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

Page 717 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

CouncilorIrissa

Senior member
Jul 28, 2023
542
2,123
96
At this point my biggest worry is that AMD already hyped up Zen5 vcache a bit saying it had some “interesting” features

Based on what we’ve seen so far of the core I’d expect it to have even more uplift than Zen4X3D vs Zen4, but if we apply the “reverse marketing lingo” filter, “interesting” doesn’t sound too good….
Looking at the lower temps of Zen 5 compared to Zen 4, I bet that "interesting part" is just lower frequency gap between vanilla and x3d parts.

In any case, this is it then. A 70mm², decent density, power efficiency oriented, productivity monster Zen 5 with fairly crappy gaming performance. That's the end result of our forever waiting.
Well, it's both good and completely meh. Server lads will absolutely buy it in infinity quantity, and I'm really eager to see if the microcode updates can somehow get better results out of their relatively poor performance.

The best thing about Zen 5 will be Turin. A true cloud chip.
STX looks like a decent purchase for webdevs honestly. Improved battery life, good browser and node.js performance.
 

Mahboi

Golden Member
Apr 4, 2024
1,033
1,897
96
STX looks like a decent purchase for webdevs honestly. Improved battery life, good browser and node.js performance.
I've been restraining from buying one of those tiny cheapo Zen 3 mini PCs for months despite the click finger itching so bad.
6 Zen 3 cores, 32Go RAM, crappy quality but a lot better than a raspberry Pi, perfect for a home server.
My ultimate copium is that when Kraken will come out, 8 Z5 cores at low as heck power will be much better for web.
AMD, you better vindicate my copium to the max. Even on that Samsung lithography.
 

poke01

Platinum Member
Mar 8, 2022
2,207
2,804
106
I think and still do that STX Halo is the best thing to come out Zen5 family. Zen5 core is no longer stuck to dual channel but quad channel for client and yeah RAM is soldered but its fast RAM.
 
  • Like
Reactions: misuspita

Mahboi

Golden Member
Apr 4, 2024
1,033
1,897
96
I think and still do that STX Halo is the best thing to come out Zen5 family. Zen5 core is no longer stuck to dual channel but quad channel for client and yeah RAM is soldered but its fast RAM.
Any info on how they ended up with that atrocious latency though? Someone pointed out that it's somehow worse than Zen freaking 2.
 

Joe NYC

Platinum Member
Jun 26, 2021
2,539
3,471
106
I've been restraining from buying one of those tiny cheapo Zen 3 mini PCs for months despite the click finger itching so bad.
6 Zen 3 cores, 32Go RAM, crappy quality but a lot better than a raspberry Pi, perfect for a home server.
My ultimate copium is that when Kraken will come out, 8 Z5 cores at low as heck power will be much better for web.
AMD, you better vindicate my copium to the max. Even on that Samsung lithography.
MinisForum has a brand new MiniPC using HawkPoint CPU, and adds some nice IO features.

The advantage of HawkPoint vs. either StrixPoint or Kraken is that it has 8 full cores and there is no need for Windows 11 for improved scheduling of dense cores.
 
  • Like
Reactions: FlameTail

Mahboi

Golden Member
Apr 4, 2024
1,033
1,897
96
Isn't it using LPRDDR? That isn't great latency wise.
Sure but it's the inter-CCX latency that is truly abysmal.
This is why you somehow get better gaming perf limiting to the 4 full cores than having all 12 active, although the power draw gets quite a lot worse.
From Anandtech review:

AMD%20Ryzen%20AI%209%20HX%20370%20Core%20to%20Core%20Latencies%202.png

AMD%20Ryzen%209%207940HS%20Core%20to%20Core%20Latency%202.png

Can't find the vid anymore, but I also saw a bench where they got about 10-15% better performance with just the 4 full cores, while the power draw jumped from 21W to 28W.

Frankly there's a lot to unpack in terms of imbalance and Zen5c sizes. David Huang also pointed out that Zen5c is so compact that it basically loses any point above 2Ghz, you're just better off using Zen 5 full.
 

RnR_au

Platinum Member
Jun 6, 2021
2,062
5,007
106
Maybe AMD is simply acknowledging that the 3D vcache models are the gaming CPUs and non are primarily work focused.
The interesting thing is that there are multiple reviews now that calls the 9600X the best budget gaming cpu. Its weird how some reviewers are calling the release a flop. According to some, AMD will have the best value gaming cpu and once the X3D parts arrive, the best gaming cpu bar none.
 

DisEnchantment

Golden Member
Mar 3, 2017
1,747
6,598
136
I think and still do that STX Halo is the best thing to come out Zen5 family. Zen5 core is no longer stuck to dual channel but quad channel for client and yeah RAM is soldered but its fast RAM.

I would be inclined to think that way too.

Z5 L3 is fairly big, so being too mem BW limited in many scenarios would indicate prefetching is not working very well or at least prefetching is tailored to address other workloads. So it would need fatter caches.
One evidence of this is that 7800X3D has higher SIR2017 1T score than 7700X at same frequency, around 6% faster (then it hits some other bottleneck likely)

I would imagine having a bigger L2/SLC would alleviate some of this.
However for Strix Halo, having LLC which can prefetch and keep some of the data ready, overhauling the SDF for improved latency and throughput and the Octa channels of LPDDR5X would really help with bandwidth and memory parallelism.

Strix Halo is going to be one interesting part.
 
Last edited:

HurleyBird

Platinum Member
Apr 22, 2003
2,760
1,455
136
Yes, and some of the ones they throw at desktop CPUs are hella obscure with peculiar performance profiles. You really have to know your stuff to read a full Phoronix review of a desktop CPU and understand what information is on display. Their server CPU tests are much more concise.

Maybe you need to know your stuff to interpret what any one particular result means, but if you're just looking at the average it really doesn't matter. When it comes to distilling performance to a single number, regression to the mean is real, and sample size is king.
 
  • Like
Reactions: CouncilorIrissa

Joe NYC

Platinum Member
Jun 26, 2021
2,539
3,471
106
There were rumours that V-cache in Ryzen 9000X3D is different.

But nothing concrete in what way it would be different. No clock regression would be the most obvious guess, but that does not make the V-Cache different, it makes the CPU with V-Cache different.

The biggest game changer would be if V-Cache covered the whole CPU, and AMD switched to Wafer over Wafer packaging. This would turn V-Cache models into mainstream, high volume parts. Also, covering the whole CPU die could double the V-Cache size.
 
  • Like
Reactions: lightmanek

Joe NYC

Platinum Member
Jun 26, 2021
2,539
3,471
106
At this point my biggest worry is that AMD already hyped up Zen5 vcache a bit saying it had some “interesting” features

Based on what we’ve seen so far of the core I’d expect it to have even more uplift than Zen4X3D vs Zen4, but if we apply the “reverse marketing lingo” filter, “interesting” doesn’t sound too good….

It would be disappointing if after 4 years, 3rd generation, if AMD still remained stalled at the same place with V-Cache, which is in one of AMD's most promising and differentiating technologies.
 

misuspita

Senior member
Jul 15, 2006
498
592
136
MinisForum has a brand new MiniPC using HawkPoint CPU, and adds some nice IO features.

The advantage of HawkPoint vs. either StrixPoint or Kraken is that it has 8 full cores and there is no need for Windows 11 for improved scheduling of dense cores.
Best one is the beelink, looking nice and very quiet. But yeah, if they manage to cram a Strix Halo in a 15cmx15cm device, I'd instabuy if it's under 1500€ barebone... Sort of, as I understand there's gonna be soldered memory, no CAMM