• 0 Posts
  • 25 Comments
Joined 3 years ago
cake
Cake day: June 1st, 2023

help-circle








  • 7900 XTX recently got support for Stable Diffusion and LLM, on paper, it’s faster than 4090 RTX for FP16 computation, it does seem faster judging my experience using rented 4090 RTX on Runpod and my 7900 XTX GPU. 14 seconds (4090 RTX) vs 6 seconds (7900 XTX.)

    7900 XTX is an option if you want $1000 cheaper than 4090 RTX and have similar sized VRAM and having comparable performance to that of 4090 RTX.






  • Servers that run the MMO with dynamic persistence would likely need GPUs for keeping track and computing vectors/matrices for each player/asset.

    I always theorized that if Eve Online developers actually bothered, they could utilize their GPU on the servers to track all of the activity in space and they would easily support 10,000+ players on the same system without issue or time dilation (considering the capability of GPU today.)

    Assuming you have for an example, a GPU Server with 8 slots for GPU, and each GPU is a 7900 XTX and you use Vulkan Compute for it. You got 192 GB of vRAM and absurd amount of computation capability with ~1 PFlops on FP16 and 528 TFlops on FP32. Assuming you have 100,000 players on the same system, that GPU Server would be able to handle it with ease even with theoretical complication that can results from programming on GPU. You could store ~2MB worth of data about each player’s ship for 100,000 players on that GPU Server alone.

    Dynamic Persistence in MMO is basically Server with GPU that keep track of everything. (You can’t just defer GPU operations to the end-user, server have to be the one to do it.)




  • Just a head up to be careful with 7900 XTX if you do plan on getting an AMD GPU like other people on here are suggesting.

    When I purchased 7900 XTX, AMD doesn’t offer me any way to control the fan speed on 7900 XTX and it always get stuck on 5% speed. I literally tried everything from using hwmon mode setting to manual (it stuck on auto and refuse to switch to manual), literally modifying the AMD GPU driver in kernel to forcibly set the manual mode for fan speed, it doesn’t work and instead it locks up the Kernel, and tried literally every application that exists for setting fan speed on 7900 XTX.

    I tried to contact the manufacturer to refund me, they refused to pay me back in full and want me to reduce what I get back, I paid $1000 for it, they want me to pay $100 shipping and to only be qualified to receive $400 from them. I ended up keeping the 7900 XTX and basically went nuclear on fixing the GPU. This was literally within 1 week of receiving the GPU mind you. AMD is ranked far below Nvidia after my absurdly negative experience with them and I would rather go with Intel than AMD at this point and that is saying a lot, because it’s not only my GPU that is a problem, but it also with their software and driver like ROCm that NEVER worked, ever.

    I created a plastic strap via 3D printing on top of the GPU and create a negative pressure fan to cool it down, it can stay under 50 degree Fahrenheit at 100% utilization.