Nvidia Pascal GP104 Chip Pictured!


At the rate things are going, Pascal may end up the most hyped graphics card release in recent memory. So far we’ve been treated to die shots of GP106, the low-end chip used in the Drive PX 2. The GP104, set to release a little over a month from now, has remained hidden from view, till today.

Coming out from Chinese forums, we can finally see what GP104 looks like. The SKU we have before us is the GP104-200, the one that is expected to bed used for the GTX 1070, the middle of the pack for GP104. The die size comes in at about 333 mm which is remarkably similar to the GM204. Around the die, we are able to see some Samsung 8Gbps 8Gb modules. This is what we have expected for the GTX 1070 and suitable for the GTX 1080 if it doesn’t use GDDRRX.

Overall, the leak today confirms what we already know about Nvidia’s plans. They will start off with the small GXx04 die as the “flagship” and follow it up with the bigger die later on. If the die size is correct, we maybe seeing a decently larger chip than the GM204 due to the die shrink. However, Pascal may bring back some of the compute cut out in Maxwell so it remains to be seen how the chip will truly perform.

FreshTech Solutions Ultimate Buyer’s Guide

FreshTech Solutions is one of the leading system integrators in the UK and forged an impressive reputation for their huge range of pre-configured PCs. Whether you’re on a tight budget looking to try out PC gaming for the first time, or a content creator seeking the absolute best performance, there’s something to suit your requirements. Each PC is backed by generous warranty period and it’s even possible to purchase additional cover for further peace-of-mind. For example, on systems costing £1000 and above, you can increase the warranty length from 3 years to 5 years for £25. If you’re opting for a cheaper rig under the £1000 mark, then it’s possible to select three years cover for £25 and five years support at a cost of £50. This level of flexibility allows you to protect your investment for a period which you expect to use the system for.

When selecting a custom PC, the number of configurations and pricing variation means it’s extraordinarily difficult for novices to judge which model to go for. It’s not always a sensible approach to spend more if you’re unlikely to reap the benefits. Therefore, we’ve compiled a list of custom PCs from FreshTech Solutions that we are happy to recommend at various price points. Hopefully, our readers will find this useful and those without technical knowledge can make an informed decision about which unit suits their usage scenario.


£529.00

Specification:

  • Skylake i5 6400
  • 8gb Crucial DDR4 2133MHz
  • 120gb Kingston V300 SSD
  • 24x DVDRW
  • EVGA 500w 80Plus White PSU
  • Nvidia GTX 950 2GB
  • Gigabyte H110M-S2H
  • Falcon Blue Case With Side Window
  • Windows 10 Home 64bit
  • 1 Year Collect and Return Warranty ( 1 Year Collect, 1 Year Parts, 1 Year Labour)

The first machine we’re taking a look at targets consumers on a very tight budget who want to enter into the Intel ecosystem. While you could compile a cheaper build using an AMD CPU and AM3+ motherboard, it doesn’t offer much headroom in the future if you suddenly demand additional horsepower. The majority of games are GPU bound but there’s some examples including ARMA III and more recently, Black Desert Online which rely heavily on a system’s CPU performance. This particular build features a large enough capacity boot SSD to install the operating system and enjoy a snappier feel in Windows.

Also, the GTX 950 2GB graphics card is a great choice for 1080P gaming providing you turn down a few settings. On another note, the lack of overclocking isn’t an issue at this price and won’t impact on the user experience during graphically intensive games. Ideally, I’d recommend paying extra for the next tier, but this is a great choice for beginners moving away from consoles for the very first time.


£699.00

Specification:

  • Skylake i5 6600K
  • 8GB Corsair Vengeance LPX 3000MHz
  • Samsung EVO 250GB SSD
  • 24x DVDRW
  • EVGA 500w 80Plus White PSU
  • NVIDIA GTX 960 2GB
  • Gigabyte Z170-Gaming K3
  • Titan Blue Case with Side Window
  • Windows 10 Home 64bit
  • 1 Year Collect and Return Warranty ( 1 Year Collect, 1 Year Parts, 1 Year Labour )

The next system which retails for £699.00 offers overclocking functionality due to the unlocked K processor and Z170 chipset. Furthermore, the Gigabyte Z170-Gaming K3 contains an illuminated LED strip on the PCB to create a more ostentatious appearance. This higher end model includes the GTX 960 2GB GPU to maintain improved frame-rates at higher settings. It’s a significant boost over the GTX 950 and worth considering given the constant influx of modern titles requiring additional graphical grunt. Of course, it’ s still a budget option and pales in comparison to higher end graphics cards, but it should provide an enjoyable user-experience with a 1920×1080 display. Also, the boot SSD’s higher capacity allows you to store your favourites games and benefit from faster load times.


£899.00

Specification:

  • Skylake i5 6600K
  • 16GB Corsair Vengeance LPX 3000MHz
  • Samsung EVO 250GB SSD
  • WD 1TB WD10EZEX HDD
  • 24x DVDRW
  • Corsair CX600 80Plus Bronze PSU
  • NVIDIA GTX 970 4GB
  • Gigabyte Z170-Gaming K3
  • Titan Red Case with Side Window
  • Windows 10 Home 64bit
  • 1 Year Collect and Return Warranty ( 1 Year Collect, 1 Year Parts, 1 Year Labour )

The next tier’s system costs £899.99 and has a number of key improvements including a secondary mechanical hard disk, 16GB RAM and GTX 970 graphics card. When discussing the previous systems it’s important to remember that I’d recommend adding a mechanical drive to store large sums of data. This allows you to easily install various games while ensuring the boot SSD doesn’t fill up rather quickly. Thankfully, this doesn’t increase the asking price by an exponential amount. On another note, the 16GB memory is useful when working with video, photo editing and other complex workloads. It’s not really necessary in games just yet but I wouldn’t be surprised if 16GB becomes the standard in the next few years.

NVIDIA’s GTX 970 offers absolutely stunning performance even with a 2560×1440 monitor and can exceed 60 frames-per-second in various titles including the 2013 Tomb Raider reboot. It’s unquestionably one of the best price to performance GPUs on the market and a suitable choice for a powerful gaming system.


£1049.00

 

Specification:

  • Skylake i5 6600K Tested Overclock to 4.6GHz
  • Corsair H80I V2 Water Cooler
  • 16GB Corsair Vengeance LPX 3000MHz
  • Samsung EVO 250GB SSD
  • Seagate 1TB HDD
  • 24x DVDRW
  • Corsair RM750x 80Plus Gold PSU
  • NVIDIA GTX 970 4GB
  • Gigabyte Z170-Gaming K3
  • Falcon Gaming Case With Side Window
  • Windows 10 Home 64bit
  • 3 Year Collect and Return Warranty ( 1 Year Collect, 2 Years Parts, 3 Years Labour )

Moving towards the higher end market is a system which retails for £1049.00. This model contains a higher wattage power supply and Corsair water cooling unit to attain better temperatures. Furthermore, FreshTech Solutions guarantees the system will reach 4.6GHz and tests each overclock extremely carefully. This will be incredibly useful to consumers who feel apprehensive about overclocking their CPU and haven’t got any experience in this field. They really shouldn’t be concerned because the process is ridiculous simple. However, whenever anything goes awry, it’s important to have an excellent level of after-care support.


