• 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 RDNA 5 / UDNA (CDNA Next) speculation

Page 118 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Win2012R2 said:
This was first shared here like a week ago.
Click to expand...
I guess with 118 pages of reaction a double post could happen.
Win2012R2 said:
Nobody outside AMD knows how it will improve perf, and if I had to guess AMD does not know for sure either, their last attempt was pathetic perf wise, made Ampere look good.
Click to expand...
Looks like Coelacanth's Dream knows where he is talking about .... just saying.🤷‍♂️ But only after release of RDNA 5 and reviews by al mayor PC hardware reviewers we will know the real deal.
 
del42sa said:
RDNA6 going to have unified rentable units, with SMT4 like operation and will quadruple performance and brand new Primitive Shader² ® with advanced Radiance AI Cores ® which going to make Jensenn begg for mercy 😎 😛

approved by MLID ®
Click to expand...
lol I could be in a random meeting and anytime someone says anything that sounds close to SMT I always think of this forum and SMT4...

Then I get weird looks because i'm smiling to myself.
 
Fun fact, one of the effects of the memory shortage is that even SMT2 is now being widely disabled.

When running typical server loads, each instance you run requires the same amount of working memory regardless of how fast it is running. If you have SMT, the throughput per core is higher, but the throughput per gigabyte of memory is lower, because tasks take longer to complete. So by turning SMT off you slightly reduce the efficiency of cpu use, but improve the efficiency of ram use. And guess which one is more expensive right now.
 
Tuna-Fish said:
Fun fact, one of the effects of the memory shortage is that even SMT2 is now being widely disabled.

When running typical server loads, each instance you run requires the same amount of working memory regardless of how fast it is running. If you have SMT, the throughput per core is higher, but the throughput per gigabyte of memory is lower, because tasks take longer to complete. So by turning SMT off you slightly reduce the efficiency of cpu use, but improve the efficiency of ram use. And guess which one is more expensive right now.
Click to expand...
Big cores ==> cheap cores 😎
 
olrak29 said:
The rumors always revolve around "Samsung will produce X Product" and ends up always wrong. It got so tiring at this point that it's easy to assume that Reverse Cramer is applicable to this kind of news.
Click to expand...
I just heard it from someone at the foundry. It might be 'Jupiter' -codename?
 
ToTTenTranz said:
Which means PSSR is probably a percursor to FSR4, and it seems likely to me that PSSR will eventually be upgraded to whatever the latest version of FSR4 Sony's QA engineers decide to validate.
Click to expand...
ToTTenTranz said:
PSSR and FSR4 are the same project.
Click to expand...

Here it is, official confirmation of this from Cerny himself.

https://twitter.com/x/status/2034751680290705451





And BTW, here's another missed prediction:
adroc_thurston said:
Yes they're completely separate, unrelated products.
Click to expand...
 
Win2012R2 said:
PSSR2 and now it is FSR4.1 seems to have been very well received, they just need to improve ray reconstruction stuff and DLSS moat is done for.
Click to expand...
That moat is never going away. Devs aren't going back to finished games to correct for AMD's DLL mistake by manually updating the games to support FSR3.1+.
Also zero reason to ensure 100% overlap with DLSS when Radeon's market share is well into the single digits and existing install base is pathetic. FSR4 on RDNA 2-3 won't fix the issue.
 
MrMPFR said:
That moat is never going away. Devs aren't going back to finished games to correct for AMD's DLL mistake by manually updating the games to support FSR3.1+.
Also zero reason to ensure 100% overlap with DLSS when Radeon's market share is well into the single digits and existing install base is pathetic. FSR4 on RDNA 2-3 won't fix the issue.
Click to expand...
Won’t fix but making the last few Radeon customers mad just will make things even worst than they already are.

I think this decision is the final nail in the coffin for the Radeon brand. Rip
 
MrMPFR said:
That moat is never going away. Devs aren't going back to finished games to correct for AMD's DLL mistake by manually updating the games to support FSR3.1+.
Also zero reason to ensure 100% overlap with DLSS when Radeon's market share is well into the single digits and existing install base is pathetic. FSR4 on RDNA 2-3 won't fix the issue.
Click to expand...
From next gen consoles onwards the FSR tech will be in built to the engines. It's a reset on additions of that kind of tech for AMD, especially since Xbox is a hybrid console.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top