Nvidia has announced it next-generation GPU called ‘Pascal’ at the GPU Technology Conference (GTC) in San Jose, California, Tuesday.
This next-gen ‘Pascal’ will be incorporated with a new interconnect called ‘NVLink’ which is expected to increase the bandwidth among GPUs over that of PCIe by five to 12 times.
“NVLink is a chip-to-chip communication,” Nvidia CEO and cofounder Jen-Hsun Huang said during his keynote at the GPU Technology Conference, adding that the its “programming model is essentially that of PCIe, but with enhanced direct memory access (DMA) capabilities, which should make it easy for developers to adapt their software to it.”
Reports said that ‘Pascal’ will vividly re-architect the traditional graphics card by replacing the PCI Express bus and combining it with a new 3D memory technology.
“Because we’re able to pile on eight blocks ? eight signals in one block ? scale up to four in the first generation and up to eight in the second generation, we’re able to take PCI Express and increase its performance, increase its bandwidth, increase its throughput, by five to 12 X,” Huang said.
Furthermore, Pascal will also come with another a 3D packaging.
“The challenge, of course, is that the GPU already has a lot of pins,” Huang said. “It’s already the biggest chip in the world. The interface is already extremely wide.” By this 3D packaging, Nvidia will “build chips on top of other chips. We’re going to pile heterogeneous chips ? meaning different types of chips ? on one wafer,” Huang said.
At this moment, Nvidia already has one major CPU partner for NCVLink in the name of IBM.
“NVLink enables fast data exchange between CPU and GPU, thereby improving data throughput through the computing system and overcoming a key bottleneck for accelerated computing today. We think this technology represents another significant contribution to our OpenPOWER ecosystem,” said IBM VP and fellow Bradley McCredie in a statement.
Nvidia will start shipping the Pascal processor in 2016.
Moreover, Nvidia also announced the following:
- its next-generation Titan card, the GeForce GTX Titan Z, starting at a whopping $2,999.
- Its Iray VCA, a $50,000 virtual computing appliance for rendering images using modeled photons
- its $192 Jetson TK1, an embedded development board to take its Tegra K1 to robots and computer vision, and disclosed the name of its next-generation Tegra chip ? Erista