Having some experience with Intel + Nvidia and Tensorflow, I would love to transition to AMD. This is in part because I am biased towards open source, and I want to invest in new machine learning hardware but don't feel like spending 3k on Nvidia-oriented gear.

The machine learning will be oriented towards Tensorflow for voice and sound generation with Wavenet.

So I'm looking into assembling a new desktop setup of one (or perhaps more) of AMD's GPU's. Anything equivalent to (or better than) the gtx 1080 ti seems like worth the time spend to get it to work.

I am hoping to get some pointers on what to look for, what works (aiming at using Tensorflow). Perhaps even complete recommendations on components that work well together.

Any tips are very welcome, many thanks!

  • Do you have any requirements for this GPU? How are you going to use it? Why do you find the Radeon TX Vega 64 interesting? Why haven't you purchased it yet? – Cfinley Mar 13 '18 at 13:56
  • Thanks for your questions @Cfinley, I've edited my question to answer most of them (deleting the unnecessary Vega pondering). Regarding why I haven't purchased it yet: I'd rather ask for advice based on experience prior to spending 2-3k :) – nick88 Mar 13 '18 at 14:03

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.