Building a 50 Teraflops AMD Vega Deep Learning Box for Under $3K
Carlos E. Perez
1.2K17

I’m also in the process of spec’ing out a ryzen 7 deep learning box- however sticking with Nvidia 1080ti due to better support for GGPU libraries, unless dual RX Vegas becomes more enticing in a few weeks (and tensorflow-gpu gets supported) — and the one reason i haven’t hit the ‘confirm purchase’ button yet.

My build is similar, except with 1700x, 2*16gb 3200 G.SKill, 500gb Samsung EVO 960 NVMe for the OS, and 2T Seagate FireCuda Hybrid drive for storage. If I add SLI later on, i just need to upgrade the PSU.

What interests me the most is (moving away from MacOS ) is the new Windows 10 “Windows Subsystem for Linux” (https://msdn.microsoft.com/en-au/commandline/wsl/about) , which means one main OS, with ‘native’ Ubuntu/Bash familiarity, that can also do VR and games (low priority- but nice to have), and no need to dual-boot or use VM’s.

updating and maintaining drivers, applications, libraries becomes cumbersome when you have to deal with more than one OS.

Like what you read? Give Adam Ezekiel a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.