Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)

Page 605 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Mopetar

Diamond Member
Jan 31, 2011
8,024
6,489
136
Well, AMD once made tri-core CPUs so....

Those were four core CPUs with one core disabled, but since the hardware wasn't physically fused off people could unlock the fourth core and get a free upgrade.

The Xbox 360 CPU (IBM POWER-based architecture) had three physical cores as I recall. Outside of custom silicon, something not using a power of two is unusual, and when you do see it (e.g., 6 core CPUs) it tends to be a case of additional physical hardware present, but disabled.

There were exceptions such as AMD Thuban (Phenom II) which had six cores, and Intel may have also had a six core HEDT chip at one point, but I think they were mainly making 8-core or 16-core CPUs and binning those. They may have had a few with an odd (unusual, not non-even) core count.
 

Mopetar

Diamond Member
Jan 31, 2011
8,024
6,489
136
Indeed but that was still odd and they were mocked, as usual.

I don't really remember anyone mocking people for it as a lot of the people buying a 3-core CPU were doing so with the intent of unlocking the fourth core and saving some decent money for the time.

It was actually a pretty good time for consumers as both AMD and Intel had compelling parts and were frequently getting into price wars with each other.
 

Hail The Brain Slug

Diamond Member
Oct 10, 2005
3,255
1,704
136
Roughly 17% over Z4, just as the slide said.
Also R23 was the only IPC metric AMD explicitly said was nT, the rest could be anything.
I think we should expect more than 16% average MT performance uplift is supposed to have a higher frequency residency.

However, whether geekbench's rapidfire 1 second long loads are enough to show that, I don't know.
 
Last edited:

PJVol

Senior member
May 25, 2020
632
561
136
Roughly 17% over Z4, just as the slide said.
To be fair, the actual core clocks in the R23 MT are still unknown. Not to mention the fact that some reviewers measured 29,3 - 29,5 MT score for the 7900X which makes it only +12%.
I think we should expect more than 16% average MT performance uplift if the 9950X is supposed to have a higher frequency residency.
Omg, those residents again...
AMD has implemented the Effective Frequency Interface since first Zen, idk why all this marketing fluff
The performance IS core effective clock, which in turn depends on PVT limits
 
Last edited:

Fjodor2001

Diamond Member
Feb 6, 2010
3,938
408
126
AI is robbing us of a bunch of neat stuff and I am not happy.

AMD has completely changed internal roadmaps.
You mean because AMD is spending die area and budget on NPU instead of CPU/iGPU/... (on laptop CPUs at least)?

But what do you mean by that they are completely changing the internal roadmap? What did it look like before, and what have they changed?
 
Reactions: Nothingness

eek2121

Diamond Member
Aug 2, 2005
3,053
4,282
136
You mean because AMD is spending die area and budget on NPU instead of CPU/iGPU/... (on laptop CPUs at least)?

But what do you mean by that they are completely changing the internal roadmap? What did it look like before, and what have they changed?

They are focusing on enterprise stuff, and nothing related to Zen 5 (yet)
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,938
408
126
They are focusing on enterprise stuff, and nothing related to Zen 5 (yet)
Not sure I follow. In what way do you mean AMD has changed the internal roadmap? Are you saying it previously focused on e.g. client desktop CPUs and now enterprise server CPUs instead?

Also, what do you mean with ”nothing related to Zen5 (yet)”?

If you can elaborate a bit it would be appreciated.
 

Abwx

Lifer
Apr 2, 2011
11,172
3,872
136
But the 7900X3D has 3D V-cache. So would a better comparison not be 7900X, but with the TDP limit set to 120W?
V-cache doesnt affect the CB R23 score significantly, the only downside is that the 7900X3D real TDP is probably lower than the 9900X one.

Best comparison is to use the average between the 7900X and 7900X3D, the 7900X score 29500 while the 7900X3D score 27200, so about 28350 pts for a theorical 120W TDP/162W PPT 7900X.

That put the 9900X at 34.5/28.35 = 21% better perf at same power, and this should be the same for the 9700X and 9950X, that is, 20-22% better perf than their Zen 4 counterparts at same power, indeed all these numbers were predictible since AMD said that the MT IPC is 17% better in CB R23 and that the N4P process provide 28%
better perf/watt than the N5 used fo the 7000 series.
 

Fjodor2001

Diamond Member
Feb 6, 2010
3,938
408
126
Best comparison is to use the average between the 7900X and 7900X3D
Why would it be better to use such an average as comparison instead of just 7900X set to 120W TDP limit?

9900X is 120W TDP and has no 3D V-cache, so 7900X which also does not have 3D V-cache should be the closest comparison if 120W TDP limit set for it.
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |