Wednesday, November 13th 2019
7nm Intel Xe GPUs Codenamed "Ponte Vecchio"
Intel's first Xe GPU built on the company's 7 nm silicon fabrication process will be codenamed "Ponte Vecchio," according to a VideoCardz report. These are not gaming GPUs, but rather compute accelerators designed for exascale computing, which leverage the company's CXL (Compute Express Link) interconnect that has bandwidth comparable to PCIe gen 4.0, but with scalability features slated to come out with future generations of PCIe. Intel is preparing its first enterprise compute platform featuring these accelerators codenamed "Project Aurora," in which the company will exert end-to-end control over not just the hardware stack, but also the software.
"Project Aurora" combines up to six "Ponte Vecchio" Xe accelerators with up to two Xeon multi-core processors based on the 7 nm "Sapphire Rapids" microarchitecture, and OneAPI, a unifying API that lets a single kind of machine code address both the CPU and GPU. With Intel owning the x86 machine architecture, it's likely that Xe GPUs will feature, among other things, the ability to process x86 instructions. The API will be able to push scalar workloads to the CPU, and and the GPU's scalar units, and vector workloads to the GPU's vector-optimized SIMD units. Intel's main pitch to the compute market could be significantly lowered software costs from API and machine-code unification between the CPU and GPU.
Image Courtesy: Jan Drewes
Source:
VideoCardz
"Project Aurora" combines up to six "Ponte Vecchio" Xe accelerators with up to two Xeon multi-core processors based on the 7 nm "Sapphire Rapids" microarchitecture, and OneAPI, a unifying API that lets a single kind of machine code address both the CPU and GPU. With Intel owning the x86 machine architecture, it's likely that Xe GPUs will feature, among other things, the ability to process x86 instructions. The API will be able to push scalar workloads to the CPU, and and the GPU's scalar units, and vector workloads to the GPU's vector-optimized SIMD units. Intel's main pitch to the compute market could be significantly lowered software costs from API and machine-code unification between the CPU and GPU.
50 Comments on 7nm Intel Xe GPUs Codenamed "Ponte Vecchio"
Intel has lost my trust personally. I have never seen so many security vulnerabilities in my life.
Another new one, Xombieland V2 was announced yesterday, and Intel announced 70 security patch fixes incoming. lol how many times this year now? wow... Ryzen... none needed... since launch lol
The real question is: will they?Spectre v2. Not many admitedly, but not "none" either.I mean, how else are people going to spin this into a brand war where intel looks bad?
See the link? Ivy, Sandy.... It's another bridge.
Will be interesting to see how much of it is Intel DNA and how mush is Koduri Radeon DNA...
Anyway, I thought Intel was working on a Graphics card for games and honestly I been waiting for this. I think it is some sort of iteration of the previous attempts for Intel like the Xeon Phi and Larrabee. maybe it is a combination of the two now although this Xe is kinda like a Xeon Phi to me though.
I'm italian so i know better i suppose.
Anyway, back to the topic , is funny how their gonna call it when right now Venice is floading .
Is gonna be a fail?
Also that Xe will be in gaming segment but not the Aurora (and probably not the exact dies used in Aurora).
I may me wrong, but didn't you spend a lot of time "hacking" Intel cpu to get rid of the management software? Maybe I am reading it wrong. I mean I don't have to mess with Ryzen at all.
This bizarre fetish for heterogeneous computing Intel has above all else needs to stop or it will continue to ruin everything they do related to these things. I know why they do this, Nivida really screwed them up in recent years and took a lot of potential customers from them and they had no way to respond to any of that but this isn't the way to do it.
How is it that all CPU producers are focused on nomenclature with italian locations in this period
Returning in topic. This looks like a mix of Larrabee and Vega ... can Intel with all its money really fail with this project? There are some precedents like P4, Itanium, Larrabee, 10nm (let's remember here that 14nm was also delayed and they said at the time that wouldn't affect 10nm which was awesome and in time; similarly as they are now saying 7nm is awesome and imminent)
That's not to say Intel isn't being hit around the face with far MORE, and far more SERIOUS vulnerabilities - theirrecent track record is pretty dire, after all. But to try and claim AMD aren't being hit with anything is just wrong at best, and a lie or outrageous fanboying at worst.
Don't create misinformation. TPU reported all of these things and you should know better if you've been following the news feed.
Corruption was the main feature from the start. More than 20 years later they started admitting that probably all the billions spent and the components installed may be useless.
one of the latest problem was that metal junctions of the dams are affected by corrosion due to salty water (who could've imagined that eh...). If i remember well they cheaped out on junctions and soldering materials. Also they discovered that while the dams are active, debris fill the space where they should fold while unactive (sand and sea, who could have imagined again). Moreover, the entire build of the dam needs so much maintainance that will be so economically unsustainable that it will probably ditched in the end. It is basically a legalized scam, funded with public money with the excuse of job creation and industry boosting.