reproduce any required sound from all directions, when used in the right way. Will Saturn 5.1 stand out as an exception or not?

[BREAK=Packaging & Contents]

The headset comes in a cardboard box with plastic outer covering. Scattered all over the box we will find mentions of the key feature of Saturn 5.1, which is none other than the real 5.1 sound system.

The manufacturer claims that it offers the mostly anticipated solution to the problem of acoustic orientation , using dedicated speakers to simulate sound from various directions . This technique has been known for several years in home cinemas, in which 4+ speakers are placed around the viewer at a certain distance. By controlling the playback source for each stream of a multichannel multimedia file, we offer a vivid playback experience to the viewer.

On the back side, the product's characteristics can be clearly seen, such as sensitivity, impedance and cable length. We can also spot the fact that each speaker cavity bears two separate 40mm speakers, a 30mm speaker and a 30mm subwoofer. The first image that comes to mind when seeing the word "subwoofer " is the bulky speaker put in an occult position, which takes charge of the reproduction of very low frequencies, which are non-directional. In this case, however, this is not true, as we shall see below.

The contents of this box are limited to the absolutely necessary. These include the headset, an instruction manual and an extra pair of leather pads. The last ones are considered of increasing importance, as more and more users are not pleased with the fabric/mesh pads, which the Saturn X2 wears by default. According to public opinion, leather pads offer better noise isolation but simultaneously cause your ears to warm up during use.

[BREAK=Close snapshots]

The color palette used by the manufacturer is at least elegant. The combination of black and bright red adds character to the product, while silver details highlight its modern design. The microphone is unfortunately not removable or telescopic, while the small bending resistance gave us the impression that the microphone will not survive any abuse.

Even though this headset is made entirely of plastic, it gives the impression that it is very durable and sturdy. The red surfaces (parts of the ear pads and the upper head pad) are made of leatherette. Both the ear pads and the head pad are quite soft [/ B], in order to adjust to the shape of your head. The pads have an internal depth of about 2 cm and are removable by taking off a plastic flange with extreme caution. As a point of size reference, we have put next to it the screen of an iphone 5.

Saturn X2 comes with an inline audio controller, which can adjust the headset's general volume, the volume of back, side and front speakers, the volume of vibrator/subwoofer and enable/disable the microphone. All pots are dimmers apart from the general volume modulator, which is a three-position switch and changes the OS's general volume.

The headset comes equipped with a USB port, which means that the sound card (or more precisely a DAC - Digital to Analog Converter) must be located somewhere on the headset. In this case, we found a CM6206 Taiwanese C-Media chip. This chip includes a 7.1CH DAC, a 2CH ADC (Analog to Digital Converter) for the microphone , the USB host, an audio buffer, PLL and other stuff. According to the manufacturer, this specific IC is recommended for outdoor high performance sound systems, while offering on-board capabilities of changing virtually the speaker placement and processing karaoke functions.

The controller has lighting at the side of its silver details. Perhaps it would be better if the lighting highlighted somehow the potentiometers' labels, as numerous times we've found into the situation of having to turn on the lights in order to adjust the appropriate potentiometer.

As we can see in the pictures above , each cabinet contains 3 speakers capable of reproducing sound from the rear, side and front. By dissecting the cabinet we notice a little clutter in both wiring and their support structure. The drum, which appears to be supported by the three speakers, is a subwoofer . It's just a vibrator, which fuctions similarly to that of a mobile phone. The vibration generated by the vibrator is of low frequencies and, in combination with the design of the cabinet and its structure, converts these vibrations to low frequency sound, simulating the behavior of a subwoofer. Finally, we observe that the wiring is distinct, meaning that each wire "carries" the signal of the corresponding speaker, and starts from the inline controller.

The headset looks particularly beautiful in low light conditions, while matching marvelously with CM Storm Trigger's [/ URL]lighting. The illumination is permanent,with the exception of the microphones, which is active as long as the microphone is enabled.

The plug is -as expected- a USB port, while the cable is braided with an alternating pattern of black and red. We would also like to see gilded end of the USB cable, for enhanced resistance to general wear and tear.

