AMD Execution Thread [2023]
- Thread starter Kaotik
- Start date
- Status
"The YouTube channel RedGamingTech"
Wait, he thinks developer support is the easier way? I cannot imagine how, especially if you don't dominate the market share. Even Reflex support is limited.
Also, I would like to have the lag-reducing feature control globally via a single switch, without checking game menus. I really hope the "hack" way will prevail.
I think it's far more likely to have developers integrate it into their games than AMD investing resources to do it for them.
DavidGraham
Veteran
AMD high end consumer/professional/AI GPUs are getting banned in China as well.
wccftech.com
Dell Prohibits Sales of AMD Radeon RX 7900 XTX, 7900 XT, PRO W7900 & Upcoming MI300 GPUs In China
AMD Radeon RX 7900 XTX, 7900 XT, PRO W7900 & MI300 GPUs sales are now prohibited by Dell in China following more aggressive US bans.
wccftech.com
Can you explain why?
Gfx12 (RDNA 4) appeared in LLVM
github.com
This usually means about 6 months until release if previous gens are anything to go by if I'm not mistaken
[AMDGPU] Define new targets gfx1200 and gfx1201 by jayfoad · Pull Request #73133 · llvm/llvm-project
Define target names and ELF numbers for new GFX12 targets gfx1200 and gfx1201. For now they behave identically to GFX11.
github.com
This usually means about 6 months until release if previous gens are anything to go by if I'm not mistaken
AMD don't really have a history of investing in developer support.
Which is why I want this injection method to work.
https://videocardz.com/newz/amd-nex...-up-to-64-venice-zen6-cores-pcie-gen6-support
AMD Next-Gen EPYC-E to feature up to 64 “Venice” Zen6 cores, PCIe Gen6 support
Bondrewd
Veteran
Not in relevant HPC workloads where APU advantages are nice to obscene in some cases (CFD, as you've seen, is particularly egregious).
How does it stack up against nVidia? That's the big question.
I was talking about availability
Bondrewd
Veteran
That's just a question of what cloud guys want and it's 300X.
DavidGraham
Veteran
According to AMD, it's basically on par with H100. Which probably means it actually isn't. Also, NVIDIA has H200 now, which is even faster.
AMD showed that MI300X does win significantly against H100 in select cases, when the H100 runs out of memory, as H100 has 80GB, while the MI300X has double that: 192GB. However, H200 has 140GB, which means it fixes these cases.
MI300 is much more expensive and much more complicated to make than H100, MI300X is more power hungry too.
- Status