Discussion Intel current and future Lakes & Rapids thread

Page 621 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
As it turns out. The 79,150 CineBench R23 MT score was for a 2S(Two socket processor for a total of 108 Cores), And 2S EPYC Rome gets 92,000 points and 2S Milan get's 113,000 points So it's looking Really Really Bad right now for Sapphire Rapids.


Not looking good for Sapphire Rapids

EPYC Rome
View attachment 60841


EPYC Milan
View attachment 60840
There is probably something wrong there with the Intel system, because Golden Cove is EXCEPTIONALLY strong in Cinebench.
 

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
There is probably something wrong there with the Intel system, because Golden Cove is EXCEPTIONALLY strong in Cinebench.
At ISO Speed it's even with Zen3, that is the desktop version that has a monolithic nature with a Ring Bus connectivity.. But Xeons(including Sapphire Rapids) use a Mesh connectivity and to make things worst, Intel is using sub-par glue to connect Sapphire Rapids(4 tiles of 14 cores each)

1651241639613.png
 
Last edited:
  • Like
Reactions: Tlh97 and ftt

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
No, it's ahead.

They are Even on CBR15
1651241713972.png


Alder lake is ahead about 16% on CBR20

1651243834147.png



Also keep in mind that it gets harder and harder for Intel to compete the larger the CPUs get, due to TDP, Thermals and their Mesh Ring connectivity, that issue has been compounded for Sapphire Rapids because now it's a MCM(Multi Chip Module)

1651244524136.png
 
Last edited:

lobz

Platinum Member
Feb 10, 2017
2,057
2,856
136
yeah all that is fine, but I doubt that a 2S 52 core SPR will score less than a Zen 2 counterpart when it's released and adequate testing is done
 
  • Like
Reactions: Tlh97 and coercitiv

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
Milan was released about a year ago, not 3-4.
Sapphire Rapids it's old tech from 3-4 years ago that has been plagued by delays after delays due to Intel's Over Ambitious plans. Had Intel been able to execute, SPR would have been facing off against ROME(at the tail end of Rome product cycle)

1651254929847.png
 

mikk

Diamond Member
May 15, 2012
4,296
2,382
136
“Intel 4 Meteor Lake has now successfully booted Windows, Chrome, and Linux. The speed at which the team was able to achieve this milestone is a significant sign of the health of both Meteor Lake and our Intel 4 process technology.”
— Pat Gelsinger, Intel CEO
We have officially powered-on our first disaggregated product: Meteor Lake. An incredible milestone resulting from the efforts of so many across @intel. Congratulations, team!

And furthermore a desktop version of Meteor Lake is coming says MLID:

 

eek2121

Diamond Member
Aug 2, 2005
3,410
5,049
136
No, it's ahead.

NO, it is not.

The issue is that you are looking at the picture Intel gave you vs. reality.

In every metric imaginable Intel is behind except 'raw performance ignoring power'.

Note that I'm not going on the typical Intel power consumption rant.

If you take ANY SPR platform and compare it to atop Milan-X platform, the Milan-X platform is going to deliver far better perf/watt while running cooler than the Intel chip, and they are doing that with FAR less silicon. Intel allowed 4S on SPR precisely because they are losing this little 'race'. AMD is all about producing inexpensive, cool running, power efficient chips. Intel is all about throwing money and climate change out the window in order to gain 2% extra performance.

I try to be neutral to both parties. I've been in this game long enough that I know Intel can deliver when pushed.

I am still waiting on that delivery Intel. I get that you are a sub, but sheesh...
 

semiman

Member
May 9, 2020
77
68
91
There is probably something wrong there with the Intel system, because Golden Cove is EXCEPTIONALLY strong in Cinebench.

That's an ES sample according to the screenshot. CPU-Z single score 515(my 12600K gets 753) means that it's barely running 3Ghz even with a single turbo. Even Xeon 8380 with old Intel 10nm used better clocks(2.3 ~ 3.4 Ghz).

So SPR isn't super bad, but it'll have a tough time fighting Genoa(at least they can keep some HPC markets thanks to HBM variants). Intel will have a hard time before they get Sierra forest and Granite Rapids in 2024~2025.
 
  • Like
Reactions: lobz and Tlh97
Jul 27, 2020
28,018
19,126
146
First I'm hearing of that;

The new Intel 15th Gen Core "Arrow Lake" CPUs will be launching on not one but three different process nodes: Intel 4, Intel 20A, and External N3 technology... but spoiler alert: it'll be TSMC.

Take that with a huge rock of salt :)
 

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
Intel allowed 4S on SPR precisely because they are losing this little 'race'. AMD is all about producing inexpensive, cool running, power efficient chips. Intel is all about throwing money and climate change out the window in order to gain 2% extra performance.

I try to be neutral to both parties. I've been in this game long enough that I know Intel can deliver when pushed.

I am still waiting on that delivery Intel. I get that you are a sub, but sheesh...

4S is Not going to Cut it when your competitor will release 96 and 128 cores in a single socket system with an IPC and Speed advantage ...So they have allowed SPR to expand to 8S. It's going to be a real challenge to set up those COD(Cluster on Demand) and SNC(Sub NUMA Clusters)

1651331973001.png
 

Exist50

Platinum Member
Aug 18, 2016
2,452
3,106
136
Arrow Lake is a 2024 platform confirmed by Intel in February, it won't or cannot come before mid 2024 because of 20A which is manufacturing ready in H1 2024.
The process is likely the least of their issues. N3 would work, if they needed something and were otherwise ready. Or just the new architectures on Intel 3 instead of 4 would be plenty of gain.

Desktop MTL isn't going to be anything this forum is excited about. Think we're squaring up for basically a repeat of 11th gen, at least in terms of products available. Going to be a long, sad wait till 2024. Not expecting mobile to be too much better either, tbh.
 
Last edited:

jpiniero

Lifer
Oct 1, 2010
16,809
7,253
136
4S is Not going to Cut it when your competitor will release 96 and 128 cores in a single socket system with an IPC and Speed advantage ...So they have allowed SPR to expand to 8S. It's going to be a real challenge to set up those COD(Cluster on Demand) and SNC(Sub NUMA Clusters)

I wouldn't read too much into SPR supporting 4S and 8S. Intel has always supported those in the past.

The process is likely the least of their issues. N3 would work, if they needed something and were otherwise ready. Or just the new architectures on Intel 3 instead of 4 would be plenty of gain.

Desktop MTL isn't going to be anything this forum is excited about. Think we're squaring up for basically a repeat of 11th gen, at least in terms of products available. Going to be a long, sad wait till 2024. Not expecting mobile to be too much better either, tbh.

My expectation is that anything that uses I4/I3/20A is going to be tiny volume at best, with the hope that they can make 18A work eventually.
 

nicalandia

Diamond Member
Jan 10, 2019
3,331
5,282
136
So SPR isn't super bad, but it'll have a tough time fighting Genoa(at least they can keep some HPC markets thanks to HBM variants). Intel will have a hard time before they get Sierra forest and Granite Rapids in 2024~2025.

I am going to leave this comment here which I believe it summarizes the ES Performance and possible performance on release samples.

"Although this is an ES chip and we expect performance to improve significantly for the Sapphire Rapids-SP chip in the QS state, the final chip may still be only competitive against AMD's EPYC Milan parts while it will be launching at the same time when AMD will come out guns blazing with its next-generation EPYC Genoa 7004 CPUs. With such high power draw and not enough performance at hand, it looks like Intel will be getting a serious beating in the server segment by AMD, something that has been the case over the past few generations"



Will Mr. Gelsinger still think it's a "Knife Fight on a Phone Booth" after the serious beat down yet another Xeon gets at the hands of an EPYC?

1651346620103.png

How many more people will get fired for buying Xeons?

1651346824762.png
 
Last edited: