Description

This is a build to meet the needs of Machine Learning, Deep Neural Networks research leveraging Nvidia's CUDA.

One of the requirements was that it be under 2k. Another was it needed to be able to support up to 4 GPUs but start with one to hit the cost requirement.

Note: http://timdettmers.com/2015/03/09/deep-learning-hardware-guide/

Log in to rate comments or to post a comment.

Comments

  • 42 months ago
  • 1 point

Running cool, quiet and fast. Loving it.

  • 42 months ago
  • 1 point

Hows the ram? +1 For that water at the end ;P

  • 42 months ago
  • 1 point

Having 64 is awesome - haven't hit /tmp yet. :)

Fresh Mint Juleps FTW.

  • 42 months ago
  • 1 point

Runs super cool and quiet - pretty jazzed. When I start using the graphics card hard during number crunching it spins up a little, but stays pretty cool. Never used a water cooler before, I'm sold.

  • 41 months ago
  • 1 point

I've been thinking about building a desktop (my first!) for use in Scientific Computing and ML. Not having much of a clue and trying to adhere to a budget of $1250.00, I chose some similar but less expensive parts, then I happened upon this...

I'm wondering if you have any suggestions on how I might reduce the cost of this a bit without sacrificing expandability?