£1599.00

Specification:

  • Skylake i7 6700K Tested Overclock to 4.5GHz
  • Corsair H80I V2 Water Cooler
  • 16GB Corsair Vengeance LPX 3000MHz
  • Samsung EVO 500GB SSD
  • Seagate 2TB HDD
  • 24x DVDRW
  • Corsair RM750x 80Plus Gold PSU
  • NVIDIA GTX 980TI 6GB
  • Gigabyte Z170X-Gaming 5
  • Fractal Design R5 With Side Window
  • Windows 10 Home 64bit
  • 3 Year Collect and Return Warranty ( 1 Year Collect, 2 Years Parts, 3 Years Labour )

The £1599.00 system swaps out the i5 6600K for Intel’s enthusiast flagship i7-6700K on the LGA1151 chipset. Once again, FreshTech Solutions guarantees an impressive minimum overclock to reach significant gains compared to stock values. Also, the 500GB SSD is a great addition which allows you to install various applications without constantly checking on the remaining capacity. The system is housed in the highly regard Fractal Design R5 chassis which creates a more premium feel. Clearly, the main improvement revolves around the GTX 980Ti and is a major step up when compared to the GTX 970.

Even though the GPU isn’t capable of running games at maximum details on a 4K display, it’s able to really reach high figures on a 2560×1440 or 3440×1440 monitor. If you want to improve the performance even further, it’s possible to add a second GTX 980Ti due to the power supply’s 750 watts rating. The Gigabyte Z170X-Gaming 5 motherboard supports USB 3.1, SATA Express, multi-GPU configurations and opts for a superb audio solution. Once combined, this offers an absolutely staggering user-experience.


£1938.00

Specification:

  • Intel Core i7 5930K 3.50GHz
  • Corsair H80I V2 Water Cooler
  • 16GB Corsair Vengeance LPX DDR4 3000MHz
  • 512GB Samsung 950 Pro M.2 Solid State Drive
  • Seagate 2TB Hard Drive
  • 24x DVDRW
  • Corsair RM750x 80Plus Gold PSU
  • NVIDIA GTX 980TI 6GB
  • Gigabyte X99-SLI
  • Fractal Design R5 With Side Window
  • Windows 10 Home 64bit
  • 3 Year Collect and Return Warranty ( 1 Year Collect, 2 Years Parts, 3 Years Labour )

Up to this point, we’ve focussed on the consumer LGA1151 chipset which offers great value for typical desktop tasks. Unfortunately, the platform only supports up to a 4-core 8-thread CPU. This severely limits the computational power during multi-threaded workloads. As a result, many professionals opt for 2011-v3 powered systems because they can leverage extra performance using 6-core or even 8-core processors. The next system totalling £1938.00 utilises a hex core processor with 40 PCI-E lanes and a 2TB hard disk.

Not only that, the Samsung 950 Pro NVMe boot drive is capable of an astounding maximum read of 2500MB/s and maximum write reaching 1500MB/s. This allows games to load faster and helps to process huge 4K video projects in a more efficient manner. Of course, this elite-grade PC opts for the best possible hardware including the GTX 980Ti and Gigabyte X99-SLI motherboard. This means you can have a powerful workstation and play demanding games in your leisure time.

If you’d like us to review any of the systems mentioned above, please let us know in the comments section below.

Nvidia Release Pascal Architecture White Paper

When AMD and Nvidia release all of those TFLOPs numbers, it’s important to realize that those are theoretical maximums. In order of a chip to reach that number, its architecture has to be extremely efficient and powerful to use. When Pascal first launched, Nvidia released some details about what it would look like. With the release of the white paper for the architecture, there are a few additional highlights worth noting.

First off, we know that Pascal has cut the SM (Stream Multiprocessor) down from 128 FP32 cores to 64. This allows for better distribution of processing power to tasks and as each SM keeps the same amount of register files and other support hardware, throughput is increased overall. Nvidia has also tweaked the SM so datapaths are more streamlined and sharing information within the SM takes less power and hardware. The scheduler has also seen some improvements and updates to ensure the SM is constantly being fed.Cache sizes have also been increased from 3MB to 4MB and a dedicated shared memory space of 64KB per SM has been added This is lower than the 96KB per SM in Maxwell but if you consider the doubled SM count relative to a same size Maxwell chip, it’s actually an increase of 16KB per SM. Finally, Nvidia detailed the P100 interposer layout for HBM2, something for us to look forward to when HBM2 finally arrives.

For other details, be sure to check out our earlier write up on the Pascal architecture.

Nvidia Pascal GTX 1070 and GTX 1080 Prices Leaked

With Computex just over a month away, leaks have been popping up daily about Nvidia’s upcoming Pascal graphics cards. So far, we’ve seen a lot of leak around the physical card but nothing yet on core specifications or pricing. Today, according to Taiwanese insiders, we are getting a look at how much GP104 will set buyers back. The GTX 1070 will reportedly sell for NT $ 19990 and the GTX1080 will sell for NT $ 27990.

Converting the prices, you get around $620 and $870 for the 1070 and 1080 respectively. That seems a bit high considering that you can find the 980 at about $500 and the 980Ti at $700. However, if you only consider NT prices, that’s exactly in line with what the GTX 980 and 980Ti cost in Taiwan respectively. This suggests to me that the GP104 cards would be priced where the GTX 980 and 980Ti are right now based on the market they are selling in

These prices are in line with what we’ve come to expect from Nvidia and mirror the GTX 670/680 and GTX 970/980 launches. The GTX 1070 will significantly less and offer competitive performance and the GTX 1080 will command the extra flagship premium. While high the prices are understandable given the new process and architecture. It’s just a shame that the top end GP100 won’t be the true flagship anymore at launch.

Ashes of the Singularity DirectX 12 Graphics Performance Analysis

Introduction


Ashes of the Singularity is a futuristic real-time strategy game offering frenetic contests on a large-scale. The huge amount of units scattered across a number of varied environments creates an enthralling experience built around complex strategic decisions. Throughout the game, you will explore unique planets and engage in enthralling air battles. This bitter war revolves around an invaluable resource known as Turinium between the human race and masterful artificial intelligence. If you’re into the RTS genre, Ashes of the Singularity should provide hours of entertainment. While the game itself is worthy of widespread media attention, the engine’s support for DirectX 12 and asynchronous compute has become a hot topic among hardware enthusiasts.

DirectX 12 is a low-level API with reduced CPU overheads and has the potential to revolutionise the way games are optimised for numerous hardware configurations. In contrast to this, DirectX 11 isn’t that efficient and many mainstream titles suffered from poor scaling which didn’t properly utilise the potential of current graphics technology. On another note, DirectX 12 allows users to pair GPUs from competing vendors and utilise multi graphics solutions without relying on driver profiles. It’s theoretically possible to achieve widespread optimization and leverage extra performance using the latest version of DirectX 12.

