Gráfica nVidia Pascal (Mobile, TITAN Xp, GTX 1000 series)

Estarei eu a entender mal, a Nvidia diz que esta nova arquitectura é 10 vezes mais rápida. Isto não significa que tenha 10 vezes mais performance, certo ?
 
10x mais rápida num caso especifico de um benchmark interno qualquer, provavelmente em machine learning, redes neurais ou algo assim que é o que eles mais têm mostrado.
 
http://wccftech.com/nvidia-pascal-g...hip-single-chip-card-feature-16-gb-hbm2-vram/

NVIDIA Pascal GPU Will Be Manufactured on TSMC 16nm FF Node – Flagship Single Chip Card To Feature 16 GB HBM2 VRAM
HARDWARE 5 hours ago by Hassan Mujtaba
NVIDIA’s upcoming Pascal GPU has been confirmed by sources to be manufactured on the TSMC 16nm FinFET node as Samsung has failed to win the contract for the latest GPU. The NVIDIA Pascal GPU is the latest NVIDIA Compute oriented graphics architecture which will be launching in 2016 and feature new technologies such as HBM2, NVLINK and Mixed Precision support.



TSMC Wins Contract To Manufacturer Next Generation NVIDIA Pascal GPU
There’s no specific reason given for NVIDIA choosing TSMC over Samsung and there were reports that NVIDIA would use both semiconductor companies to mass produce Pascal GPUs but at the end of the day, NVIDIA had to choose TSMC and their 16nm FinFET node even though Samsung already has 14nm FinFET in production as was demonstrated by Apple with their A9 SOC demonstrated just a few weeks back.

It was revealed a few weeks ago that NVIDIA’s Pascal GP100 chip has already been taped out on TSMC’s 16nm FinFET process, last month. This means that we can see a launch of these chips as early as Q2 2016. Doubling of the transistor density would put Pascal to somewhere around 16-17 Billion transistors since the Maxwell GPUs already feature 8 Billion transistors on the flagship GM200 GPU core.



TSMC’s 16FF+ (FinFET Plus) technology can provide above 65 percent higher speed, around 2 times the density, or 70 percent less power than its 28HPM technology. Comparing with 20SoC technology, 16FF+ provides extra 40% higher speed and 60% power saving. By leveraging the experience of 20SoC technology, TSMC 16FF+ shares the same metal backend process in order to quickly improve yield and demonstrate process maturity for time-to-market value.

According to industry sources on Sept. 15, Nvidia decided to let TSMC mass produce the Pascal GPU, which is scheduled to be released next year, using the production process of 16-nm FinFETs. Some in the industry predicted that both Samsung and TSMC would mass produce the Pascal GPU, but the U.S. firm chose only the Taiwanese firm in the end. Since the two foundries have different manufacturing process of 16-nm FinFETs, the U.S. tech company selected the world’s largest foundry for product consistency.



Advertisements


The reason for Samsung’s determination to win the contract for the Pascal GPU lies in the fact that Nvidia’s new GPU is highly likely to mark a milestone in the next-gen graphic market. Experts are saying that Samsung’s failure to obtain the contract is mainly attributable to its lack of experience. The fact that the Korean tech giant has become TSMC’s rival only two years after it started to produce GPUs itself is considered to have special meaning at the moment. via Business Korea

Furthermore, NVIDIA has their GTC conference being held in Japan on 18th September 2015 and have already released some material in regards to the new announcements that will be made at the graphics focused event. While not much in details in regards to specific architecture specifications or features, the presentation does show some new details about the NVIDIA Pascal GPU. As we already know, NVIDIA Pascal GPU will feature Mixed Precision, 3D Memory integration and NVLINK. The new details show that the top of the line NVIDIA Pascal (Single chip) card will feature up to 16 GB HBM2 VRAM that will deliver up to 1 TB/s bandwidth. More details about Pascal are expected to be revealed at the conference.

16 GB and up to 32 GB of HBM memory across a massive 4096bit memory interface which will dominate the next high-resolution 4K and 8K gaming panels. With 8Gb per DRAM die and 2 Gbps speed per pin, we get approximately 256 GB/s bandwidth per HBM2 stack. With four stacks in total, we will get 1 TB/s bandwidth on NVIDIA’s GP100 flagship Pascal which is twice compared to the 512 GB/s on AMD’s Fiji cards and three times that of the 980 Ti’s 334GB/s. Samsung and SK Hynix are planning to commence mass production later this year so NVIDIA will have the choice to select from two manufacturers while AMD is expected to stick with SK Hynix as their partner for HBM2 memory that will be featured on their Arctic Islands cards that are also suggested to feature around 16-32 GB HBM2 VRAM. The flagship NVIDIA Pascal GPU will come with up to 16 GB of HBM2 VRAM while the dual chip offerings that will be aimed at GeForce and Tesla markets later on will ship with 32 GB VRAM.

