Announcement

Collapse
No announcement yet.

A Lot More AMD RDNA4/GFX12 Graphics Driver Code Has Landed In Mesa 24.2

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • qarium
    replied
    Originally posted by Panix View Post
    https://www.cgdirector.com/amd-radeo...-w7800-review/
    The AMD Radeon Pro W7900 / W7800 might lack the horsepower to truly rival the Nvidia RTX 6000 Ada, but they can compete on price/performance

    I guess every site is biased towards Nvidia? Do you know any Blender professionals (other than some Linux users on here?) who pick an AMD gpu over Nvidia? I don't and haven't discovered any on any website or link - other than ppl picking it to use on Linux - even then, I think it's a really small sample/minority. Again, not because they prefer or want an Nvidia (workstation) gpu.
    you clearly lag unterstanding of the industry. all benchmarks you have ever seen is with dataset who fits in the 24gb vram of a 4090...
    but this is not how the industry works. believe it or not big kino/hollywood-style movie production companies buy CPU clusters with 512GB ram or even more and render it on the CPUs.
    there are 2 rationality behind this first point is OptiX is not an option because the output is not the same and they invest a lot of money to make sure the result is always the same.
    second point is they can not even buy 48GB vram cards because their dataset do not fit in 48gb...

    you really do not unterstand that performance is irrelevant for them if the final product renders 2 weeks on their server cluster or 4 weeks makes litterally no difference even this means 100% higher performance they just wait the 2 weeks for the result.

    the 48gb vram nvidia GPUs are not a option because their data set do not fit in.

    Leave a comment:


  • Panix
    replied
    AMD's new RDNA3 PRO GPU line-up, the W7900 and W7800, bring some very interesting new developments to the table. Is this heralding the downfall of Nvidia's Pro GPUs? A closer look.

    The AMD Radeon Pro W7900 / W7800 might lack the horsepower to truly rival the Nvidia RTX 6000 Ada, but they can compete on price/performance

    I guess every site is biased towards Nvidia? Do you know any Blender professionals (other than some Linux users on here?) who pick an AMD gpu over Nvidia? I don't and haven't discovered any on any website or link - other than ppl picking it to use on Linux - even then, I think it's a really small sample/minority. Again, not because they prefer or want an Nvidia (workstation) gpu.

    Leave a comment:


  • qarium
    replied
    Originally posted by Panix View Post
    LOL! You AMD shills are something else. The A6000 is older but still wipes the floor with crappy amd hardware - which, btw, has not shown any progress in Blender support. If a professional Blender user is going to buy a workstation card worth several thousands and needs 48gb of vram - they're probably going with nvidia anyway - an A6000 is still better than the AMD gpu.
    AMD's has released the Radeon PRO 7000 series of graphics cards featuring up to 48 GB of VRAM. How do they compare to NVIDIA's RTX Ampere and Ada GPUs?

    It's unfortunate but reality - I am looking at performance and objective results - not shilling for nvidia or amd - sure, you zealots who pick amd regardless of performance, that's fine but don't claim it is better (performance per $ either). The A6000 probably isn't much more.
    pugetsystems . com​ is known for biased benchmarks and also they spread cyber infections over their websites.

    also you clearly did not check any prices on Price search engines​ because the minimum price of a A6000 is 4.898€ https://tease-shop.de/PNY-NVIDIA-RTX...R6/11113524000

    the AMD PRO w7900 is at 3416​€ https://www.proshop.de/Grafikkarte/A...aign=pricesite

    its not clear how this does annihilate my argument the nvidia card is still more expensive for 48GB vram.

    the price per performance for blender is still better on AMD side.