Of course, Vulkan is another alternative which works on various operating systems and adopts an open-source ideology. Although, the focus will likely remain on DirectX 12 for the foreseeable future unless there’s a sudden reluctance from users to upgrade to Windows 10. Even though the adoption rate is impressive, there’s a large number of PC gamers currently using Windows 7, 8 and 8.1. Therefore, it seems prudent for developers to continue with DirectX 11 and offer a DirectX 12 render as an optional extra. Arguably, the real gains from DirectX 12 will occur when its predecessor is disregarded completely. This will probably take a considerable amount of time which suggests the first DirectX 12 games might have reduced performance benefits compared to later titles.

Asynchronous compute allows graphics cards to simultaneously calculate multiple workloads and reach extra performance figures. AMD’s GCN architecture has extensive support for this technology. In contrast to this, there’s a heated debate questioning if NVIDIA products can even utilise asynchronous compute in an effective manner. Technically, AMD GCN graphics cards contain 2-8 asynchronous compute cores with 8 queues per core which varies on the model to provide single cycle latencies. Maxwell revolves around two pipelines, one designed for high-priority workloads and another with 31 queues. Most importantly, NVIDIA cards can only “switch contexts at draw call boundaries”. This means the switching process is slower and gives AMD and a major advantage. NVIDIA has dismissed the early performance numbers from Ashes of the Singularity due to its current development phase. Finally, the game’s release has exited the beta stage which allows us to determine the performance numbers after optimizations were completed.

Zotac Announces PCI-E x1 GeForce GT 710 Graphics Card

Some time ago, NVIDIA unveiled the GT 710 graphics card designed for HTPCs and relatively basic usage scenarios. The company claimed that you could experience performance gains up to 10 times better than integrated graphics solutions. Of course, it’s not suited to demanding applications which is reflected in the price point and form factor. The GT 710 doesn’t require any dedicated power connectors and utilizes the PCI Express format instead. Up to this point, custom GT 710 cards from manufacturers including Inno3D, EVGA and others have employed the PCI-E x16 interface. Zotac’s latest model bucks the trend and opts for the x1 interface.

The GPU is passively cooled and supports D-Sub, HDMI and DVI-D. Furthermore, it’s capable of driving displays up to 2560×1600 and opts for a WDDM 2.0 compliant driver. Technically, the Zotac version is clocked at 954MHz and includes 1GB DDR3L memory at 1600MHz. The PCI-E interface means you can use the card in expansions slots which traditionally remain free. This allows you to keep the x16 slots full with fibre-channel cards, enterprise HBAs and more. Clearly, the GT 710’s gaming credentials are fairly basic but they are a better option than many iGPUs. Saying that, I wouldn’t recommend it and there’s greater value when purchasing a higher performing product. The Zotac GT 710 might be useful if you’re watching videos and want to install a dedicated card.

Are you a fan of low power cards like the GT 710 or feel they are pointless due to the good performance levels on APUs?

Nvidia GTX 1080/1070 Shroud Teardown Appears Online

Earlier in the week, we got our first glimpse of what might be the GTX 1080 might look like. That was pretty much what we expected given what saw earlier of leaked GTX 1080/1070 cooler shrouds. Today, VideoCardz has found someone who has been able to tear down the cooler shroud of the 1080/1070 and show us just how they are made. The shroud reportedly comes in 4 separate parts that are put together to form the entire component.

In the case of this specific leak, the process used is die-sinking. The biggest part comes with the GTX 10_0 punched out, with the third spot left uncut. This will allow for easier reuse to turn it either in GTX 1070 or GTX 1080 shrouds as production needs dictate. The polygonal fan bracket itself is made of magnesium alloy and is the next largest piece. Finally, we have a Nvidia logo and a side panel that has GeForce GTX punched out.

More and more leaks are expected given the fact that GP104 has already entered mass production. There is no way for GP104 to arrive at Computex unless Nvidia already has everything being made. At this point, we should be seeing GP104 cards starting to be shipped to stores in preparation for the launch. All we need now is a leak confirming the GTX 1060Ti.

Nvidia Launches GM206 Based NVIDIA Quadro M2000

Even with Pascal just around the corner, Nvidia isn’t letting up with new graphics card launches. After AiB partners launched a series of low power GTX 950s based off of the GM206, Nvidia has launched their own card. Slotting into the professional lineup, the new Quadro M2000 features a full GM206 and is pretty much a 75W GTX 950 rebranded for the professional use.

At 768 CUDA cores and 1.3 TFLOPS of SP performance, the M2000 is the lowest member of the Maxwell Quadro lineup. The card replaces the Kepler-based K2200 which featured 640 CUDA cores. The card features 4GB of GDDR5 over a 128bit, offering 106 GB/s of bandwidth, 26GB/s more than it’s predecessor. Just like the recent slew of 75W GTX 950 cards, the M2000 won’t require a PCIe power connector, coming in with a slim single slot form factor with four DisplayPort 1.2.

With Maxwell just over a month away, it’s a really odd time for Nvidia to be releasing a new card. This leads me to suggest that Nvidia may not have any low-end Pascal cards out for the rest of 2016 except maybe a few mobile chips. From what we’ve heard, only GP104 will be unveiled at Computex so GP106 may not arrive for a while. A single Quadro M2000 will set you back 569 EUR.

Nvidia Pascal GP106 To Arrive Late 2016

Both AMD and Nvidia have tended to launch higher end flagships before moving to fill the rest of the lineup. The same is expected for Nvidia’s Pascal launch, with the “flagship” GP104 based cards set to arrive before the midrange GP106 does. As we know GP104 will arrive with the GTX 1080 and 1070 at Computex and reports are surfacing that we won’t see GP106 till the autumn of this year.

By holding back the release, Nvidia balances out their sales and gives a chance for last gen cards to get be cleared out of stock first. This is especially true of the low-end as GM206, the current Maxwell midrange chip, is still in production. Many of Nvidia’s AiB partners also recently launched new 75W GTX 950 cards, raising suggestions that Nvidia is preparing cards to counter the more power efficiency Polaris 11 which should arrive before GP106.

With GP104 reaching down to the 1060Ti, the GP106 will likely power the GTX 1060, 1050Ti and maybe even the GTX 1050 as well. According to the source, the chip will also not require a PCIe power connector, with a TDP of at least under 75W. This would put is TDP near that of Polaris 11 which is also under 75W. With GP106 already spotted on the Drive PX2, it’s only a matter of time though till Nvidia launches the cards,

Is This Photo of The GTX 1080 Genuine?

NVIDIA’s upcoming architecture codenamed Pascal is rumoured to launch in June and apparently invites have already been sent out to members of the press for an early briefing on its unique features. The new range is built on the 16nm manufacturing process and could utilize the latest iteration of high bandwidth memory. Although, the mid tier options might opt for GDDR5X to cut costs and maintain a lower price point. Of course, there’s always leaks whenever a new architecture is being readied from either graphics company. However, NVIDIA has kept the information extremely secret and there’s not much to go on when trying to work out the specification across various products. Some time ago, a leaked image supposedly showcased the new cooler shroud and GTX 1000 branding. This looked fairly credible given the environmental setting and high quality of workmanship.

