News Intel Shows Off Multi-Chiplet Sapphire Rapids CPU with HBM

Sep 2, 2021
17
17
15
"Therefore, it looks like HBM-equipped SPRs will only be offered to select
clients (just like Intel's Xeon Scalable 9200 CPUs with up to 56 cores) and will mostly be aimed at supercomputers. "

The question is who would be willing enough to integrate those in its supercomputers anymore, after the fiasco with Aurora. Which is still a running fiasco...
 
Sep 2, 2021
17
17
15
You did not really get my point. Of course Aurora is supposed to use these. But Intel practically broke multiple contract agreements, they were not able to hold any time line of whatsoever. Its still not finished, after dozens of frign years. In the end Aurora was a very expensive prestige project for intel, that pretty much back fired. They did not really make a single dime with it, on the contrary charges were in the hundreds of millions. And even worse, they simply lost quite some credibility with it. So I ask you again, who is willing to take the risk.
 

JayNor

Reputable
May 31, 2019
426
85
4,760
I think the SPR chips have been sampling since last Nov.. ... pcie5, cxl, ddr5, gen3 optane, DSA, AMX matrix math on bfloat16... and 8 stacks of HBM. Pretty amazing if it all works.

6x 9k units of Ponte Vecchio GPUs may be the hold-up now. Who else is doing 47 tile GPUs?

Maybe we'll see a demo in a couple of weeks.
 

JayNor

Reputable
May 31, 2019
426
85
4,760
The hbm is not so much of an innovation, but we haven't heard the whole story yet. One of the diagrams I saw shows 4 lanes of slingshot going directly into each SPR. They also haven't announced how they are using the DSA.

https://ecpannualmeeting.com/assets/overview/sessions/SYCL_Programming_Model_for_Aurora.pdf

Intel also hasn't shown their Rambo Cache details yet. I saw an Intel patent app with the whole foveros base layer of a GPU populated with a grid of SRAM chiplets, so Intel has some show-and-tell coming.
 
You did not really get my point. Of course Aurora is supposed to use these. But Intel practically broke multiple contract agreements, they were not able to hold any time line of whatsoever. Its still not finished, after dozens of frign years. In the end Aurora was a very expensive prestige project for intel, that pretty much back fired. They did not really make a single dime with it, on the contrary charges were in the hundreds of millions. And even worse, they simply lost quite some credibility with it. So I ask you again, who is willing to take the risk.
Maybe they did not make a single dime with it specifically, but if they used the wafers that where to be used for aurora to supply the mobile and desktop market in the last two years then they made a crap ton of money from it even with the fines.

And they also gained a lot of credibility with OEMs and "the common man" by being able to supply the whole market as if nothing had happened and without increasing the prices at all or reducing the value.

Having a business is all about weighing the pros and cons of everything and going with what you believe will benefit your company the most.
 
Sep 2, 2021
17
17
15
jeez its like talking against a wall. How did you come up with this fairy tell.
One of their big problems was with their process node, yields were crappy as it only can get. So those wafers werent reallocated for other purposes.
And dont talk like they were able to supply, because clearly the last ERs showed more or less, growth was catastrophic considering we were and still are in a mega semi conductor cycle.
So supply is one thing, but apparently the demand for their products was not the same, because everybody else benefited much more from the cycle than intel was able to, although they were less capacity strained. think about that rather.
 
jeez its like talking against a wall. How did you come up with this fairy tell.
One of their big problems was with their process node, yields were crappy as it only can get. So those wafers werent reallocated for other purposes.
So they didn't spend money or resources on wafers for aurora and had that money/resources to make wafers for other things... huge difference, changes everything.
And dont talk like they were able to supply, because clearly the last ERs showed more or less, growth was catastrophic considering we were and still are in a mega semi conductor cycle.
So supply is one thing, but apparently the demand for their products was not the same, because everybody else benefited much more from the cycle than intel was able to, although they were less capacity strained. think about that rather.
Sure demand for intel products must be really bad for them to be making more and more money while keeping prices the same.

https://www.macrotrends.net/stocks/charts/INTC/intel/revenue
  • Intel annual revenue for 2020 was $77.867B, a 8.2% increase from 2019.
  • Intel annual revenue for 2019 was $71.965B, a 1.58% increase from 2018.
  • Intel annual revenue for 2018 was $70.848B, a 12.89% increase from 2017.
 

JayNor

Reputable
May 31, 2019
426
85
4,760
I don't think it is a wafer processing yield problem for SPR. They have been in high volume mfg of Ice Lake Server chips with 38 cores, while the SPR chips require only 14 cores per tile. The SPR only uses EMIB stitching, which they also say is in high volume mfg.

My guess is Ponte Vecchio is the hold-up. The most recent presentation stated it was first silicon. They did say it will be shipping to all the customers in early 2022.
 

Ogotai

Reputable
Feb 2, 2021
323
219
5,060
So they didn't spend money or resources on wafers for aurora and had that money/resources to make wafers for other things... huge difference, changes everything.
Sure demand for intel products must be really bad for them to be making more and more money while keeping prices the same.

https://www.macrotrends.net/stocks/charts/INTC/intel/revenue
  • Intel annual revenue for 2020 was $77.867B, a 8.2% increase from 2019.
  • Intel annual revenue for 2019 was $71.965B, a 1.58% increase from 2018.
  • Intel annual revenue for 2018 was $70.848B, a 12.89% increase from 2017.
more intel PR ??
 
like most of your posts are when some one says something negative about intel ? as i said before, most of your posts, read like they come from intel PR.
Sure, but why?! What exactly do you disagree with? I at least give the data I think is relevant and comment on it, I'm not just saying "no, yo moma" like you do.
If you have an issue comment on the issue, say that you don't think that demand is measured by how much sales a company does or something.
 
I was taking the mickey at them - that's what Intel was saying about Ryzen in general and Threadripper in particular. Pot, this is kettle.
At least intel was much more careful in gluing the parts together as closely as possible.
At least if this SPR image is accurate.
Threadripper had huge gaps in-between them causing much lag.
wnBRQfaMLLBYmougbS9Rv6-970-80.jpg.webp

KH6Zio9CgFHSPbSQsvXuMo-1024-80.png.webp
 
At least intel was much more careful in gluing the parts together as closely as possible.
At least if this SPR image is accurate.
Threadripper had huge gaps in-between them causing much lag.
wnBRQfaMLLBYmougbS9Rv6-970-80.jpg.webp

KH6Zio9CgFHSPbSQsvXuMo-1024-80.png.webp
....yeah, no, Intel had to stick them very close together to fit the HBH chips around the silicon - that's what AMD did with Vega, and it caused problems when sticking the heat spreader on top. Keeping the chiplets apart gives a bit of tolerance while the distance is too small to really have any impact on the interconnects.