Check out the impressive work done for the Computex GIGABYTE Case Mod Show

Subject: General Tech | June 23, 2016 - 01:15 PM |
Tagged: case mods, gigabyte, computex 2016

Gigabyte hosted a showcase of impressive case mods from all across the planet at Computex and TechARP posted a slideshow of the best of them.  They are all quite incredible, going beyond basic modding to the creation of truly unique enclosures, from Gatling guns to Ghostbusters.  It is a pity the forklift wasn't powered by a Steamroller or Bulldozer though.  Check out the full slide show and videos of the cases here.

GIGABYTE-Case-Mod-icon.jpg

"The GIGABYTE case mod showcase featured incredible case mods by top case modders from around the world, including Maciel Barreto from Brazil and Suchao Prowphong from Thailand."

Here is some more Tech News from around the web:

Tech Talk

Source: TechARP

Podcast #405 - AMD RX 480 Hands-on, 32-core Zen rumors, VBIOS scandal and more!

Subject: General Tech | June 23, 2016 - 01:02 PM |
Tagged: Zen, western digital, video, vbios, SM961, sli, Samsung, rx 480, radeon, podcast, My Passport Wireless Pro, msi, GTX 1080, evga, drobo, be quiet, asus, amd, 960 PRO

PC Perspective Podcast #405 - 06/23/2016

Join us this week as we discuss an AMD RX 480 hands-on, 32-core Zen rumors, the ASUS/MSI VBIOS scandal and more!

You can subscribe to us through iTunes and you can still access it directly through the RSS page HERE.

The URL for the podcast is: http://pcper.com/podcast - Share with your friends!

This episode of the PC Perspective Podcast is sponsored by Kaspersky Labs!

Hosts:  Ryan Shrout, Allyn Malventano, Jeremy Hellstrom, and Josh Walrath

Program length: 1:33:07
  1. Week in Review:
  2. This episode is sponsored by Kaspersky Labs!
  3. News items of interest:
  4. Hardware/Software Picks of the Week
    1. Allyn: devCalc Pro - Engineering Mode calculator for iOS
  5. Closing/outro

Dennaton Games Releases Hotline Miami 2 Level Editor

Subject: General Tech | June 22, 2016 - 10:35 PM |
Tagged: pc gaming, hotline miami, devolver digital, dennaton games

The (Edit June 26th: latest entry in the) Hotline Miami franchise was released a little over a year ago, and it was quite popular with both fans and critics. It is a fast-paced, top-down action game that is unforgiving enough that it tends to feel a little bit like a puzzle game as well, at least to me, in the sense that you need to figure out how to catch enough NPCs off-guard to easily and quickly take them out. As such, level design should have a huge impact on gameplay.

devolver-2016-hotlinemiami2level.png

And now users can make their own maps. Dennaton Games and Devolver Digital have released the level editor in today's patch. If you don't like user-generated fun, or you experience bugs or something, then you can stay on Hotline Miami 2 version 1.05.

Hotline Miami 2 and Hotline Miami 1+2 combo are both 75%-off for the rest of today.

Several more RX 480's appear

Subject: Graphics Cards | June 22, 2016 - 04:18 PM |
Tagged: xfx, sapphire, Radeon RX 480, powercolor, gigabyte, asus, amd

An astute reader spotted several more RX 480's on Newegg, lacking clock speeds but providing physical dimensions, albeit with what looks to be a stock image.  All three cards seem to be dual slot designs, XFX's card measuring 10" x 5", ASUS' at 11.8" x 5.4" and Sapphire's a wide bodied 11.8" x 6.5". This could indicate a custom cooler or merely that the cards have rough dimensions listed as opposed to the exact size.

480 two.PNG

Unfortunately the comparison and details page is unavailable so we don't have a way to see the listed clock speeds but we can be sure that they will have three DP 1.2 ports and an HDMI out.  We will keep an eye out for any more leaks we can share with you.

