Showing posts with label Graphics Card. Show all posts
Showing posts with label Graphics Card. Show all posts

Wednesday, 3 August 2016

NVIDIA Titan X: The Ultimate Graphics Card Unleashed – Features Pascal GP102 GPU, 12 GB G5X Memory and a $1200 US Price

Today, NVIDIA launches their $1200 US, Titan X graphics card aiming the enthusiast and professional market. Do not mistake this card with last year’s GeForce GTX Titan X which featured the Maxwell GPU as the latest Titan X adopts the Pascal GPU architecture. Already having the fastest gaming product in the market, the Titan X would further strengthen NVIDIA’s lead in the high-end GPU sector.
NVIDIA Titan X has been launched, the fastest Pascal graphics card now available at $1200 US.

NVIDIA Titan X Unleashed – Features Brand New Pascal GP102 GPU With 12 GB G5X Memory and 11 TFLOPs of Compute Crunching Power

The NVIDIA Titan X is one of many things, NVIDIA calls it the ultimate graphics card but it’s not just any Ultimate graphics card. The Titan X is the fastest gaming card available, its the fastest deep learning card and its also the fastest professional card available at its price. The NVIDIA Titan series is mostly branded as a prosumer series as they are targeted at consumers and professionals. This is one of the reasons NVIDIA has dropped the GeForce GTX branding from the Titan X.

NVIDIA Titan X Review Round-Up:


Recently, AMD had also announced their own take at the prosumer market with the Radeon Pro Duo, a dual Fiji based, $1500 US solution. While Pro Duo is meant for enthusiast gamers and content creators, the NVIDIA Titan X aims the enthusiast gaming, deep learning and professional content creation markets. The latest Pascal chip significantly helps deep learning instructions throughput with its multiple TFLOPs of compute processing power.

NVIDIA Titan X Specifications in Detail – The First GP102 Card

The Pascal GP102 GPU features 12 Billion transistors, a massive increase over 7.2 Billion transistors on GP104 and lower than the 15.4 Billion transistors on GP100 die. The chip housing a lower transistor count comes with a smaller die size of 471mm2 compared to 610mm2 on the GP100 die. The GP102 rids of the dual precision FP 64 hardware featured on the GP100. The GPU does get the same 3584 CUDA cores as featured on the NVIDIA Tesla P100 GPU which are clocked 1.41 GHz core and 1.53 GHz boost clocks.
NVIDIA Titan X Graphics Card Official
The compute numbers of these cards are a leap beyond what we have seen on any consumer card with the GPU touching 11 TFLOPs of single precision and 44 TOPs INT8 calculations (new deep learning instructions). NVIDIA’s GP104 GPU comes very close to the Titan X with its 9 TFLOPs of compute power and around 10 TFLOPs with OC. The Titan X is said to boost around 1.8 GHz in majority of workloads so 12 TFLOPs on this card wouldn’t be hard to hit at stock configs.
NVIDIA Titan X Launch
The NVIDIA Titan X features 12 GB of GDDR5X which is clocked at 10 GB/s and delivers 480 GB/s which is just about the same amount that HBM1 delivered (512 GB/s). We can see that HBM2 is still not that widely available for use in consumer grade cards but GDDR5X can cover it up with enough bandwidth to feed the GPU. NVIDIA is using G5X memory on all of their GP102 GPUs with the Quadro P6000 featuring up to 24 GB G5X memory. The Tesla P100 on the other hand houses 16 GB of HBM2 memory.

NVIDIA Titan X PCB – Premium Components and Design For Fastest Pascal GPU

The NVIDIA Titan X PCB has the highest quality components with 7+1 Phase Dual-FET design that deliver unprecedented power to the GPU core. The GP102 GPU is surrounded by 12 GDDR5X memory chips that deliver 480 GB/s bandwidth. The side of the card has two SLI-HB gold fingers that allow up to Quad-Way interface.

NVIDIA Titan X With GP102 PCB:

NVIDIA Titan X_Official_PCB

NVIDIA GTX 1080 With GP104 PCB:NVIDIA GeForce GTX 1080 PCB


NVIDIA GeForce 10 Pascal Family:

