News Posts matching "PCI-Express 4.0"

Return to Keyword Browsing

AMD's CES 2019 Keynote - Stream & Live Blog

CPUs or GPUs? Ryzen 3000 series up to 16 cores or keeping their eight? Support for raytracing? Navi or die-shrunk Vega for consumer graphics? The questions around AMD's plans for 2019 are still very much in the open, but AMD's Lisa Su's impending livestream should field the answers to many of these questions, so be sure to watch the full livestream, happening in just a moment.

You can find the live stream here, at YouTube.

18:33 UTC: Looking forward, Lisa mentioned a few technology names without giving additional details: "... when you're talking about future cores, Zen 2, Zen 3, Zen 4, Zen 5, Navi, we're putting all of these architectures together, in new ways".

18:20 UTC: New Ryzen 3rd generation processors have been teased. The upcoming processors are based on Zen 2, using 7 nanometer technology. AMD showed a live demo of Forza Horizon 4, using Ryzen third generation, paired with Radeon Vega VII, which is running "consistently over 100 FPS at highest details at 1080p resolution". A second demo, using Cinebench, pitted an 8-core/16-thread Ryzen 3rd generation processor against the Intel Core i9-9900K. The Ryzen CPU was "not final frequency, an early sample". Ryzen achieved a score of 2057 using 135 W, while Intel achieved a score of 2040 using 180 W.. things are looking good for Ryzen 3rd generation indeed. Lisa also confirmed that next-gen Ryzen will support PCI-Express 4.0, which doubles the bandwidth per lane over PCI-Express 3.0. Ryzen third generation will run on the same AM4 infrastructure as current Ryzen; all existing users of Ryzen can simply upgrade to the new processors, when they launch in the middle of 2019 (we think Computex).
Ryzen third generation uses a chiplet design. The smaller die on the right contains 8-cores/16-threads using 7 nanometer technology. The larger die on the left is the IO die, which consists of things like the memory controller and PCI-Express connectivity, to shuffle data between the CPU core die and the rest of the system.

AMD Vega 20 GPU Could Implement PCI-Express gen 4.0

The "Vega 20" silicon will be significantly different from the "Vega 10" which powers the company's current Radeon RX Vega series. AMD CEO Dr. Lisa Su unveiled the "Vega 20" silicon at the company's 2018 Computex event, revealing that the multi-chip module's 7 nm GPU die is surrounded by not two, but four HBM2 memory stacks, making up to 32 GB of memory. Another key specification is emerging thanks to the sharp eyes at ComputerBase.de - system bus.

A close inspection of the latest AMDGPU Linux driver includes PCI-Express link speed definitions for PCI-Express gen 4.0, which offers 256 Gbps of bandwidth per direction at x16 bus width, double that of PCI-Express gen 3.0. "Vega 20" got its first PCIe gen 4.0 support confirmation from a leak slide that surfaced around CES 2018. AMD "Vega" architecture slides from last year hinted at a Q3/Q4 launch of the first "Vega 20" based product. The same slide also hinted that the next-generation EPYC processor, which we know are "Zen 2" based and not "Zen+," could feature PCI-Express gen 4.0 root-complexes. Since EPYC chips are multi-chip modules, it could also hint at the likelihood of PCIe gen 4.0 on "Zen 2" based 3rd generation Ryzen processor family.

PCI-Express 4.0 Pushes 16 GT/s per Lane, 300W Slot Power

The PCI-Express gen 4.0 specification promises to deliver a huge leap in host bus bandwidth and power-delivery for add-on cards. According to its latest draft, the specification prescribes a bandwidth of 16 GT/s per lane, double that of the 8 GT/s of the current PCI-Express gen 3.0 specification. The 16 GT/s per lane bandwidth translates into 1.97 GB/s for x1 devices, 7.87 GB/s for x4, 15.75 GB/s for x8, and 31.5 GB/s for x16 devices.

More importantly, it prescribes a quadrupling of power-delivery from the slot. A PCIe gen 4.0 slot should be able to deliver 300W of power (against 75W from PCIe gen 3.0 slots). This should eventually eliminate the need for additional power connectors on graphics cards with power-draw under 300W, however, the change could be gradual, as graphics card designers could want to retain backwards-compatibility with older PCIe slots, and retain additional power connectors. The PCI-SIG, the special interest group behind PCIe, said that it would finalize the gen 4.0 specification by the end of 2016.

Intel 14-nanometer Skylake Platform To Support DDR4, PCIe 4.0, SATA Express

Intel's first chips based on the company's new, and Industry first, 14-nanometer manufacturing process are expected to hit markets in late 2015. With Skylake, Intel will introduce their new 9th-generation Intel HD IGP. The new platform will be the first to bring dual-channel DDR4 memory support. Skylake won't be the first platform to support DDR4 memory. In the 2H of 2014, Intel will launch their enthusiast grade Haswell-E platform, with support for quad-channel DDR4 memory. Skylake will be more of an evolution of Broadwell, which in turn is essentially an die shrink of Haswell to 14nm.

Additionally, the new mainstream platform will bring in support for PCI-E 4.0, essentially doubling the bandwidth offered by the current PCI-E 3.0 standard. More powerful GPUs from NVIDIA and AMD should be able to take advantage of the improved bandwidth, as their cards keep getting more and more powerful with each passing generation. Skylake will also introduce support for SATA Express. The advantage? SATA Express allows for a max bandwidth of about 16 Gb/s, more than 2.5x the 6 Gb/s bandwidth offered by the current SATA standard. While the product slide doesn't specify exactly as to when the first Skylake based products are scheduled to hit the market, our best guess places it at the end of 2015.

PCI-SIG Announces PCI-Express 4.0 Evolution to 16 GT/s, Twice That of PCIe 3.0

PCI-SIG, the organization responsible for the widely adopted PCI Express (PCIe) industry-standard input/output (I/O) technology, today announced the approval of 16 gigatransfers per second (GT/s) as the bit rate for the next generation of PCIe architecture, PCIe 4.0. This decision comes after the PCI-SIG completed a feasibility study on scaling the PCIe interconnect bandwidth to meet the demands of a variety of computing markets.

After technical analysis, the PCI-SIG has determined that 16 GT/s on copper, which will double the bandwidth over the PCIe 3.0 specification, is technically feasible at approximately PCIe 3.0 power levels. The data also confirms that a 16GT/s interconnect can be manufactured in mainstream silicon process technology and can be deployed with existing low-cost materials and infrastructure, while maintaining compatibility with previous generations of PCIe architecture. In addition, the PCI-SIG will investigate advancements in active and idle power optimizations, key issues facing the industry.
Return to Keyword Browsing