 

Source: Newegg

Dawn of War III is looking ... different

Subject: General Tech | June 22, 2016 - 02:44 PM |
Tagged: gaming, dawn of war III, warhammer 40k

Relic was showing off what DoW III will look like in the usual E3 tradition, with an enhanced 'game play' video.  Heroes are somewhat different than in the previous game, instead of leading a squad they operate on their own, however Gabriel does have an impressive upgrade to his Thunder Hammer.  Also featured is something which is totally not a warjack; an Imperial Knight which is a scaled down Titan with a single pilot.  Generally found guarding Ag. worlds they are the first of the Super Units to be revealed.  Heroes will be Elite Units, faster but somewhat squishier than Super Units which will be much slower, vulnerable to anti-vehicle attacks but able to shrug off most other attacks.  They can be chosen at the beginning of a mission and then deployed with Elite Points which you gain during the mission.  The quotes over at Rock, Paper, SHOTGUN don't have a lot of detail about how the game will play but the video sure is pretty.

"For the benefit of good warboys and wargirls, here’s the not-really-gameplay-despite-what-Relic-say look at a grizzled Gabriel Angelos duffing up some Eldar with the help of his Space Marine chums and a 14-metre mech named Imperial Knight Solaria"

Here is some more Tech News from around the web:

Gaming

Intel still hasn't paid AMD the 1.2 billion USD anti-trust fine

Subject: General Tech | June 22, 2016 - 01:33 PM |
Tagged: Intel, amd, antitrust

This is a saga for the ages and a snit worthy of any 2 year old child.  11 years ago AMD filed suit against Intel citing questionable business tactics Intel had been using worldwide.  Intel was offering discounted parts to retailers if they would use Intel chips exclusively.  For instance, if a company like Dell offered an AMD alternative then Intel would raise the price of every Intel component sold to Dell across the board.  This is, of course, illegal. 

The court cases were settled in 2009, in the US Intel agreed to pay AMD $1.25 billion USD to settle all outstanding court cases in the US and several overseas.  In the UK there was a seperate court case which also went against Intel, the courts there requiring Intel to pay AMD  €1.06bn, the largest ever fine in the UK.  Since then Intel has been fighting tooth and nail to find a way not to pay the fine and while they have not succeeded in their legal battle they have succeeded in not paying AMD one single cent.  Their initial appeal was dismissed in 2014 but that has not stopped Intel from delaying the payment and as of today that fine still remains unpaid.  The Inquirer posted today about their latest challenge to the ruling, Intel's legal team claims that it somehow unfair to be punished for unfair business practices.

Six years on and over 1 billion dollars that should be AMDs is still under a couch cushion in Intel's offices somewhere.

images.jpg

"CHIPMAKER Intel ain't giving up and continues to fight the €1.06bn (around £815m) antitrust fine levied on the firm six years ago."

Here is some more Tech News from around the web:

Tech Talk

Source: The Inquirer

EVGA Shows Off New High Bandwidth "Pro SLI Bridge HB" Bridges

Subject: Graphics Cards | June 22, 2016 - 02:40 AM |
Tagged: SLI HB, nvidia, EVGA SLI HB

Earlier this month we reported that EVGA would be producing its own version of Nvidia's SLI High Bandwidth bridges (aka SLI HB). Today, the company unveiled all the details on its new bridges that we did not know previously, particularly pricing and what the connectors look like.

EVGA is calling the new SLI HB bridges the EVGA Pro SLI HB Bridge and it will be available in several sizes to accommodate your particular card spacing. Note that the 0 slot, 1 slot, 2 slot, and 4 slot spacing bridges are all for two graphics card setups; you will not be able to use these bridges for Tri SLI or Quad SLI setups. While Nvidia did not show the underside of the HB bridges when it first announced them alongside the GTX 1080 graphics card, thanks to EVGA you can finally see what the connectors look like.

EVGA Pro SLI HB Bridge.jpg

As many surmised, the new high bandwidth bridges use both fingers of the SLI connectors on each card to connect the two cards together. Previously (using the old-style SLI bridges), it was possible to connect card A to card B using one set of connectors and Card B to Card C using the second set of connectors for example. Now, you are limited to two card multi-GPU setups. That is the downside; however, the upside is that the HB bridges promise to deliver all of the necessary bandwidth to allow for high speed 4K and NVIDIA Surround display setups. While you will not necessarily see higher frame rates, the HB bridges should allow for improved frame times which will mean smoother gameplay on those very high resolution monitors!

sli_rgb_full_animated.gif

The new SLI bridges are all black with an EVGA logo in the middle that is backlit by an LED. Users are able to use a switch along the bottom edge of the pcb to select from red, green, blue, and white LED colors. In my opinion these bridges look a lot better than the Nvidia SLI HB bridge renders from our computex story (hehe).

Now, as for pricing: EVGA is pricing its SLI HB bridges at $39.99 with the 2 slot spacing and 4 slot spacing bridges available now and the 0 slot and 1 slot spaced bridges set to be available soon (you can sign up to be notified when they are available for purchase). Hopefully reviews will be updated shortly around the net with the new bridges to see what impact they really have on multi-GPU gaming performance (or if they will just be better looking alternatives to the older LED bridges or ribbon bridges)!

Also read: 

Source: EVGA

UCDavis Manufactures a 1000-Core CPU

Subject: Processors | June 21, 2016 - 10:00 PM |
Tagged: ucdavis

Update (June 22nd @ 12:36 AM): Errrr. Right. Accidentally referred to the CPU in terms of TFLOPs. That's incorrect -- it's not a floating-point decimal processor. Should be trillions of operations per second (teraops). Whoops! Also, it has a die area of 64sq.mm, compared to 520sq.mm of something like GF110.

So this is an interesting news post. Graduate students at UCDavis have designed and produced a thousand-core CPU at IBM's facilities. The processor is manufactured on their 32nm process, which is quite old -- about half-way between NVIDIA's Fermi and Kepler if viewed from a GPU perspective. Its die area is not listed, though, but we've reached out to their press contact for more information. The chip can be clocked up to 1.78 GHz, yielding 1.78 teraops of theoretical performance.

These numbers tell us quite a bit.

ucdavis-2016-thousandcorecpu.jpg

The first thing that stands out to me is that the processor is clocked at 1.78 GHz, has 1000 cores, and is rated at 1.78 teraops. This is interesting because modern GPUs (note that this is not a GPU -- more on that later) are rated at twice the clock rate times the number of cores. The factor of two comes in with fused multiply-add (FMA), a*b + c, which can be easily implemented as a single instruction and are widely used in real-world calculations. Two mathematical operations in a single instruction yields a theoretical max of 2 times clock times core count. Since this processor does not count the factor of two, it seems like its instruction set is massively reduced compared to commercial processors. If they even cut out FMA, what else did they remove from the instruction set? This would at least partially explain why the CPU has such a high theoretical throughput per transistor compared to, say, NVIDIA's GF110, which has a slightly lower TFLOP rating with about five times the transistor count -- and that's ignoring all of the complexity-saving tricks that GPUs play, that this chip does not. Update (June 22nd @ 12:36 AM): Again, none of this makes sense, because it's not a floating-point processor.

"Big Fermi" uses 3 billion transistors to achieve 1.5 TFLOPs when operating on 32 pieces of data simultaneously (see below). This processor does 1.78 teraops with 0.621 billion transistors.

On the other hand, this chip is different from GPUs in that it doesn't use their complexity-saving tricks. GPUs save die space by tying multiple threads together and forcing them to behave in lockstep. On NVIDIA hardware, 32 instructions are bound into a “warp”. On AMD, 64 make up a “wavefront”. On Intel's Xeon Phi, AVX-512 packs 16, 32-bit instructions together into a vector and operates them at once. GPUs use this architecture because, if you have a really big workload, you, chances are, have very related tasks; neighbouring pixels on a screen will be operating on the same material with slightly offset geometry, multiple vertexes of the same object will be deformed by the same process, and so forth.

This processor, on the other hand, has a thousand cores that are independent. Again, this is wasteful for tasks that map easily to single-instruction-multiple-data (SIMD) architectures, but the reverse (not wasteful in highly parallel tasks that SIMD is wasteful on) is also true. SIMD makes an assumption about your data and tries to optimize how it maps to the real-world -- it's either a valid assumption, or it's not. If it isn't? A chip like this would have multi-fold performance benefits, FLOP for FLOP.