    Leave a comment:


  • Panix
    replied
    Originally posted by qarium View Post

    well just compare 48GB vram cards with other 48GB vram cards then you will discover that the AMD card costs 3300€ and the Nvidia cards 8000€...

    if you translate "Sucks in Blender" into: Sucks per Dollar then AMD wins big.

    but sure you can freely and openly spend the 8000€ and claim you are superior.
    LOL! You AMD shills are something else. The A6000 is older but still wipes the floor with crappy amd hardware - which, btw, has not shown any progress in Blender support. If a professional Blender user is going to buy a workstation card worth several thousands and needs 48gb of vram - they're probably going with nvidia anyway - an A6000 is still better than the AMD gpu.
    AMD's has released the Radeon PRO 7000 series of graphics cards featuring up to 48 GB of VRAM. How do they compare to NVIDIA's RTX Ampere and Ada GPUs?

    It's unfortunate but reality - I am looking at performance and objective results - not shilling for nvidia or amd - sure, you zealots who pick amd regardless of performance, that's fine but don't claim it is better (performance per $ either). The A6000 probably isn't much more.

    Leave a comment:


  • qarium
    replied
    Originally posted by Panix View Post
    So? It still sucks in Blender despite more vram and it being expensive...for what?
    well just compare 48GB vram cards with other 48GB vram cards then you will discover that the AMD card costs 3300€ and the Nvidia cards 8000€...

    if you translate "Sucks in Blender" into: Sucks per Dollar then AMD wins big.

    but sure you can freely and openly spend the 8000€ and claim you are superior.

    Leave a comment:


  • Panix
    replied
    Originally posted by qarium View Post

    benchmarks already show that vram is king in blender even on nvidia cards. even with similar compute power cards with 12gb vram do better than cards with 10gb vram. and cards with 24gb vram do better than cards with 8gb vram.

    this has exactly zero to do with AMD... by the way i have a AMD PRO w7900 with 48gB vram you see vram size sell cards if you want it or not.
    So? It still sucks in Blender despite more vram and it being expensive...for what?

    Leave a comment:


  • qarium
    replied
    Originally posted by Panix View Post
    LOL except Blender doesn't love AMD and AMD doesn't love Blender - so mutual. Even if AMD put 48gb or more on their gpu, it will still suck eggs in Blender.
    benchmarks already show that vram is king in blender even on nvidia cards. even with similar compute power cards with 12gb vram do better than cards with 10gb vram. and cards with 24gb vram do better than cards with 8gb vram.

    this has exactly zero to do with AMD... by the way i have a AMD PRO w7900 with 48gB vram you see vram size sell cards if you want it or not.

    Leave a comment:


  • Panix
    replied
    Originally posted by qarium View Post

    you are right its a miracle why AMD does not understand that people expect to tun their LLM/AI on their GPU...
    also vram LLM need much vRAM this means the AMD radeon 8800XT should have 32gb VRAM at minimum.
    my Vega64 is more or less obsolete because it can only run 7billion parameter AI... because it only has 8GB vram...
    so even 16gb vram is not enough the Mistral 8x20B opensource modell expect at minimum 20-24GB vram.

    and its not only AI/LLM people of course also expect Blender and other software to run well.

    Blender also loves a lot of Vram.
    LOL except Blender doesn't love AMD and AMD doesn't love Blender - so mutual. Even if AMD put 48gb or more on their gpu, it will still suck eggs in Blender.

    Leave a comment:


  • Panix
    replied
    Originally posted by timofonic View Post
    AMD!!!

    GPU Computing for all RDNA4 family, derivated and sucessors. Please.
    LMAO. That's pretty funny.

    Leave a comment:


  • qarium
    replied
    Originally posted by timofonic View Post
    AMD!!!
    GPU Computing for all RDNA4 family, derivated and sucessors. Please.
    you are right its a miracle why AMD does not understand that people expect to tun their LLM/AI on their GPU...
    also vram LLM need much vRAM this means the AMD radeon 8800XT should have 32gb VRAM at minimum.
    my Vega64 is more or less obsolete because it can only run 7billion parameter AI... because it only has 8GB vram...
    so even 16gb vram is not enough the Mistral 8x20B opensource modell expect at minimum 20-24GB vram.

    and its not only AI/LLM people of course also expect Blender and other software to run well.

    Blender also loves a lot of Vram.

    Leave a comment:

Working...
X