• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

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

Page 645 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

O M G !!!

The lies! They never end!

View attachment 103405

View attachment 103406

Oh, the unbearable agony of going through all the screenshots! I hope he didn't end up in hospital. Real journalism is so hard these days!
Pretty impressive it can hit 11.4 GHz at only 40 W...
 
R15 through R23 doesnt scale with RAM speed at all. Its why the 9950X ES leaker ran his R23 at 4800MT/s. Actually helps the score a tiny bit by feeding less power to memory, thus having more to give to CPU.

I think R24 does a little bit, probably no more than 2% - 3%.
 
Not impressed?
No. I anticipated max OC -30CO ST score not ~Stock 😛
I guess il degrade that 9700X very fast trying 6Ghz like i degraded my 5600x 4years ago only took me 45days of running ~4.9ghz and 2100fclk 1.15vsoc for it to degrade. Then it needed 1.464V for 4.85G and could only do 2000fclk. Thank god 5800x3d released as by the end it could only do 4.65G and 2000fclk started giving whea's.
 
R15 through R23 doesnt scale with RAM speed at all. Its why the 9950X ES leaker ran his R23 at 4800MT/s. Actually helps the score a tiny bit by feeding less power to memory, thus having more to give to CPU.

I think R24 does a little bit, probably no more than 2% - 3%.
While my ST only saw a 120->126 score going from jedec 4800 to fast tuned ram, MT sees a much larger gain. I'll run it real quick for reference.

4800 JEDEC - 1,950
6400 Tuned - 2,280 2,190

edit: wrong MT score from memory, 2,280 was probably with PBO+CO
 
Last edited:
While my ST only saw a 120->126 score going from jedec 4800 to fast tuned ram, MT sees a much larger gain. I'll run it real quick for reference.
I went over a couple of detailed reviews of Raphael CPUs where it was shown that the effective clocks in ST load like SuperPI for the two best cores reached 5700-5750 MHz at stock and 5850 MHz with +100 PBO boost and a tuned curve, with Fmax 5850 and 5950 respectively. Looks like something is really wrong with the setup where the ST clocks can't exceed 4650Mhz.
The rate at wich the score increase as he increase power, we ll know more once Computerbase do their review since they ll make tests at 45, 65, 88 and 142W like they did for the 7950X.
Yes, we'll know for sure, although I was asking why you assumed that the VRM voltage is held at a fixed value, while the telemetry indicated otherwise.
 
Last edited:
I went over a couple of detailed reviews of Raphael CPUs where it was shown that the effective clocks in ST load like SuperPI for the top two cores reaches 5700-5750 MHz at stock and 5850 MHz with +100 pbo boost and a tuned curve, with Fmax 5850 and 5950 respectively. I think something is wrong with the setup where the ST clocks can't exceed 4650Mhz
It runs 5650 in CB 2024 ST. In other singlethreaded workloads, it can 5700 or 5750. FMAX/peak reported frequency when unloaded hits 5850.
 

The updated design was introduced by Mark Papermaster, AMD's chief technology officer, and he began by pointing out that the changes culminated from its collaboration with Samsung, which licenses AMD's graphics tech for the Exynos range of smartphone and tablet processors.
That's very interesting (emphasis mine).

Cross pollination of technologies between smartphone and PC!
 
If it’s N4X that’s even better, as it’s meant for extreme performance and no wonder AMD is able clock highly so easily.

One could say N4X is better than N3B for desktop.
 
Back
Top