Source: UCDavis

Whoops! AMD Radeon RX 480 Specifications on Newegg

Subject: Graphics Cards | June 21, 2016 - 08:49 PM |
Tagged: rx 480, Radeon RX 480, polaris 10, Polaris, amd

The AMD Radeon RX 480 is set to launch on June 29th, but a VisionTek model was published a little early (now unpublished -- thanks to our long-time reader, Arbiter, for the heads up). Basically all specifications were already shared, and Ryan wrote about them on June 1st, but the final clock rates were unknown. The VisionTek one, on the other hand, has it listed as 1120 MHz (5.16 TFLOPs) with a boost of 1266 MHz (5.83 TFLOPs).

amd-2016-polaris-rx480visionteknewegg.jpg

Granted, it's possible that the VisionTek model could be overclocked, even though the box and product page doesn't mark it as a factory-overclocked SKU. Also, 5.16 TFLOPs and 5.83 TFLOPs align pretty close to AMD's “>5 TFLOPs” rating, so it's unlikely that the canonical specifications slide underneath this one. Also, TFLOP ratings are basically a theoretical maximum performance, so real-world benchmarks need to be considered for a true measure of performance. That said, this would put the stock RX 480 in the range of a GTX 980 (somewhere above its listed boost clock, and slightly below its expected TFLOP rating when overclocked).

There is no price listed for the 8GB model, but the 4GB version will be $199 USD.

Will people Flip over ASUS' new Chromebook?

Subject: Mobile | June 21, 2016 - 06:50 PM |
Tagged: asus, Chromebook, Chromebook Flip

ASUS' new Chromebook Flip convertible laptop can be yours for about ~$250, not too shabby for a tablet, let alone a laptop.  However for this price a few sacrifices must be made, including the use of Chrome OS as it is a Chromebook after all.  The hardware is a quad-core, 32-bit ARM chip from Rockchip called the RK3288C which can reach up to 1.8GHz.  It also has 4GB of RAM and 16GB of local storage using eMMC flash and a two year subscription to Google drive to give you 100GB of additional storage.  The Tech Report were quite enamoured of this little 10.1", 1280x800 IPS touch screen device, it may not be the fastest machine out there but for the price they felt it to be quiet impressive.

main2.jpg

"Asus' Chromebook Flip is an all-aluminum convertible PC that runs Google's Chrome OS. Its $240-ish price tag puts it in contention with the budget Windows PCs we usually suggest in our mobile staff picks. We put the Flip to the test to see whether it's a worthy Windows alternative."

Here are some more Mobile articles from around the web:

More Mobile Articles

Issues with ASMedia and Zen, or much ado over nothing?

Subject: General Tech | June 21, 2016 - 06:33 PM |
Tagged: amd, asmedia, Zen, usb 3.1

DigiTimes has heard rumours of a possible defect with the ASMedia USB 3.1 controller which will appear on motherboards for AMD's upcoming Zen, which ASMedia have denied and AMD ignored.  The supposed issue stems from increased degradation of transmission speeds over distance which requires the inclusion of additional retimer and redriver chips.  If the issue does exist the worst repercussion will be an increase in manufacturing costs of $2 to $5 per board; even when that charge is passed on to the consumer it will have a very small impact on MSRP and is not likely to raise prices to the realm of Intel motherboards.  As with all rumours take this with a grain of salt, even if it is true it is unlikely to have any major effect on pricing.

asmedia-85228274.jpg

"Commenting on the news, AMD said it is pleased that Zen is on track and will not comment on customer specific board-level solutions., while ASMedia clarified that this is purely a market rumor and its product's signal, stability and compatibility have all passed certification."

Here is some more Tech News from around the web:

Tech Talk

Source: DigiTimes

Fermi, Kepler, Maxwell, and Pascal Comparison Benchmarks

Subject: Graphics Cards | June 21, 2016 - 05:22 PM |
Tagged: nvidia, fermi, kepler, maxwell, pascal, gf100, gf110, GK104, gk110, GM204, gm200, GP104

Techspot published an article that compared eight GPUs across six, high-end dies in NVIDIA's last four architectures: Fermi to Pascal. Average frame rates were listed across nine games, each measured at three resolutions:1366x768 (~720p HD), 1920x1080 (1080p FHD), and 2560x1600 (~1440p QHD).

nvidia-2016-dreamhack-1080-stockphoto.png

The results are interesting. Comparing GP104 to GF100, mainstream Pascal is typically on the order of four times faster than big Fermi. Over that time, we've had three full generational leaps in fabrication technology, leading to over twice the number of transistors packed into a die that is almost half the size. It does, however, show that prices have remained relatively constant, except that the GTX 1080 is sort-of priced in the x80 Ti category despite the die size placing it in the non-Ti class. (They list the 1080 at $600, but you can't really find anything outside the $650-700 USD range).

