Tag: pascal

Nvidia talks up Pascal

nvidiaNvidia CEO Jen-Hsun Huang has been telling the Nvidia Computex 2016 press conference about his latest Pascal-based GPUs.

He claimed they were being broadly adopted in applications including data centers, car-use electronics and deep-learning platforms.

He said Pascal GPUs will continue to use TSMC 16nm FinFET manufacturing process to provide enhanced performance than previous-generation products.

He said next-generation servers built by Nvidia’s Tesla P100 GPUs can compete against servers that are equipped with several hundreds of CPUs. Nvidia also showcased a server system manufactured by Quanta Computer.

He also showed the Pascal-based GeForce GTX 1080 reference design, displayed at the conference, the board is manufactured by Foxconn.

Nvidia has been pushing its Pascal GPUs into industry sectors, including virtual reality, car-use electronics and deep-learning. As for mobile devices such as smartphones and tablets, Huang noted that Nvidia has no interest in smartphones.

Nvidia shows off Pascal GPU

nvidiaNvidia has unveiled its next-generation Pascal-based GeForce GPUs.

Dubbed the GeForce GTX 1080 and GeForce GTX 1070, the pair of chips are based around Nvidia’s Pascal 16nm FinFET technology.  They are similar to Nvidia’s high-end data centre Tesla P100 processing engine but geared for the consumer gaming market.

The GP104 GPU at the heart of the new GeForce cards has eight billion transistors and features a 256-bit memory interface with 8GB of Micron GDDR5X graphics memory on the GeForce GTX 1080. The GTX 1070 uses standard GDDR5.

The GeForce GTX 1080 hit 2.1GHz at one point during the demonstration and Nvidia is not talking about the GTX 1070 clockspeed.

Nvidia CEO Jen-Hsun Huang claimed the new GeForce GTX 1080 is faster than a pair of GeForce GTX 980 cards in SLI and the company’s very expensive Titan X graphics card but at half the price.

The new GeForce GTX 1080 will be offered in two versions, a standard card for  $599 or an overclockable Founders Edition for $699. The standard GTX 1070 will cost at $379, while a Founders Edition will be priced at $449. Availability for the GTX 1080 will be in the shops on 27 May and the GTX 1070 for 10 June.


Nvidia gives the world a new road map

Nvidia has announced a new roadmap for its GPU families at the GTC conference and it appears to have carried out some major surgery.

Pascal, Nvidia’s latest GPU architecture, is being introduced in between Maxwell and Volta. It has absorbed old Maxwell’s unified virtual memory support and old Volta’s on-package DRAM, integrating those feature additions into a single new product. We always thought that Pascal was a rubbish name for a GPU, after all, who wants a chip which breaks because you forgot to put a semi-colon in the middle of a nest.

Volta was supposed to follow Maxwell in 2014. Volta’s marquee feature would be on-package DRAM, using Through Silicon Vias (TSVs) to die stack memory and place it on the same package as the GPU.

What appears to have changed is Nvidia’s definition of Maxwell and Volta. Maxwell has lost its promised unified virtual memory feature which will now be under the bonnet of the chips after Maxwell. All users can hope for from Maxwell will get is the software based unified memory feature being rolled out in CUDA 6.

Nvidia is not telling anyone about its second generation Maxwell GPUs and how those might be integrated into professional products.

Maxwell’s best feature will be DirectX 12 support and will ship in 2014 as scheduled.

Meanwhile Volta has been pushed back and stripped of most of what people will find interesting, Its on-package DRAM will be promoted to the GPU before Volta, and while the name Volta still exists, all anything knows about the chip is that will come after the 2016 GPU.

Nvidia has not said anything else directly about the unified memory plans that Pascal has inherited from old Maxwell. Pascal will get NVLink which is an attempt to supplant PCI-Express with a faster interconnect bus.

Nvidia thinks that the 16GB/sec made available by PCI-Express 3.0 is not enough when compared to the 250GB/sec+ of memory bandwidth available within a single card. PCIe 4.0 will bring higher bandwidth yet but Nvidia wants to push its own bus to achieve the kind of bandwidth it wants.

According to the roadmap, the result is a bus that looks a whole heck of a lot like PCIe but uses tighter requirements and a true point-to-point design. NVLink uses differential signalling, with the smallest unit of connectivity being a “block.”

A block contains eight lanes, each rated for 20Gbps, for a combined bandwidth of 20GB/sec. In terms of transfers per second this puts NVLink at roughly 20 gigatransfers/second, as compared to an already staggering 8GT/sec for PCIe 3.0.

Nvidia has knocked up a Pascal prototype and it will be put on a motherboard parallel to the board with each Pascal card connected to the board through the NVLink mezzanine connector. This allows GPUs to be cooled with CPU-style cooling methods in a server rack.