Discussion RDNA4 + CDNA3 Architectures Thread

[DisEnchantment]

With the GFX940 patches in full swing since first week of March, it is looking like MI300 is not far in the distant future!
Usually AMD takes around 3Qs to get the support in LLVM and amdgpu. Lately, since RDNA2 the window they push to add support for new devices is much reduced to prevent leaks.
But looking at the flurry of code in LLVM, it is a lot of commits. Maybe because US Govt is starting to prepare the SW environment for El Capitan (Maybe to avoid slow bring up situation like Frontier for example)

See here for the GFX940 specific commits

History for llvm/lib/Target/AMDGPU - llvm/llvm-project

The LLVM Project is a collection of modular and reusable compiler and toolchain technologies. - History for llvm/lib/Target/AMDGPU - llvm/llvm-project

github.com

Or Phoronix

More AMD "GFX940" Enablement Work Landing In LLVM - Phoronix

www.phoronix.com

There is a lot more if you know whom to follow in LLVM review chains (before getting merged to github), but I am not going to link AMD employees.

I am starting to think MI300 will launch around the same time like Hopper probably only a couple of months later!
Although I believe Hopper had problems not having a host CPU capable of doing PCIe 5 in the very near future therefore it might have gotten pushed back a bit until SPR and Genoa arrives later in 2022.
If PVC slips again I believe MI300 could launch before it

This is nuts, MI100/200/300 cadence is impressive.



Previous thread on CDNA2 and RDNA3 here

Question - Speculation: RDNA3 + CDNA2 Architectures Thread

Man I have been dying to make this one for a while now. First rumours for RDNA3 are here so new thread time! Just going to start off with this one for now: kopite7kimi on Twitter: "@VideoCardz Ah, I mean a simple mcm design with 10240 cores is not enough. Because the lift from RDNA2 to RDNA3...

forums.anandtech.com

 

[ToTTenTranz]

"soon" -> "in a month"?

Rumors are pointing to a dedicated event in late February.

 

[PJVol]

Yes, full details are coming soon.

Missed «™»

 

[gaav87]

Soo

Now I'm starting to remember some interesting research from years ago.
If elements of this shared L1 paradigm are in RDNA4 that would be very cool indeed.

https://adwaitjog.github.io/docs/pdf/sharedl1-pact20.pdf

Interesting so instead of going full shared l1 cache with its negatives they introduced gl1 buffer to mitigate the interconnect bandwidth/latency hit and data replication ?

 

[Z O X]

Soo

Interesting so instead of going full shared l1 cache with its negatives they introduced gl1 buffer to mitigate the interconnect bandwidth/latency hit and data replication ?

Don't understand why the performance drop for some compute.
Is sampling running always, or on compute switch (e.g. different shaders running if I understand correctly)?
If it's running once, why should there be a drop?

 

[branch_suggestion]

Solving the Dense Geometry Problem

Discover how AMD's Dense Geometry Compression Format (DGF) revolutionizes graphics by compressing complex models for efficient real-time rendering, bridging the gap between rasterization and ray tracing.

gpuopen.com

Hopefully it is accelerated in RDNA4.
And arguably the bigger tidbit is the desire to implement it in APIs, seeing a lot of recent momentum for features to be implemented in future DX12/DXR versions.

 

[RnR_au]

Frank is teasing with a single slide... grr...

https://twitter.com/x/status/1887560644787920961

 

[poke01]

Frank is teasing with a single slide... grr...

https://twitter.com/x/status/1887560644787920961

Annoying

 

[blckgrffn]

Annoying

You gotta admit, if they can stretch out this limelight from nvidias underwhelming launch they are wise to. They need a long runway for the hype train for the greater gamerdom.

Then we can all dump on this thread for my birthday! 😂 (early March)