The Pascal GPU would also introduce NVLINK which is the next generation Unified Virtual Memory link with Gen 2.0 Cache coherency features and 5 – 12 times the bandwidth of a regular PCIe connection. This will solve many of the bandwidth issues that high performance GPUs currently face. One of the latest things we learned about NVLINK is that it will allow several GPUs to be connected in parallel in HPC focused platforms that will feature several nodes fitted with Pascal GPUs for compute oriented workloads. The latest NVLINK interconnect path will allow multi-processors featured inside HPC blocks to have faster interconnect than traditional PCI-e Gen3 lanes at 80 GB/s speeds. Pascal GPUs will also feature Unified memory support allowing the CPU and GPU to share the same memory pool and finally we have Mixed precision support.

http://wccftech.com/nvidia-pascal-g...p-card-feature-16-gb-hbm2-vram/#ixzz3lv8UHWrX
 
Pensava que o nvlink ia ser aplicado somente em sistemas com processadores da IBM.

A Intel e a AMD vão pagar o licenciamento e vão alterar as arquitecturas?
 
até acredito que para colocar a coisa no "terreno" a nvidia ofereça o licenciamento.

o que me preocupa é ter de trocar a maquina toda lol.
 
NVIDIA Confirmed To Be Supplied With Second Gen. HBM From Both Samsung And SK Hynix
HardwareIndustry 2 hours ago by Omar Sohail

In addition to SK Hynix, Samsung has also entered the HBM market and would commence mass production of the memory during Q1, 2016. Both South Korean memory manufacturing giants would be mass producing second-generation HBM for NVIDIA and it would be for the company’s Pascal lineup of GPUs.



Mass Production Of Second-Gen HBM For NVIDIA Will Commence From Q1, 2016, But Not After The Pilot Production Phase Starts By The End Of 2015
Industry sources have stated that both Samsung Electronics and SK Hynix are planning to mass produce second-gen High Bandwidth Memory (HBM) for NVIDIA’s Pascal lineup of graphics processors. The mass production phase is expected to start during the first quarter of 2016, but before that, pilot production and reliability tests completion will be taking place by the end of 2015. SK Hynix, the world’s second largest memory chipmaker was already providing AMD and NVIDIA with first-generation before Samsung decided to jump into this market.



Unfortunately, Micron is not going to be a part of the deal since the firm uses HMC (Hybrid Memory Cubes). While it is a technology that is very similar to HBM, its productivity levels are not up to par. With second-generation HBM incorporated into the Pascal GPU lineup from NVIDIA, the company will effortlessly be able to achieve far more than what first-generation HBM present on AMD Fury X, Fury, Nano, and Fury X2 brought to the table.

Additionally, NVIDIA will also have access to denser chips, which will no doubt result in graphic cards boasting VRAM in between the range 16 and 32GB, along with a 4096 memory bus-width that will deliver astounding performance in future gaming titles that are being played on the 4K and 8K resolutions. Furthermore, HBM will be able to process data 4-8 times faster when compared to traditional DDR memory while at the same time, consume 40 percent less power than the latter.



Advertisements


NVIDIA Pascal GPU module

Apart from graphic cards, both Samsung and SK Hynix are also gearing up to supply second-generation HBM to servers and supercomputers, and after that, branch out to provide memory chips for Sony’s and Microsoft’s next-generation gaming consoles. According to Business Korea, the following has been quoted:

“There are clear signs of a change in the structural design of CPUs, GPUs, DRAM memory, and storage led by Intel and Nvidia, which control standards in the market. Samsung and SK Hynix are both expanding their technical bases for various types of next-gen memory, including HBM chips.”


Apart from featuring second-generation HBM, NVIDIA Pascal graphics processors are also going to feature NVLink. NVLink will allow data to move between GPUs and CPUs 5-12 times faster compared to the level of speeds delivered by PCI-Express. The first products sporting Pascal GPUs will arrive during the second quarter of 2016.


Source

Read more: http://wccftech.com/nvidia-confirmed-supplied-gen-hbm-samsung-sk-hynix/#ixzz3mN2YoINQ


http://wccftech.com/nvidia-confirmed-supplied-gen-hbm-samsung-sk-hynix/
 
NVIDIA’s next-gen GP100 ‘Pascal’ GPU enters testing phase

NVIDIA’s next-gen GP100 ‘Pascal’ GPU appears to have entered testing phase, as it was recently imported to India for internal testing purposes.

The next-gen GPU has been listed on the Indian import tracking portal Zauba as ‘graphics processor integrated circuits 433X3’ and was imported to India from Taiwan last week. Each chip has is claimed to have a customs value of around $3150. The GPU was taped out in June this year, and the first samples were sent to India for testing in August 2015. It is expected that the first products based on the GPU will be out by the second quarter of 2016.



According 3DCenter.org, the NVIDIA GP100 GPU could feature around 17 billion transistors and will include HBM2 memory interface with up to 32GB HBM2 memory on the flagship professional graphics card. Gaming oriented card could pack a maximum of 16GB HBM2 memory. Performance is expected to jump by around 60 to 90% compared to the Maxwell GM200 GPU which forms the base for the NVIDIA GTX Titan X.
 
Performance is expected to jump by around 60 to 90% compared to the Maxwell GM200 GPU which forms the base for the NVIDIA GTX Titan X.
Boa noticia do ponto de vista tech, péssima para quem não está a par e anda agora a gastar 400 500 e 800€, a comprar 970/980/Ti as we speak, para daqui a 10 meses estarem a sair gráficas que fazem quase o dobro, e provavelmente até o mainstream vir com performance semelhante (960 + 90% = 980), e com 4GB de HBM2....por 200 220€.
 
Back
Topo