8 Cores and that’s not enough.

AI and CPU churn

I continued with the AI/ML. Since I am not using Jupyter but running from the command line, I could not display an image that the instructor was but only after I trashed PIP in the process. PIP is a package manager for Python. It usually makes it easy to add packages so you can import and use them in projects. I got an error and after actually blasting and reinstalling PIP I finally was able to display the image.

There are almost 170K images so just clicking an viewing the image was not an option. Anyway I followed along and ran the code to a point. It was running quite a while. It was running all 8 cores on my system at around 80 percent. I let it run for over a half hour but wound up canceling it.

I then just watched the rest of the video and found that the guy then ran it in a loop for about 40 iterations. At over a half hour each I decided to cut my losses. The computer he was using was running each at less than 630 seconds. I figured based on past performance of my rig I was not that far off, but I was wrong. I have 16GB RAM it only got up to about half that so memory was not the limit. It is possible that some of the calculations are off loaded to the GPU. I had just upgraded my graphics card so I can only imagine what it would have done with the old one.


I continued with the book Figure Drawing for Comics and Graphic Novels. It has the basics and some good tips, lots of photos and illustrations.

I finally finished the course on using Infinite Painter on Android. It was a whale of a time.