Today, a brand new image has arrived courtesy of Chinese website Baidu which apparently shows the GTX 1080 in full for the first time:

The cooler opts for a dynamic appearance with sharp edges compared to the older reference design. It also corresponds with the previous leak which suggests both images are credible. On the other hand, it’s important to adopt a cynical approach just in case someone made the effort of modding the stock cooler to foil people around the globe. Honestly, this is very unlikely and it’s probably the final design for the GTX 1080. Sadly, it’s impossible to tell if the GPU contains a stylish backplate. Although, this should be revealed soon if the release date reports are correct. Whatever the case, it looks like NVIDIA has tweaked the cooler design and opted for the GTX 1000 series name. I’m not so sure this is a clever move as the 1080 might cause some confusion.

Do you think the image discussed in the article is genuine?

Colorful Announces GTX 980 Ti iGame KUDAN

Even though the current Maxwell line-up is set to be replaced by Pascal relatively soon if recent revelations are true, it hasn’t stopped new custom models from entering production. Colourful, a vendor offering high-end solutions primarily in the Chinese market has announced their latest product based on the GM200 core. This particular GPU entitled, the GTX 980 Ti iGame KUDAN utilizes an overclocked core of 1203MHz and boost reaching 1304MHz. On another note, the graphics card is powered by two 8-pin PCI-E connectors and includes three DisplayPort connectors, one HDMI and a DVI-I port. Of course, it’s built on the 28nm process and comes with 6GB GDDR5 memory using a 384 bit bus.

The GTX 980 Ti iGame KUDAN opts for a unique cooling solution with an integrated water cooling loop and five heatpipes. Furthermore, the densely packed fin array provides exceptional thermal dissipation and should allow for some impressive manual overclocking performance. The water cooling aspect even allows you to add the GPU to a custom loop via the embedded block mount.

Colourful have employed silver plating technology to improve the PCB’s reliability compared to traditional copper alternatives. This also reduces oxidation and should allow for a longer lifespan.

The graphics card employs an impressive power delivery to enable large overclocks, and there’s EMI shielding throughout. This hefty cooling solution and impeccable components makes it a great choice of users demanding the ultimate in performance. While the timing is a little strange given the upcoming release of Pascal, I highly doubt the GTX 980 Ti is suddenly going to become outclassed by lower end options. However, its price might reduce by a decent amount once the new cards arrive.

Do AMD Drivers Really Deserve Such a Hostile Reception?

Introduction


AMD has a serious image problem with their drivers which stems from buggy, unrefined updates, and a slow release schedule. Even though this perception began many years ago, it’s still impacting on the company’s sales and explains why their market share is so small. The Q4 2015 results from Jon Peddie Research suggests AMD reached a market share of 21.1% while NVIDIA reigned supreme with 78.8%. Although, the Q4 data is more promising because AMD accounted for a mere 18.8% during the last quarter. On the other hand, respected industry journal DigiTimes reports that AMD is likely to reach its lowest ever market position for Q1 2016. Thankfully, the financial results will emerge on April 21st so we should know the full picture relatively soon. Of course, the situation should improve once Polaris and Zen reach retail channels. Most importantly, AMD’s share price has declined by more than 67% in five years from $9 to under $3 as of March 28, 2016. The question is why?

Is the Hardware Competitive?


The current situation is rather baffling considering AMD’s extremely competitive product line-up in the graphics segment. For example, the R9 390 is a superb alternative to NVIDIA’s GTX 970 and features 8GB VRAM which provides extra headroom when using virtual reality equipment. The company’s strategy appears to revolves around minor differences in performance between the R9 390 and 390X. This also applied to the R9 290 and 290X due to both products utilizing the Hawaii core. NVIDIA employs a similar tactic with the GTX 970 and GTX 980 but there’s a marked price increase compared to their rivals.

NVIDIA’s ability to cater towards the lower tier demographic has been quite poor because competing GPUs including the 7850 and R9 380X provided a much better price to performance ratio. Not only that, NVIDIA’s decision to deploy ridiculously low video memory amounts on cards like the GTX 960 has the potential to cause headaches in the future. It’s important to remember that the GTX 960 can be acquired with either 2GB or 4GB of video memory. Honestly, they should have simplified the process and produced the higher memory model in a similar fashion to the R9 380X. Once again, AMD continues to offer a very generous amount of VRAM across various product tiers.

Part of the problem revolves around AMD’s sluggish release cycle and reliance on the Graphics Core Next (GCN) 1.1 architecture. This was first introduced way back in 2013 with the Radeon HD 7790. Despite its age, AMD deployed the GCN 1.1 architecture on their revised 390 series and didn’t do themselves any favours when denying accusations about the new line-up being a basic re-branding exercise. Of course, this proved to be the case and some users managed to flash their 290/290X to a 390/390X with a BIOS update. There’s nothing inherently wrong with product rebrands if they can remain competitive in the current market. It’s not exclusive to AMD, and NVIDIA have used similar business strategies on numerous occasions. However, I feel it’s up to AMD to push graphics technology forward and encourage their nearest rival to launch more powerful options.

Another criticism regarding AMD hardware which seems to plague everything they release is the perception that every GPU runs extremely hot. You only have to look on certain websites, social media and various forums to see this is the main source of people’s frustration. Some individuals are even known to produce images showing AMD graphics cards setting ablaze. So is there any truth to these suggestions? Unfortunately, the answer is yes and a pertinent example comes from the R9 290 range. The 290/290X reference models utilized one of the most inefficient cooler designs I’ve ever seen and struggled to keep the GPU core running below 95C under load.

Unbelievably, the core was designed to run at these high thermals and AMD created a more progressive RPM curve to reduce noise. As a result, the GPU could take 10-15 minutes to reach idle temperature levels. The Hawaii temperatures really impacted on the company’s reputation and forged a viewpoint among consumers which I highly doubt will ever disappear. It’s a shame because the upcoming Polaris architecture built on the 14nm FinFET process should exhibit significant efficiency gains and end the concept of high thermals on AMD products. There’s also the idea that AMD GPUs have a noticeably higher TDP than their NVIDIA counterparts. For instance, the R9 390 has a TDP of 275 watts while the GTX 970 only consumes 145 watts. On the other hand, the Fury X utilizes 250 watts compared to the GTX 980Ti’s rating of 275 watts.

Eventually, AMD released a brand new range of graphics cards utilizing the first iteration of high bandwidth memory. Prior to its release, expectations were high and many people expected the Fury X to dethrone NVIDIA’s flagship graphics card. Unfortunately, this didn’t come to fruition and the Fury X fell behind in various benchmarks, although it fared better at high resolutions. The GPU also encountered supply problems and emitted a large whine from the pump on early samples. Asetek even threatened to sue Cooler Master who created the AIO design which could force all Fury X products to be removed from sale.

The rankings alter rather dramatically when the DirectX 12 render is used which suggests AMD products have a clear advantage. Asynchronous Compute is the hot topic right now which in theory allows for greater GPU utilization in supported games. Ashes of the Singularity has implemented this for some time and makes for some very interesting findings. Currently, we’re working on a performance analysis for the game, but I can reveal that there is a huge boost for AMD cards when moving from DirectX11 to DirectX12. Furthermore, there are reports indicating that Pascal might not be able to use asynchronous shaders which makes Polaris and Fiji products more appealing.