It would be interesting to see this data set compared against AMD. It's informative for an NVIDIA-only article, though.

Source: Techspot

Brace Yourselves, Samsung SM961, PM961, 960 PRO and 960 EVO SSDs Are Coming!

Subject: Storage | June 21, 2016 - 04:02 PM |
Tagged: V-NAND, SM961, Samsung, PM961, 960 PRO, 960 EVO, 48-layer

We've known Samsung was working on OEM-series SSDs using their new 48-layer V-NAND, and it appears they are getting closer to shipping in volume, so here's a peek at what is to come:

s-l1600.jpg

First up are the SM961 and PM961. The SM and PM appear to be converging into OEM equivalents of the Samsung 'PRO' and 'EVO' retail product lines, with MLC flash present in the SM and TLC (possibly with SLC TurboWrite cache) in the PM. The SM961 has already been spotted for pre-order over at Ram City. Note that they currently list the 1TB, 512GB, and 256GB models, but at the time of this writing, all three product titles (incorrectly) state 1TB. That said, pricing appears to be well below the current 950 PRO retail for equivalent capacities.

s-l500.jpg

These new parts certainly have impressive specs on paper, with the SM961 claiming a 25-50% gain over the 950 PRO in nearly all metrics thanks to 48-layer V-NAND and an updated 'Polaris' controller. We've looked at plenty of Samsung OEM units in the past, and sometimes specs differ between OEM and retail parts, but it is starting to make sense for Samsung to simply relabel a given OEM / retail part at this point (minus any vendor-requested firmware detuning, like reduced write speeds in favor of increased battery life, etc).

With that are the other two upcoming parts that do not appear on the above chart. Those will be the 960 PRO and EVO, barring any last second renaming by Samsung. Originally we were expecting Samsung to add a 1TB SKU to their 950 PRO line, but it appears they have changed gears and will now shift their 48-layer parts to the 960 series. The other big bonus here is that we should also be getting an EVO, which would mark Samsung's first retail M.2 PCIe 3.0 x4 part sporting TLC flash. That product should come in a lot closer to 850 EVO pricing, but offer significantly greater performance over the faster interface. While we don't have specs on these upcoming products, the safe bet is that they will come in very close (if not identical) to that of the aforementioned SM961 and PM961.

48-V-NAND.png

Samsung's 48-Layer V-NAND, dissected by TechInsights
(Similar analysis on 32-Layer V-NAND here)

All of these upcoming products are based on Samsung's 48-layer V-NAND. Announced late last year, this flash has measurably reduced latency (thanks to our exclusive Latency Percentile testing) as compared to the older 32-layer parts. Given the performance improvements noted above, it seems that even more can be extracted from this new flash when connected to a sufficiently performant controller. Previous controllers may have been channel bandwidth limited on the newest flash, where Polaris can likely open up the interface to higher speed grades.

We await these upcoming launches with baited breath. It's nice to see these parts inching closer to the saturation point of quad lane PCIe 3.0. Naturally there will be more to follow here, so stay tuned!

Drobo Updates 5D to Turbo Edition 5Dt

Subject: Storage | June 21, 2016 - 02:43 PM |
Tagged: usb 3.0, Thunderbolt 2, raid, hdd, drobo, DAS, BeyondRAID, 5Dt, 5D

Today Drobo updated their 5D, shifting to Thunderbolt 2, an included mSATA caching SSD, and faster internals:

5bay-front-header.png

The new 5Dt (t for Turbo Edition) builds on the strengths of the 5D, which launched three years ago. The distinguishing features remain the same, as this is still a 5-bay model with USB 3.0, but the processor has been upgraded, as well as the USB 3.0 chipset, which was a bit finicky with some earlier implementations of the technology.

rear.png

The changes present themselves at the rear, as we now have a pair of Thunderbolt 2 (20 Gb/s) ports which support display pass-through (up to 4k). Rates speeds climb to 540 MB/s read and 250 MB/s write when using HDDs. SSDs bump those figures up to 545 / 285 MB/s, respectively.

bottom.jpg

Another feature that has remained was their Hot Data Cache technology, but while the mSATA part was optional on the 5D, a 128GB unit comes standard and pre-installed on the 5Dt.

