I grew up in Virginia and my father who was an amateur sculptor was a major influence in my love of art. I took a few art classes in college, but at that time the university had one mainframe computer and there was no such thing as a personal computer. I became enchanted by the wide open spaces and colors of the Southwest and moved to New Mexico in 1988. I became aware of the possibility of making art on the computer in 1993, but it was the late 1990s when digital art became my passion and has been ever since. My day job as a psychiatrist is demanding but my art is on my mind every day. Music, especially heavily rythmic music (e.g. Mickey Hart), helps me get into a trance-like state exploring color, form, light, texture, and composition to bring to life my version of beauty. I have put together an AI setup with Stable Diffusion 1.5 that is very powerful for my purposes. I start with a picture I have made and use the image to image function to make a new image. The text prompt goes through 3 separate levels where I have trained the AI on ShivaFooL artwork: the main multi-gigabyte model or checkpoint file, embeddings referenced in the text prompt, and Loras. Then the work in progress goes to IPAdapter where it is instructed to use another one of my original artworks to modify the starting image. Using my AI art setup is an active process working with about 10 different adjustable parameters and the text prompt. I generate many output images, refining the output as I go. From the images produced I start to see the picture I want to make, and put together elements from multiple images for the new picture. The elements may be from just a few images or up 20. I use my Photoshop skills to bring the elements together into a cohesive picture. Often there will be painting I need to do with things the AI got wrong or changes that are called for by the picture I’m making. My process uses 2 of my pictures in a system that has been highly trained on years worth of ShivaFooL art