Do AMD GPUs Lack Essential Hardware Features?


When selecting graphics hardware, it’s not always about pure performance and some consumers take into account exclusive technologies including TressFX hair before purchasing. At this time, AMD incorporates with their latest products LiquidVR, FreeSync, Vulkan support, HD3D, Frame rate target control, TrueAudio, Virtual Super resolution and more! This is a great selection of hardware features to create a thoroughly enjoyable user-experience. NVIDIA adopts a more secretive attitude towards their own creations and often uses proprietary solutions. The Maxwell architecture has support for Voxel Global Illumination, (VGXI), Multi Frame Sampled Anti-Aliasing (MFAA), Dynamic Super Resolution (DSR), VR Direct and G-Sync. There’s a huge debate about the benefits of G-Sync compared to FreeSync especially when you take into account the pricing difference when opting for a new monitor. Overall, I’d argue that the NVIDIA package is better but there’s nothing really lacking from AMD in this department.

Have The Drivers Improved?


Historically, AMD drivers haven’t been anywhere close to NVIDIA in terms of stability and providing a pleasant user-interface. Back in the old days, AMD or even ATI if we’re going way back, had the potential to cause system lock-ups, software errors and more. A few years ago, I had the misfortune of updating a 7850 to the latest driver and after rebooting, the system’s boot order was corrupt. To be fair, this could be coincidental and have nothing to do with that particular update. On another note, the 290 series was plagued with hardware bugs causing black screens and blue screens of death whilst watching flash videos. To resolve this, you had to disable hardware acceleration and hope that the issues subsided.

The Catalyst Control Center always felt a bit primitive for my tastes although it did implement some neat features such as graphics card overclocking. While it’s easy enough to download a third-party program like MSI Afterburner, some users might prefer to install fewer programs and use the official driver instead.

Not so long ago, AMD appeared to have stalled in releasing drivers for the latest games to properly optimize graphics hardware. On the 9th December 2014, AMD unveiled the Catalyst 14.12 Omega WHQL driver and made it ready for download. In a move which still astounds me, the company decided not to release another WHQL driver for 6 months! Granted, they were working on a huge driver redesign and still produced the odd Beta update. I honestly believe this was very damaging and prevented high-end users from considering the 295×2 or a Crossfire configuration. It’s so important to have a consistent, solid software framework behind the hardware to allow for constant improvements. This is especially the case when using multiple cards which require profiles to achieve proficient GPU scaling.

Crimson’s release was a major turning point for AMD due to the modernized interface and enhanced stability. According to AMD, the software package involves 25 percent more manual test cases and 100 percent more automated test cases compared to AMD Catalyst Omega. Also, the most requested bugs were resolved and they’re using community feedback to quickly apply new fixes. The company hired a dedicated team to reproduce errors which is the first step to providing a more stable experience. Crimson apparently loads ten times faster than its predecessor and includes a new game manager to optimize settings to suit your hardware. It’s possible to set custom resolutions including the refresh rate, which is handy when overclocking your monitor. The clean uninstall utility proactively works to remove any remaining elements of a previous installation such as registry entries, audio files and much more. Honestly, this is such a revolutionary move forward and AMD deserves credit for tackling their weakest elements head on. If you’d like to learn more about Crimson’s functionality, please visit this page.

However, it’s far from perfect and some users initially experienced worse performance with this update. Of course, there’s going to be teething problems whenever a new release occurs but it’s essential for AMD to do everything they can to forge a new reputation about their drivers. Some of you might remember, the furore surrounding the Crimson fan bug which limited the GPU’s fans to 20 percent. Some users even reported that this caused their GPU to overheat and fail. Thankfully, AMD released a fix for this issue but it shouldn’t have occurred in the first place. Once again, it’s hurting their reputation and ability to move on from old preconceptions.

Is GeForce Experience Significantly Better?


In recent times, NVIDIA drivers have been the source of some negative publicity. More specifically, users were advised to ignore the 364.47 WHQL driver and instructed to download the 364.51 beta instead. One user said:

“Driver crashed my windows and going into safe mode I was not able to uninstall and rolling back windows would not work either. I ended up wiping my system to a fresh install of windows. Not very happy here.”

NVIDIA’s Sean Pelletier released a statement at the time which reads:

“An installation issue was found within the 364.47 WHQL driver we posted Monday. That issue was resolved with a new driver (364.51) launched Tuesday. Since we were not able to get WHQL-certification right away, we posted the driver as a Beta.

GeForce Experience has an option to either show WHQL-only drivers or to show all drivers (including Beta). Since 364.51 is currently a Beta, gamers who have GeForce Experience configured to only show WHQL Game Ready drivers will not currently see 364.51

We are expecting the WHQL-certified package for the 364.51 Game Ready driver within the next 24hrs and will replace the Beta version with the WHQL version accordingly. As expected, the WHQL-certified version of 364.51 will show up for all gamers with GeForce Experience.”

As you can see, NVIDIA isn’t immune to driver delivery issues and this was a fairly embarrassing situation. Despite this, it didn’t appear to have a serious effect on people’s confidence in the company or make them re-consider their views of AMD. While there are some disgruntled NVIDIA customers, they’re fairly loyal and distrustful of AMD’s ability to offer better drivers. The GeForce Experience software contains a wide range of fantastic inclusions such as ShadowPlay, GameStream, Game Optimization and more. After a driver update, the software can feel a bit unresponsive and takes some time to close. Furthermore, some people dislike the notion of GameReady drivers being locked in the GeForce Experience Software.  If a report from PC World is correct, consumers might have to supply an e-mail address just to update their drivers through the application.

Before coming to a conclusion, I want to reiterate that my allegiances don’t lie with either company and the intention was to create a balanced viewpoint. I believe AMD’s previous failures are impacting on the company’s current product range and it’s extremely difficult to shift people’s perceptions about the company’s drivers. While Crimson is much better than CCC, it’s been the main cause of a horrendous fan bug resulting in a PR disaster for AMD.

On balance, it’s clear AMD’s decision to separate the Radeon group and CPU line was the right thing to do. Also, with Polaris around the corner and more games utilizing DirectX 12, AMD could improve their market share by an exponential amount. Although, from my experience, many users are prepared to deal with slightly worse performance just to invest in an NVIDIA product. Therefore, AMD has to encourage long-term NVIDIA fans to switch with reliable driver updates on a consistent basis. AMD products are not lacking in features or power, it’s all about drivers! NVIDIA will always counteract AMD releases with products exhibiting similar performance numbers. In my personal opinion, AMD drivers are now on par with NVIDIA and it’s a shame that they appear to be receiving unwarranted criticism. Don’t get me wrong, the fan bug is simply inexcusable and going to haunt AMD for some time. I predict that despite the company’s best efforts, the stereotypical view of AMD drivers will not subside. This is a crying shame because they are trying to improve things and release updates on a significantly lower budget than their rivals.