The Drobo 5Dt is available today starting at $899. That price is a premium over the 5D, but the increased performance specs, included SSD, and Thunderbolt connectivity come at a price.

lineup.png

The current (updated) Drobo product lineup.

Full press blast after the break.

Source:

Looking to build your first PC this summer? We have a guide for you!

Subject: General Tech | June 21, 2016 - 02:32 PM |
Tagged:

We don't usually do this, but I have been getting a lot of emails and messages on social media from gamers looking to build their first PC this summer. With the release of the high end GeForce GTX 1080 and GTX 1070 cards this month, and the pending release of the Radeon RX 480 for more budget-minded gamers, there will likely never be a better time to get into PC gaming than now!

Back in February my nephew wanted to undertake building his own gaming PC for the first time. I took that opportunity to build an article and three video series for enthusiasts and DIYers that were either new to the game or needed a refresher on how to put screws to PCB, so to speak. With the numerous emails and messages I've been getting, I thought now would be a great time to bump our story back here and showcase to everyone how easy it can be to build your own PC, whether it be for gaming, VR, productivity or anything else.

gbsponsor.jpg

You can find the original story right here, sponsored by Gigabyte, but I have also re-embedded the videos below. Yes, the component selections we used in February could use some updating on the graphics card, monitor and maybe power supply, but the rest of the build summary is spot on and the build process remains unchanged.

Good luck to all the budding enthusiasts out there!

Allyn isn't the only one with a Lapdog on his couch

Subject: General Tech | June 20, 2016 - 05:40 PM |
Tagged: RGB, mouse, lapdog, keyboard, gaming control center, couchmaster, Couch, corsair

The Tech Report would like to back Al up in saying that gaming on a TV from the comfort of your couch is not as weird as some would think.  In their case it was Star Wars Battlefront and Civilization V which were tested out, Battlefront as it is a console game often played on a TV and Civ5 as it is not a twitch game and the extra screen real estate is useful.  They also like the device although they might like a smaller version so that keyboards without a numpad did not leave as much room ... perhaps a PocketDog?  Check out their quick review if Al's review almost sold you on the idea.

comparison2.png

"Corsair's Lapdog keyboard tray is built to bridge the gap between the desk and the den by giving gamers a way to put a keyboard and mouse right on their laps. We invited the Lapdog into our living room to see whether it's a good boy."

Here is some more Tech News from around the web:

Tech Talk

Windows 10 versus Ubuntu 16.04 versus NVIDIA versus AMD

Subject: Graphics Cards | June 20, 2016 - 04:11 PM |
Tagged: windows 10, ubuntu, R9 Fury, nvidia, linux, GTX1070, amd

Phoronix wanted to test out how the new GTX 1070 and the R9 Fury compare on Ubuntu with new drivers and patches, as well as contrasting how they perform on Windows 10.  There are two separate articles as the focus is not old silicon versus new but the performance comparison between the two operating systems.  AMD was tested with the Crimson Edition 16.6.1 driver, AMDGPU-PRO Beta 2 (16.20.3) driver as well as Mesa 12.1-dev.  There were interesting differences between the tested games as some would only support one of the two Linux drivers.  The performance also varies based on the game engine, with some coming out in ties, others seeing Windows 10 pull ahead and even some cases where your performance on Linux was significantly better.

NVIDIA's GTX 1080 and 1070 were tested using the 368.39 driver release for Windows and the 367.27 driver for Ubuntu.  Again we see mixed results, depending on the game Linux performance might actually beat out Windows, especially if OpenGL is an option. 

Check out both reviews to see what performance you can expect from your GPU when gaming under Linux.

image.php_.jpg

"Yesterday I published some Windows 10 vs. Ubuntu 16.04 Linux gaming benchmarks using the GeForce GTX 1070 and GTX 1080 graphics cards. Those numbers were interesting with the NVIDIA proprietary driver but for benchmarking this weekend are Windows 10 results with Radeon Software compared to Ubuntu 16.04 running the new AMDGPU-PRO hybrid driver as well as the latest Git code for a pure open-source driver stack."

Here are some more Graphics Card articles from around the web:

Graphics Cards

Source: Phoronix

NVIDIA Announces PCIe Versions of Tesla P100

Subject: Graphics Cards | June 20, 2016 - 01:57 PM |
Tagged: tesla, pascal, nvidia, GP100

GP100, the “Big Pascal” chip that was announced at GTC, will be coming to PCIe for enterprise and supercomputer customers in Q4 2016. Previously, it was only announced using NVIDIA's proprietary connection. In fact, they also gave themselves some lead time with their first-party DGX-1 system, which retails for $129,000 USD, although we expect that was more for yield reasons. Josh calculated that each GPU in that system is worth more than the full wafer that its die was manufactured on.

nvidia-2016-gp100tesla.jpg

This brings us to the PCIe versions. Interestingly, they have been down-binned from the NVLink version. The boost clock has been dropped to 1300 MHz, from 1480 MHz, although that is matched with a slightly lower TDP (250W versus the NVLink's 300W). This lowers the FP16 performance to 18.7 TFLOPs, down from 21.2, FP32 performance to 9.3 TFLOPs, down from 10.6, and FP64 performance to 4.7 TFLOPs, down from 5.3. This is where we get to the question: did NVIDIA reduce the clocks to hit a 250W TDP and be compatible with the passive cooling technology that previous Tesla cards utilize, or were the clocks dropped to increase yield?

They are also providing a 12GB version of the PCIe Tesla P100. I didn't realize that GPU vendors could selectively disable HBM2 stacks, but NVIDIA disabled 4GB of memory, which also dropped the bus width to 3072-bit. You would think that the simplicity of the circuit would want to divide work in a power-of-two fashion, but, knowing that they can, it makes me wonder why they did. Again, my first reaction is to question GP100 yield, but you wouldn't think that HBM, being such a small part of the die, is something that they can reclaim a lot of chips by disabling a chunk, right? That is, unless the HBM2 stacks themselves have yield issues -- which would be interesting.

There is also still no word on a 32GB version. Samsung claimed the memory technology, 8GB stacks of HBM2, would be ready for products in Q4 2016 or early 2017. We'll need to wait and see where, when, and why it will appear.

Source: NVIDIA

If you bought directly from Acer over the past year, double check your spam and email

Subject: General Tech | June 20, 2016 - 01:21 PM |
Tagged: acer, security

North American customers of Acer who bought directly from them between May 12, 2015 and April 28, 2016 may have had their credit card numbers compromised.  Their less than secure customer database contained customer names, addresses, card numbers, and three-digit security verification codes all of which have been siphoned off at least once.  If this breach effected your account Acer will be sending a notification to you, you can see an example at The Register if you want to be sure you are receiving a valid notification.  For those who have seen fraudulent charges already this will be too late to mitigate their pain but anyone who used Acer's online shop during that time period would do well to get their cards changed.

Acer_logo_new.jpg

"Acer's insecure customer database spilled people's personal information – including full payment card numbers – into hackers' hands for more than a year."

Here is some more Tech News from around the web:

Tech Talk

Source: The Register

NVIDIA Releases 368.51 Hotfix Driver

Subject: Graphics Cards | June 18, 2016 - 10:37 PM |
Tagged: nvidia, graphics drivers

GeForce Hotfix 368.51 drivers have been released by NVIDIA through their support website. This version only officially addresses flickering at high refresh rates, although its number has been incremented quite a bit since the last official release (368.39) so it's possible that it rolls in other changes, too. That said, I haven't heard too many specific issues with 368.39, so I'm not quite sure what that would be.

nvidia-2015-bandaid.png

As always with a hotfix driver, NVIDIA pushed it out with minimal testing. It should pretty much only be installed if you have a specific issue (particularly the listed one(s)) and you don't want to wait until one is released that both NVIDIA and Microsoft looked over (although Microsoft's WHQL certification has been pretty lax since Windows 10).

Oddly enough, they only seem to list 64-bit links for Windows 8.1 and Windows 10. I'm not sure whether this issue doesn't affect Windows 7 and 32-bit versions of 8.1 and 10, or if they just didn't want to push the hotfix out to them for some reason.

Source: NVIDIA