• 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 171 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
igor_kavinski said:
If that app isn't very CPU intensive and not a multimedia app, it should run fine in a VMware VM of MacOS. It's easy to find pre-built VMDKs for those if you look around.
Click to expand...

I'm playing with those. At this point, they're either too slow or too hard to make work on AMD or prone to graphic glitches or unable to got past 1027*768 definition (ouch!).
Not gonna do it.
 
Hitman928 said:
Good thing they continued to invest as the AI boom (LLM in particular) could potentially be their most profitable market ever, at least for as long as that market can sustain itself.
Click to expand...
AI: yes. LLM? Hopefully not. There are a number of non-LLM models that are showing a ton of promise, including one in development by Meta.

I am hoping that we get beefed up GPGPU compute engines (that can be used for any workload) for all CPUs moving forward. Apple is doing things right in this regard. Intel Meteor Lake and AMD mobile products also do okay.

Zen 4 desktop? Not so much.
 
eek2121 said:
AI: yes. LLM? Hopefully not. There are a number of non-LLM models that are showing a ton of promise, including one in development by Meta.

I am hoping that we get beefed up GPGPU compute engines (that can be used for any workload) for all CPUs moving forward. Apple is doing things right in this regard. Intel Meteor Lake and AMD mobile products also do okay.

Zen 4 desktop? Not so much.
Click to expand...

LLM is what the current boom is all about and what AMD is both really competitive in and trying to capitalize on. Of course the future can bring a lot more prospects, but in the near term, LLM is where the money is for AMD. As far as what Meta is working on, I'm not sure what you are referring to. The thing they keep talking about, at least that I see, is Llama, which is their own LLM. Is there something else they've been promoting outside of that?
 
TESKATLIPOKA said:
Strix Halo will be the most costly APU they have for the client segment, thanks to IOD.
I don't think they will throw extra cache on top of that, when It's aimed for mobile, which doesn't have high margins.
Click to expand...

V-Cache is about $10 cost for $100 MSRP increase, so a very profitable proposition for AMD.

I think AMD will broaden the scope of Strix Halo to small form factor / Mini PCs, which in turn broadens the scope (target market) of Mini PCs to much higher performance.

If the graphics performs on level of Radeon 6600, then it will be good enough for a very large percentage of games.

Aside from the First Person AAA games, pretty much all of the strategy / simulation / survival / base or city building games all have hardware requirements that below Radeon 6600.

And, BTW, these games above I mentioned like the V-Cache.
 
eek2121 said:
AI: yes. LLM? Hopefully not. There are a number of non-LLM models that are showing a ton of promise, including one in development by Meta.

I am hoping that we get beefed up GPGPU compute engines (that can be used for any workload) for all CPUs moving forward. Apple is doing things right in this regard. Intel Meteor Lake and AMD mobile products also do okay.

Zen 4 desktop? Not so much.
Click to expand...

But the reason to get the desktop PC is to get a dGPU, which will have more power to run any of these models that any notebook chip.
 
Joe NYC said:
V-Cache is about $10 cost for $100 MSRP increase, so a very profitable proposition for AMD.

I think AMD will broaden the scope of Strix Halo to small form factor / Mini PCs, which in turn broadens the scope (target market) of Mini PCs to much higher performance.
Click to expand...
OEMs won't be willing to pay that much, AMD would have to sell It much much cheaper.
 
I know this has been discussed to death but does it seem like the author is saying expectation should be tempered?

chipsandcheese.com

Zen 5’s Leaked Slides

A YouTuber called Moore’s Law is Dead recently leaked a couple AMD slides about Zen 5.
chipsandcheese.com chipsandcheese.com

"AMD is prone to this because they’re an underdog that people expect to one-up its bigger competitors, so fanciful rumors get a lot of attention. Whenever Zen 5 comes out, I would encourage everyone to look at its performance with respect to how Intel and other CPU manufacturers are progressing, and not based on rumors."

So what I can deduce from reading the article / reading between the lines: ipc uplift from zen 4 to zen 5 should be greater than that of rpl to arl. Still, something is very off about zen 5 rumors and the slides.
 
adroc_thurston said:
Yes, clam loves saying that.
Click to expand...
Which is kind of funny because of their wild claims about zen 4 and zen 5. Different author but I'd doubt they didn't discuss this before publication.

chipsandcheese.com

AMD’s Past and Future CPUs (Formal Retraction)

Editor’s Note: With the release of our Zen 4 article, I issued a formal retraction of this article.
chipsandcheese.com chipsandcheese.com

"I was told from a trusted source that a Genoa engineering sample (Zen 4 server chip) was 29% faster than a Milan (Zen 3) chip with the same core config at the same clocks. Factor this in with what I have heard about the possible clock gains that N5 will enable over N7 and Zen 4 sounds like it is going to be a monster of a CPU.

Now I said I had Zen 5 info, unfortunately this comes from a different, less-proven source than my Zen 4 info, however they have said that the jump [to Zen 5 from 4] from will be about as much as Piledriver to Zen 1 design goal, which if you recall to earlier in this article was 40%. I was told from a 3rd source that Zen 5’s original design goal was 2.5 to 3 times the IPC of Zen 1 which roughly lines up with the perspective of a “Piledriver to Zen 1”-like jump."
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top