Gigabyte Joins 75W Party With Latest GTX 950

Even with pascal so close, Nvidia and their partners are still pushing out new Maxwell based cards. Over the course of the last month or so, ASUS, MSI, and EVGA have all come out with new 75W TDP GTX 950 graphics cards. These new cards delete the PCIe 6 pin power connector, making them a more flexible option. Today, Gigabyte has quietly joined the party with their own 75W GTX 950 offering.

The GIGABYTE GV-N950D5-2GD graphics card is based on the same old GTX 950 GM206 core with 768 stream processors, 48 texture units, 32 ROPs,  and a 128-bit GDDR5 memory interface. Somehow, Nvidia has been able to reduce the TDP from 90W to 75W without reducing base performance. Unlike the versions with power, the new card features a tamer factory overclock, with only 1051 Mhz base and 1228 MHz boost clocks respectively. The cooler is a relatively standard slot with 1 fan. Without the need for a PCIe power connector, it allows the card to be used to upgrade systems that feature a weaker PSU. The value of such a card is diminished though with Pascal being so close. In my view this suggests that Nvidia may leave the low end untouched for a while, with only GP104 launching anytime soon. This may leave AMD with the Polaris 11 relatively unchallenged, only doing battle with older Maxwell cards.

Nvidia May Launch 3 Pascal GP104 SKUs at Computex

For those of you hoping for massive performance jump with the launch of Pascal, prepare to be disappointed. Every new generation tends to improve performance but some more than others. According to previous rumours, Nvidia is using their GP104 die to replace the GTX 980Ti with the GTX 1080 and 1070. Now, the latest reports are suggesting that Nvidia will launch 3 different Pascal SKUs, all based off of GP104, at Computex.

As the xx4 die, GP104 has traditionally been viewed as the smaller chip to the larger x10 or x00 dies that traditionally power flagships. Due to this, don’t expect Pascal to surpass the 980Ti by any large amount. Today’s news also furthers that impression. By splitting GP104 into 3 SKUs, we can expect performance between the 3 cards to be pretty close. It wouldn’t make sense to have so many close performing cards to the flagship which suggests that GP104 won’t be real flagship material.

By slipping GP104 into 3 SKUs, we will likely run into the same situation as the GTX 560Ti 448/570/580 and the 660Ti/670/680. If we take our past experience with those cards as the guideline, we can expect differentiation, not just on the core but the memory bandwidth as well. This makes the previous rumours about the GTX 1070 using GDDR5 while the GTX 1080 will use faster GDDR5X. The 1060Ti as I am calling it may feature either a gimped 192bit bus or the same situation faced by the GTX 970 with a section of VRAM being slower.

Right now, all we have to differentiate the 3 SKUs are the model numbers, the GTX 1080 will be GP104-400-A1, the GTX 1070 GP104-200-A1, and the 1060Ti will be using the GP104-150-A1. It will be interesting to see how Nvidia will differentiate the cards and how they compete against current Maxwell models. Computex can’t come soon enough!

Nvidia Sending Out Invites to Pascal Event

If there is any doubt in our minds that real Pascal cards are coming soon, this latest report clears that all out. According to PCTuning, Nvidia has started sending out invites to select press outlets for a Pascal event. This suggests that Nvidia already has all of their Pascal chips that are launching at Computex ready for the press to review and test out. As of right now, we still don’t know the exact date of Pascal’s launch.

Right now, whether or not the event will be a private event for press only or a live webcast sort of affair remains to be seen. This may well be the event Nvidia sometimes holds with the press beforehand in order to brief them on the chip a bit before shipping out review samples.

What’s more, the leaker is suggesting that Nvidia is going to allow reviews and other information to be released in as soon as three weeks and four weeks at the latest. This would place it at least 2 weeks ahead of Computex which has been the suggested launch. The possibilities are that the leaker is wrong, Computex will be the hard launch with a soft launch before; either way, Pascal is coming soon.

NVIDIA DRIVE PX2 Powered by Two GP106 Chips

NVIDIA showed of its DRIVE PX 2 system – the new iteration of its autonomous and driver assistance AI platform – at last week’s GTC 2016 conference, and eagle-eyed viewers may have noticed that the board shown to the audience by CEO Jen-Hsun Huang was sporting a pair of integrated GP106 GPUs, eschewing the two Maxwell-based NVIDIA Tegra X1 chips that powered the original DRIVE PX, and confirming a rumour that we reported last week.

The GP106 runs on NVIDIA’s new Pascal architecture – set to hit the market in the latest line of GeForce graphics cards this Summer – which can perform at 24 DL TOPS or 8 TFLOPS, and features up to 4GB GDDR5.

NVIDIA hopes that the new DRIVE PX 2 will power the next generation of driverless cars – the DRIVE PX has so far only be used to power the ‘infotainment’ system on-board a Tesla, for example – and has already shipped to a number of unnamed Tier 1 customers.

https://www.youtube.com/watch?v=KnVVJSIiKpY

“DRIVE PX platforms are built around deep learning and include a powerful framework (Caffe) to run DNN models designed and trained on NVIDIA DIGITS,” according to NVIDIA. “DRIVE PX also includes an advanced computer vision (CV) library and primitives. Together, these technologies deliver an impressive combination of detection and tracking.”

First Pascal GPU’s to be Unveiled at Computex 2016?

Some of the first cards to run utilizing the all new “Pascal” architecture made by Nvidia, may debut at Computex 2016. The show is going to be in Late may / Early June in Taipei and is one of the biggest ICT shows in the world and you can be sure the eTeknix team will be there to catch the latest news from the event!

Mass shipments should start sometime in July according to Digitimes, the Taiwan based industry observer. With Nvidia unveiling the new cards via its add-in card (AIC) partners, with large manufacturers such as ASUS, MSI, and GIGABYTE being at the event.

The new GPU will be based on the GP104 chip and utilize GDDR5X VRAM; a whopping 8GB is rumored to be the amount. The leaked specs show it having a single eight-pin power connector, meaning that (due to electrical capacity) the max power usage would be 225W, though it could use a lot less power. The 980 is only 165W so this card can’t be a huge amount more. The leaked specs also tell us that it could feature up to 6144 CUDA cores and a whopping 12.6 Teraflops. We’re not sure how accurate these specs are as they have been sourced from various places, only time will tell. Either way, Computex 2016 is going to be huge this year.

So far, we can accurately say:

  • 2x performance per watt estimated improvement over Maxwell
  • DirectX 12_1 or higher
  • Successor to the GM200 GPU in the 980TI
  • Built on 16nm manufacturing process

It will be interesting to see the Polaris release too, as there is going to be some very tough competition on the GPU market shortly and that’s obviously great news for consumers.

Which cards are you most excited about this year, AMD’s or Nvidia’s latest? Let us know in the comments section below.

Nvidia Pascal GP104 May Use GDDR5 and GDDR5X

With rumours pointing to a May/June launch at Computex, more information is coming about Nvidia’s upcoming GTX 1070 and 1080. At times taking on an x70 and x80 moniker, the two chips are slated to use the GP104 Pascal die and take up the role currently filled by the GTX 970 and 980. Today, another leak has come out detailing what the two cards will look like and it seems the GTX 1080 will have a lot more memory bandwidth than the GTX 1070.

