• 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 562 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Since TDP for 8C Zen5 CPUs is low at 65W, can we expect MT perf on 16C Zen5 to be high at 170W TDP compared to Zen4 16C? I’m thinking there will be more power budget per core on 16C Zen5 compared to the 8C variant (vs Zen4 which was 105W at 8C and 170W at 16C), so high perf on multiple cores for longer should be possible on 16C Zen5.
 
Since TDP for 8C Zen5 CPUs is low at 65W, can we expect MT perf on 16C Zen5 to be high at 170W TDP compared to Zen4 16C? I’m thinking there will be more power budget per core on 16C Zen5 compared to the 8C variant (vs Zen4 which was 105W at 8C and 170W at 16C), so high perf on multiple cores for longer should be possible on 16C Zen5.
Maybe something like Cinebench or Vray will run at 5,7GHz all-core 🙂
 
I would like to point out one thing.

If the scores for Strix Point arer correct, and there is zero reason for them to be fake, or mistaken, then we are very close to 2060 mobile performance, and Zen 6 APUs might actually be within the range of that performance.

2060 is 6000 pts in 3D Mark TS. With 24 CUs/12 WGPs, and any form of SLC/Infinity cache - we might get it in an APU.

Strix Halo should be able to exceed it before Zen 6
 
Phoenix shows diminishing returns if it is allowed to go up to 72W almost as it it was a desktop APU:
72 / 54 = 1.33 power increase, 3218 / 2791 = 1.152 score increase
Really? )
Since when did performance increase linearly with power?
3712 / 4221 = 1.13 >> something in RDNA 3.5 consumes significantly less power
Or maybe the GFX clock in Rdna 3.5 decently higher than that in Rdna 3.0 GPU at the same power?
And btw, where does it state that these scores are "graphics"?
 
Last edited:
From just eyeballing the Strix Point notebooks previewed / leaked, a lot of them come with NVidia GPU. Priced in $1500+ range. Which would make them at rough price parity with Strix Halo.
STX1 indeed explicitly targets >$1200 SRP.
I hope there are also some Strix Point models without dGPU at lower prices...
Yeah but they're a wee bit later.
 
I hope there are also some Strix Point models without dGPU at lower prices...

Asus ZenBook S14/S16 as well as Vivobook S14/S15 are all four without dGPUs, other OEMs should follow suite as a mean to increase margins through cheaper yet same profit laptops.

 
2060 is 6000 pts in 3D Mark TS. With 24 CUs/12 WGPs, and any form of SLC/Infinity cache - we might get it in an APU.
If you meant to compare TS graphics scores, I didn't see it explicitly stated on the benchmarks page. They look more like an overall scores to me.
 
The new 890M should be on par with M4 GPU performance wise.( I can why Apple focuses on gaming more in their keynotes their GPUs are no joke but it’s not plug and play using the toolkit. )

That 890M should rule windows ultrabook/handheld gaming for a while.

IMG_0457.jpeg

 
The new 890M should be on par with M4 GPU performance wise.( I can why Apple focuses on gaming more in their keynotes their GPUs are no joke but it’s not plug and play using the toolkit. )

That 890M should rule windows ultrabook/handheld gaming for a while.

View attachment 101424

What is the expectation for Lunar Lake iGPU. Vs. Meteor Lake and vs. Strix Point?
 
Back
Top