Graphics Card NameNVIDIA GeForce GTX 1060 6 GBNVIDIA GeForce GTX 1070NVIDIA GeForce GTX 1080NVIDIA Titan X
Graphics CoreGP106GP104GP104GP102
Process Node16nm FinFET16nm FinFET16nm FinFET16nm FinFET
Die Size200mm2314mm2314mm2471mm2
Transistors4.4 Billion7.2 Billion7.2 Billion12 Billion
CUDA Cores1280 CUDA Cores1920 CUDA Cores2560 CUDA Cores3584 CUDA Cores
Base Clock1506 MHz1506 MHz1607 MHz1417 MHz
Boost Clock1708 MHz1683 MHz1733 MHz1530 MHz
FP32 Compute4.4 TFLOPs6.5 TFLOPs9.0 TFLOPs11 TFLOPs
VRAM6 GB GDDR58 GB GDDR58 GB GDDR5X12 GB GDDR5X
Bus Interface192-bit bus256-bit bus256-bit bus384-bit bus
Power ConnectorSingle 6-Pin PowerSingle 8-Pin PowerSingle 8-Pin Power8+6 Pin Power
TDP120W150W180W250W
Display Outputs3x Display Port 1.4
1x HDMI 2.0b
1x DVI
3x Display Port 1.4
1x HDMI 2.0b
1x DVI
3x Display Port 1.4
1x HDMI 2.0b
1x DVI
3x Display Port 1.4
1x HDMI 2.0b
1x DVI
Launch Date13th July 201610th June 201627th May 20162nd August 2016
Launch Price$249 US$379 US$599 US$1200 US


Read more: http://wccftech.com/nvidia-titan-x-pascal-unleashed/#ixzz4GHQAYHOe

NVIDIA Titan X Performance Metrics and Efficiency Numbers – Going Through The Sky!

When it comes to efficiency, the GP104 architecture was already leaps ahead of anything that came before it. Maxwell was great when it came to efficiency but Pascal is a whole new level for NVIDIA. NVIDIA also claimed up to 70% better performance than the GTX Titan X on the latest GP102 card. That seems to be accurate as the card is up to 35% faster than the GeForce GTX 1080 (30% faster Than Titan X) which utilizes the GP104 core.

Ryan Shrout – PCPerspective:

One area that I think needs some attention – AMD’s lack of competition on the high end is starting to get ridiculous. In every game we tested, except Hitman, the Titan X is 70-120% faster than the fastest single GPU AMD graphics card, the AMD Fury X. Obviously, there is a process technology gap, a cost gap, and a timing gap – but AMD is falling not just slightly behind, but PAINFULLY behind NVIDIA when it comes to flagship performance. The Radeon RX 480 is a great card and gives AMD a competitive option at the $250 price point but there are plenty of gamers buying at higher prices, where margins are fattening NVIDIA up to do this battle again in 12-18 months.
At the end of the day (and I am 9 minutes from that as I type this), the new NVIDIA Titan X based on the Pascal GP102 GPU is the fastest graphics card on the market, period. If you want the best, and have the wallet to support your addiction, you can’t get anything better than this.

Gordon Mah Ung – PCWorld:

Most people don’t “get” the point of a $189,000 car or a tin of caviar that costs more than a dinner for four, and well, most people probably won’t “get” the point of a pair of powerful new Titan X cards either—unless they’re data scientists. When it comes to gaming, a glorious pair of SLI’d new-look Titan X cards are made for high rollers with bleeding-edge displays alone.

SKYMTL – HardwareCanucks:

With all of this being said, I am going to sit back and look at the TITAN X for what it is: an expensive technological tour de force which thumbs its nose at withholding performance for the sake of price. It simply says “I’m doing it my way”, then drops the mic and walks away.

NVIDIA Titan X Official Renders:

NVIDIA Titan X Founders Edition – Black Die-Cast Aluminum Shroud With Premium Design

The NVIDIA Titan X is a beautiful card that features the same NVTTM cooler found on other Pascal based cards. The NVIDIA Titan X comes with a die-cast aluminum shroud with a metallic black finish. Housed beneath this shroud is a large vapor chamber cooler and additional heatsink based cooling fins that are provided air through a blower fan that pushes heat out of the card. The card bears the Titan X logo on the front and surprisingly has no GeForce GTX logo on the sides which was the case with previous Titan cards.
NVIDIA Titan X Front
The card features a high-end PCB design that should allow the card to reach clock speeds close to 2 GHz. The power is provided through an 8+6 Pin configuration and NVIDIA officially lists the TDP at 250W. Display output include three DP 1.4, one HDMI 2.0b and a single DVI-D connector.
The NVIDIA Titan X is currently the fastest single-chip graphics card that solidifies NVIDIA’s lead in the high-end and professional graphics department. Based on their GP102 GPU, the card offers insane amount of graphics performance but as with all enthusiast cards, it comes with a high price of $1200 US. We can expect NVIDIA to launch a more consumer and gamer focused GP102 GPU later this year for a reasonable price.

Monday, 9 May 2016

Nvidia: The Geforce GTX 1080 Graphics Card Can Do Asynchronous Compute

The Async Compute problem is probably one of the most controversial issues surrounding the older generation of Geforce graphics cards from Nvidia. Something very interesting, however, is present in the press release that they sent out to, well, the press. According to the official statement, the GTX 1080 is fully capable of performing Async Compute. If this turns out to be true, then this will give negate a major edge that Radeon graphics cards from AMD have enjoyed this past year.

Nvidia: GTX 1080 is capable of Async Compute

Keep in mind however, that even Maxwell featured Asynchronous Compute on paper. Unfortunately, due to the fact that expensive software based context switching had to be employed before it could be used, (since it did not have a dedicated hardware scheduler like AMD’s GCN) resulted in lowered performance on Maxwell based graphics cards. Nvidia’s style had been a technique called preemption, which it had perfected to an impressive degree. The reviews are going to be out on May 27th and if independent reviews confirm this fact, than it will be a huge win for the green camp. 
AMD_DirectX 12_Asynchronus Compute 2
Asynchronous compute has been a deal sweetener for Radeon buyers ever since the DirectX 12 API hit the stage. AMD currently leads in Hitman and AOTS which utilize their Asynchronous shader technology developed around DirectX 12 API. Interestingly, Nvidia GPUs historically perform much better without ASync turned on. This is probably due to the fact that Nvidia had apparently disabled ASync from their driver suite. The rationale given for that move is that its GPUs cannot process ASync concurrently on the hardware level, rather they need context switching which is expensive in terms of frame rate.

The Async Compute Story Distilled Down To Its Core

Async Compute has been a hot subject of debate ever since gamers became aware of its very existence. We dove deep a couple of months ago into this peculiar DirectX 12 feature in our two thousand word analysis piece dubbed “AMD’s Secret DirectX 12 Weapon That Nvidia Had To Trade Off – Demystifying Async Compute“. We explained the inherent architectural differences between Nvidia and AMD graphics cards and distilled the key reasons as to why they deal and perform so differently with asynchronous game code. We’d highly recommend giving it a read if you’re looking to wrap your head around this topic and get down to the core of the issue before proceeding.
Five Marvels of Pascal: NVIDIA engineered the Pascal architecture to handle the massive computing demands of technologies like VR. It incorporates five transformational technologies:
  • Next-Gen GPU Architecture. Pascal is optimized for performance per watt. The GTX 1080 is 3x more power efficient than the Maxwell Architecture.
  • 16nm FinFET Process. The GTX 1080 is the first gaming GPUs designed for the 16nm FinFET process, which uses smaller, faster transistors that can be packed together more densely. Its 7.2 billion transistors deliver a dramatic increase in performance and efficiency.
  • Advanced Memory. Pascal-based GPUs are the first to harness the power of 8GB of Micron’s GDDR5X memory. The 256-bit memory interface runs at 10Gb/sec., helping to drive 1.7x higher effective memory bandwidth than that delivered by regular GDDR5.
  • Superb Craftsmanship. Increases in bandwidth and power efficiency allow the GTX 1080 to run at clock speeds never before possible — over 1700 MHz — while consuming only 180 watts of power. “New asynchronous compute advances improve efficiency and gaming performance.” And new GPU Boost™ 3 technology supports advanced overclocking functionality.
  • Groundbreaking Gaming Technology. NVIDIA is changing the face of gaming from development to play to sharing. New NVIDIA VRWorks™ software features let game developers bring unprecedented immersiveness to gaming environments. NVIDIA’s Ansel™ technology lets gamers share their gaming experiences and explore gaming worlds in new ways.
Async Compute on the GTX 1080 will allow developers to execute some tasks that would otherwise be allocated to the CPU, on the GPU. This means that if a game is being CPU-bound (that is to say the CPU is the bottlenec present), it will drastically increase frame rates. It may even improve performance in games that are GPU bound, by allowing full use of the GPUs resources. The thing we have to keep in mind however that Preemption and Asynchronous compute are both different approaches to achieve the same end result: maximizing the utilization of a GPU. And while AMD will have you believe Async is drastically superior choice, badly implemented Async will fare much worse than properly implemented preemption.
Due to the pressure exerted by the industry to make Async compatible graphics cards however, Nvidia has been working actively to implement Async in their GPUs but were held back due to the fact that this was something that had to be implemented at a hardware level. Chip design usually takes a lot of time (in the lieu of many years) and if Nvidia has actually managed to properly implement Async in the Pascal based GTX 1080 – that would be quite an accomplishment. Something their CEO stated a few weeks back (regarding the P100 being capable of advanced preemption) made us think that Pascal might stick with the Preemption approach for now, but the press release from Nvidia states otherwise. So consider us pleasantly surprised!