According to the leak, both the GTX 1080 and GTX 1070 will be based on the GP104 die. This will slot into the Pascal lineup just like the GK104 and GM204. While previous cards have mostly differentiated in the core specifications, it looks like this time, memory bandwidth and a lot of it will be the difference. The GTX 1080 will reportedly use GDDR5X while the 1070 will use GDDR5. This should give potentially up to 100% more bandwidth for the GTX 1080 and better energy efficiency to boot. Due to the use of different memory, the 1080 will boast 20 more pins. The 1080 will use the GP104-400-A1 and the 1070 the GP104-200-A1.

For two similar cards based of off the same GP104, there seems little reason to split the memory between two different types since the required memory bandwidth should be similar. One possibility is that the GTX 1080 may be using relatively slower GDDR5X that isn’t much faster than the fastest GDDR5. This makes sense if GDDR5X is supply constrained for the top end models. Another possibility is that both were meant to get GDDR5X but supply meant only one of them could. Finally, it could also be a way to differentiate the GTX 1080 as it may remain the GeForce flagship for quite a while.

Last of all, we have also gotten information about the connectors, Both cards will feature 3x Displayport, 1 HDMI, and 1 DVI for display connectivity. Power will be provided by 2 x 8pin PCIe power connectors which is actually more than what the Titan X has. Given the efficiency from moving to 16nm, this points to either a monster chip or lots of dual precision hardware being left in. With only several weeks left, will be interesting to finally see what Nvidia has cooked up for us.

Is Nvidia GTC Drive PX 2 Pascal GP106?

All throughout this week, we’ve been treated to leak after leak and release after release of new Polaris and Pascal information. After we’ve gotten die shots of  GP100 and GP104, what may be another Pascal chip has surfaced. According to Videocardz, the Drive PX 2 module Jen-Hsun Huang waved in front of our faces might actually be Pascal this time around. More specifically, it might be GP106 or GP107.

Nvidia took some heat earlier in the year when CEO Jen-Hsun Huang took to the stage and waved a Drive PX 2 around that was Maxwell. This time around though, it looks like Pascal has come far enough along to make it into engineering samples. Based off of size comparisons to GM206, the Drive PX 2 GPU is too small in comparison. This means it is either the budget GP106 or the smaller mobile GP107.

This stems from the fact that the die is really similar to GM206 and given what we know of GP104 and GP100, the GP106 and GP107 should be of a similar size to GM207 and GM206. I believe that given the MXM format, we may be seeing the GP107 as that chip is meant for the mobile market. Using a chip that hits both the lucrative laptop and car markets would make a lot of sense. Either way, budget Pascal may be coming sooner than expected.

Nvidia Pascal GP104 Spotted?

So far all of the rumours around the GP104 and GTX 1000 series have mostly been about release date and specifications. The closest we’ve gotten to physical evidence have been the shrouds for the GTX 1080 and GTX 1070. For the first time, we’re getting a picture of the physical die and parts of the GPU board around it. According to ChipHell, the die shot you see below belongs to the GP104, the mainstream Pascal GPU.

From the die shot picture, GP104 appears to be about 15.35mm x 19.18mm for a total of about 290 to 300mm². This is the same as GK104 which was also a die shrink and came in at 294mm² and much smaller than GM204 which was a relatively massive 398mm². This shows that Nvidia is starting out with small dies first with the GTX 1070 and 1080 and releasing a GP100/102 Titan and 980Ti later on.

For now, we still don’t know what GP104 will look like, but it seems that most of the FP64 units in GP100 will likely be stripped out and replaced by the more ‘useful’ FP32 ones. The leak also suggests that total FP32 CUDA core count will be around the same as the Titan X but the TMU and ROP count seems closer to the GTX 980. I expect that clock for clock, GP104 won’t be much faster than the Titan X but it will be ahead and much more efficient.

Finally, we can see what appears to be Samsung 2Ghz 1GB GDDR5 DRAM modules for 8GB total. This suggests that that GDDR5X isn’t ready in time or will be reserved for the GP100/102 consumer release. This follows the same trend set by the GTX 680 which was more powerful than the 580 but featured lower memory bus width but faster VRAM and more memory overall.

While the leak is promising, it is a leak after all and I would make sure to take all of this with a shipload of salt. Given the information we know though, this leak may very well reflect reality.

Nvidia Pascal GTX 1000 Will Arrive at Computex

From the many leaks and rumours that have come out, the expected release of Pascal will come later this year at Computex. During the Taiwanese event, Nvidia will finally unveil the GTX 1000 lineup to the public. Today, we’re getting yet another report confirming this. In addition, Nvidia AiB partners like ASUS, Gigabyte and MSI will showcase their reference cards then as well. As revealed yesterday, mass shipments won’t begin till July.

As we’ve covered earlier, the sources appear to suggest that Nvidia will have a head start over AMD in launching Pascal ahead of Polaris. However, the lead might not amount to much. The report suggests that Nvidia will only ship large amounts till July. AMD on the other hand is also launching Polaris in June, the same month as Pascal. Given AMD’s previous history, we will probably see Polaris cards out by July as well. If Nvidia does have a lead, it won’t be for very long.

Right now, there is no word yet if GDDR5X will be utilized for top end Pascal chips. While there are some reports out that suggest GDDR5X, the timeline is very tight as GDDR5X probably won’t reach enough capacity till May/June at the earliest. Perhaps this is why we won’t be seeing Pascal or Polaris in numbers till July.

NVIDIA Pascal to Enter Mass Shipments in July

Yesterday, we reported on AMD’s plans to supposedly launch their next graphics architecture, codenamed ‘Polaris’ in June. The details surrounding NVIDIA’s response with Pascal aimed at consumers was unknown but it seemed likely the range would arrive at a similar date. According to new information sourced by Digitimes, Pascal will be unveiled during Computex and enter mass shipments in July:

“The graphics card players will begin mass shipping their Pascal graphics cards in July and they expect the new-generation graphics card to increase their shipments and profits in the third quarter.”

Interestingly, the report claims that AMD might only unveil the Polaris range in June, and shipments could occur at a later date. Apparently, NVIDIA will have the advantage and be the first company to release their new line-up to the public:

“Meanwhile, AMD has prepared Polaris-based GPUs to compete against Nvidia’s Pascal; however, the GPUs will be released later than Nvidia’s Pascal and therefore the graphics card players’ third-quarter performance will mainly be driven by demand for their Nvidia products.”

Thankfully, both graphics card manufacturers look set to release brand new products and I cannot wait to see the performance improvements and pricing across various tiers. AMD appears to be focussing on performance per watt on Polaris and the demonstrations thus far have been impressive. Sadly, we haven’t really seen anything from NVIDIA’s new consumer line-up, so it will be fascinating when the samples finally arrive. It’s still unknown which products will opt for HBM2, if any. It’s clear that the majority of models from both companies are set to utilize GDDR5X. While this isn’t a patch on HBM2, it’s a good improvement from the older GDDR5 standard.

Recently, there were some murmurings about NVIDIA delaying mainstream Pascal under 2017. This doesn’t look like the case at all, and if anything, reports suggest they will be the first to market.

