A770 LE vs RTX 3060 12GB TensorFlow Grudge Match



A770 LE vs RTX 3060 12GB TensorFlow Grudge Match

A770 LE vs RTX 3060 12GB TensorFlow Grudge Match

Intel sent me an A770 LE as part of a research program, so enjoy my very long winded examination of its TensorFlow performance vs an RTX 3060.

Documentation I followed to get the card running: https://www.intel.com/content/www/us/en/developer/articles/technical/running-tensorflow-stable-diffusion-on-intel-arc.html

0:00 Intro
0:44 What am I testing
1:35 WSL
3:17 Test Setup
4:18 Why RTX 3060
4:46 Code Background
6:07 First Results, Small Net
7:33 Medium Net
8:59 Large Net
10:30 First Proper Results, Large Net
11:00 XLarge Net
11:36 Limit of A770
11:57 Messy Hair Conclusion