[BREAK=Performance]

General useAs general use we define the use on daily desktop operations, such as navigating the operating system's UI, skype, watching videos on youtube and other multimedia streaming websites.In this area we haven't detected any problems. The headset worked just fine. It proved to be really comfortable and its use in skype conversations the microphone preserved the overall image. The microphone's end light proved particularly useful as it allows us to know if the microphone is enabled or not. We recommend that you do not use this specific headset on official conversations, as the various lights may not emit the necessary serious aura of the situation.One peculiarity we noticed is that when we activate the subwoofer by slightly turning the little dimmer,white noise of low intensity starts appearing. This noise is not distinguishable when playing a music track.Music

In the field of music we chose pieces that cover the whole spectrum of the world of music, from classical to hard rock. A prerequisite for a proper test is to know by heart the songs we listen to, in order to detect any deformation or loss of frequencies in the song. Therefore, we preferred albums of Scorpions, ACDC, Michael Jackson, Elton John, Bryan Adams, John Denver, Beethoven and official film soundtracks such as Frozen, The Hobbit and Inception. All pieces tested were formated in FLAC 44.1KHz 16bits resolution with approximately 900kbps bitrate.Since these headphones are not considered suitable for audiophiles, we didn't have great expectations in this sector. The tracks sounded rich and full, with clear highs and mids . At low frequencies there is a lack of depth, with the bass sounding muffled. The related frequencies exist, but their performance is not as rich as one would expect . In case you choose 2CH input instead of 6CH , all speakers will be used, including the bass vibrators. The latter might probably annoy some, so it is advisable to regulate them at will by the corresponding potentiometer and software .

Movies

We watched fullHD movies with hifi audio of DTS 5.1 48KHz 1500kbps bitrate and more specifically The Hobbit and Riddick. In movies, the behavior is similar to that of the music playback, the main difference being its 5.1 content. Its audible content, that simulates sounds in front and sides of the viewer, was accurately positioned. The side audio effects that are played through the dedicated speakers, which simulate speakers placed directly beside and slightly behind the viewer, were perceived slightly forward than regular.The bass vibrator is a pleasant addition to the whole experience, although in some cases it gave the impression of simply vibrating our ears. Besides, the component implementing the vibration must contain some sort of spring and tapping slightly the cabinet produces a metallic oscillation. This oscillation must be the cause of lack of depth. The situation improved considerably when we introduced the leather pads .Its almost certain that the permanent illumination of the headset may seem annoying to some users with extended peripheral vision and even more to those who wear prescription glasses -as does the writer- because they create a small but significant reflection on the lenses, especially while watching movies in dark environments.

Gaming

There was no way of finishing this review without testing the headset's performance in games and specifically in Battlefield 3 and World of Tanks, where the accuracy of the direction of sounds is vital for surviving the game.Several times I caught myself sitting at the beginning of the map, rotating unnecessarily the player in circles just to hear the curses of my team, which was waiting for me, and the sounds of explosions around me.

A special moment was when a missile that was targeting me, missed me completely, when I happened to rotate the turret of my tank. The experience was very convincing, so much in fact that it disorientated me in the middle of the battle and later I found the next missile with my name on it. My personal assessment is that the increased number of speakers helps the playback of multichannel audio, not so much regarding the direction of the sound as in terms of better playback of sounds to be reproduced simultaneously.

[BREAK=Epilogue]

By purchasing X2's Saturn 5.1, the user must be aware that he will receive exactly what the manufacturer promised. The system's 5.1 function exists and works well, allowing the use of the headset as the playback source of multichannel content found in games and movies . However in some cases the playback quality was not quite as expected. For this reason , the demanding user and audiophile, who is not interested in gaming, could opt for a pair of stereo headphones. Finally, although the design of the lights is beautiful and futuristic, there could be an option to turn them off, either through software or through a dedicated switch, for example when using them in dark environments or during teleconference.

Pros

Real 5.1 experience
Excellent build quality
Extra pair of pads
Beautiful design
Comfortable fit

Cons

Permanent "on" side illumination LEDs
Limited depth in low frequencies
Strong vibration on ears due to vibrator-subwoofer
Adjustment of general volume by changing OS's volume
Absence of illumination where it is definately needed

Many thanks to X2 from Netherlands for offering this review sample. Its price value is estimated at 60€.

Miltiadis Alexios Papadopoulos