EVGA Announces Low Power Nvidia GTX 950

Even with Pascal just around the corner, Nvidia and their partners are continuing to release some new Maxwell GPUs. Today, we’ve been treated to a new lineup of Nvidia’s GTX 950 from EVGA. The biggest and one of the few differences between the new GTX 950’s and the old ones are the reduction of TDP from 90W to 75W. In all other regards, the cards are just the same old GTX 950 we’ve come to expect.

Like the ASUS and MSI cards from before, these new GTX 950’s are somehow able to reduce their power consumption from 90W down to 75W while using the same die. Some of the 75W TDP models retain the 6pin PCIe power connector, allowing them to reach 1127Mhz base and 1317Mhz boost clock. For the ones that delete the 6pin PCIe power connector, they peak at a lower 1075Mhz base and 1253 boost.

For the segment of the market the GTX 950 targets, any reduction of power consumption is important. Not everyone has a PSU with a 6pin connector and removing it makes sense if you can keep the same performance characteristics. There is a total of 8 new cards EVGA has launched today and they are all available now.

Take a Look at the New Nvidia Pascal Architecture

With the reveal of the Tesla P100, Nvidia has taken the wraps off of their new Pascal architecture. Originally set to debut last year, delays with 16nm kept Pascal from being a reality, leading to Maxwell on 28nm. Now that Pascal is finally here, we are getting an architecture that combines the gaming abilities of Maxwell with much improved compute performance. The new Unified Memory and Compute Pre-Emption are the main highlights.

First off, Pascal changes the SM (Stream Multiprocessor) configuration yet again. Kepler featured 192 CUDA cores per SM, Maxwell had 128 and Pascal will now have 64. By reducing the number of CUDA cores per SM, it increases the fine grain control over compute tasks and ensure higher efficiency. Interestingly, 64 is also the same amount of cores GCN has in each CU, AMD’s equivalent to SM. The TMU to CUDA core ratio remains the same as Maxwell with 4 per SM instead of 8, in line with the drop in cores/SM.

For compute, the gains mostly come from increasing the number of FP64 or Dual Precision CUDA cores. DP is important for scientific and compute workloads though game rarely make use of them. Kepler started cutting out some FP64 units and Maxwell went even further, with virtually no FP64 even in the Tesla’s. This was one reason why Maxwell cards were so efficient and Nvidia only managed to hold onto their leadership in compute due to CUDA and their Single Precision performance.

With Pascal, the ratio of SP to DP units goes to 2:1, significantly higher than the 32:1 of Maxwell and 3:1 of Kepler. GP100 in particular has about 50% of its die space dedicated to FP32, about 25% to DP and the last 25% split between LD/ST and SFUs. This suggests that Pascal won’t be changing much in terms of gaming performance. The only gains will be from a slight increase in efficiency due to the smaller SMs and the die shrinking from 16nmFF+. GeForce variants of Pascal may have their FP64 units trimmed to cram in more FP32 resources but again, most of the gains will be due to increased density.

Lastly, Pascal brings forward unified memory to allow threads to better share information. This comes along with improved L2 cache sizes and the more than double register file sizes. P100, the first Pascal chip, also uses HBM2, with 16GB of VRAM over a 4096bit bus for a peak bandwidth of  720 GB/s. For CUDA compute tasks, a new Unified Memory model allows Pascal GPUs to utilize the entire system memory pool with global coherency. This is one way to tackle AMD’s advancement with HSA and GCN and Intel’s Xeon Phi’s.

Overall, Pascal looks to be an evolutionary update for Nvidia. Perhaps, Nvidia has reached the point that Intel has, making incremental progress. In other ways though, the reduction in SM size has great potential and provides a more flexible framework to build GPUs. Now all we are waiting for is for the chips to finally drop.

Nvidia Details Tesla P100 Pascal GPU Specifications

After revealing their next flagship Telsa earlier, Nvidia has let loose with a few more details and specifications. Based on the new Pascal architecture, the P100 will be utilizing TSMC’s latest 16nmFF+ process. As we know from the keynote, the chip will feature 15.3 billion transistors and the latest HBM2 memory. The P100 also features what Nvidia is calling the “5 miracles”.

First off, the P100 will run at an impressive 1328 MHz base clock and 1480 MHz boost. This is high for a professional Tesla card though well in line with GeForce clocks. The card won’t be using the full GP100 die with 60 SMs and 3840 CUDA cores, rather it will use a cut-down version with 56 SMs with 3584 cores. This mirrors Kepler’s launch where the cut-down Titan came before the Titan Black. In addition to the usual FP32 CUDA cores, there are also 1792 FP64 CUDA cores for Dual Precision Work. This gives a 2SP/1DP ratio, higher than anything from Kepler or Maxwell. The P100 also has 224 TMUs and massive amounts of cache and register files.

Next, we have the massive 610 mm² die on 16nmFF+. About 50% of that is FP32 CUDA cores, 25% is FP64 and rest on other parts. This means despite the massive die size, the P100 and GP100 derivatives won’t be great gamers, as games generally only use FP32 CUDA cores. There may be a GP100 variant though that swaps out the FP64 cores for FP32 ones. Even saddled with compute though, GP100 will still beat the Titan X by a good margin. TDP is a relatively tame 300W, as expected from the use of 16nm and 16GB of HBM2.

Finally, most marketing statements are hyperbole and the “5 miracles” are no exception. They are the Pascal Architecture, 16nm FinFET, CoWoS with HBM2, NVLink, and New AI Algorithms. Honestly, none of these are really that amazing on their own and have been expecting. Combining all of them in one go on such a massive chip though is pretty amazing though. While the P100 will be shipping soon, don’t expect many till Q1 2017.

NVIDIA Unveil Tesla P100 GPU

GTC 2016: As part of NVIDIA’s GPU Tech Conference, Jen-Hsun unveiled the latest product in the Tesla family with the P100. Branded as the most advanced hyperscale datacentre GPU, it features 150 billion transistors and is based on the latest Pascal architecture.

Built on a 16nm FinFET process and featuring HBM2, this product is the latest in a whole host of new technologies from NVIDIA and should be the start of what we’re going to see across other NVIDIA products.

With AI and deep learning now at the forefront of NVIDIA’s thinking, the Tesla P100 GPU has been created to assist with making AI and deep learning among other tasks as fast as physically possible.

Steve Wozniak Experience Mars Through VR

GTC 2016: As a special guest experience, NVIDIA enlisted Apple co-creator Steve Wozniak to take the helm in the Mars Rover.

As an person who has made it clear in the past of wanting to sign up for the one-way ticket to Mars, NVIDIA have made it possible to experience the same thing without leaving the comfort of your own couch. ‘Woz’ as part of the experience chucked a VR headset on and showed us how real the overall feel of being on Mars was.

Jen-Hsun and Wozniak joke about finally finding Matt Damon but you straight away get the feel as to how immersive the experience of Mars can be.

This was made possible from utilising the GeForce GTX Titan to give the best possible quality taking into consideration how realism is of the keypoint and this leads to IRAY VR technology.

IRAY VR utilises a pre-rendered source of light probes that is then rasterized and reconstructs the image based on what the eye expects to see as a completed composition.