Hello,
What card do you currently have?
Github has a post for the update 14th of April, the specs are "GPU was NVidia GeForce 840M 2Gb" for the PC. So, it doesn't need a "new" card from the looks of things. Its got some hints and tips, trouble shooting etc in it as well.
Listed prerequisities listed are
"Prerequisites:
Python and Tensorflow, preferably Tensorflow-GPU if you have an NVidia GPU. In this case you will also need CUDA and CuDNN libraries.
I tested it in Python 3.6.3 (Anaconda) + TensorFlow-GPU 1.4.0
Environment: I used Win 10 + Cygwin
GPU was NVidia GeForce 840M 2Gb, compute capability 5.0, CUDA version 9.1"
Also try Cloudnights, they will have a discussion on there somewhere.
Steve
https://github.com/nekitmm/starnet#readme
Its been a few years since I bought a GPU, old rule of thumb would have been $200 is a highend entry level card. $300/350 is a good mid-range entry card, but with world supplies etc, god knows.