Description

I use this machine for deep learning as it is part of my research.

It works exquisitely well. I currently have it connected to three monitors (will eventually upgrade to 5 monitors total).

Since I can game on it, I decided to sell my PS4 and strictly game on this machine. I currently play Insurgency: Sandstorm with maxed out settings with a cool 60fps.

I also am re-playing the Mass Effect Trilogy with the latest upgraded texture mods and it looks beautiful.

The only problem I faced was finding the wire that connects the front USB ports to the motherboard. I worked around this by just buying USB extension cables and using the ports in the back.

The BIOS also displays a much higher temperature for the CPU than the utilities I use such as the Threadripper's Ryzen Master, and CAM.

CAM doesn't display fan speed for the GPU, but I suppose that's more of a software issue.

Comments

  • 9 months ago
  • 1 point

in my opinion you should have gotten a vega 7 which preforms just as good in games as a 2080 but out preforms it in work task which it seems like your doing.

  • 8 months ago
  • 1 point

The RTX line is meant for deep learning though.

  • 8 months ago
  • 0 points

Incorrect actually the other way around. Rtx is more toward gamers with less vram and higher clocks. Vega 7 have 16gb hbm2 memory which is clocked higher than gddr6 and also more memory for work task (like deep learning) will 100% help oht with what your re doing.

  • 7 months ago
  • 2 points

Nah buddy. The RTX's have 16bit fp operations and tensor cores specifically designed for deep learning. The only 'gaming' feature is the ray-tracing which isn't widely implemented be devs yet. AMD gpu's don't have CUDA/CUDNN which 99% of the SOTA DL algorithms use.

[comment deleted]