interesting, but the thing most of us probably need is just how to get started, I mean just loading the simplest image was impossible for me, and I ran out of my 15 minutes free time before I had even done anything.
Love your lecture Rob. Hope you will do more like this. I am beginner myself in comfyui and ai graphic in general there are some points that's not very clear.. Anyway I plan to watch all of your channel videos 😅❤ Greetings (from Thailand.)
yeah i guess in this entire tutorial i learned one thing that do not add too much negative prompt anyway i am gonna manipulate your workflow... If it work i will share back with you
Thanks, it helpful to get back to basics. One thing that has evaded me thus far is how the cfg fits into the puzzle. Using this workflow as a test bed could be a ggod start
CFG is a big deal with SD1.5 but less so with SDXL and hardly at all with Turbos and Lightning models. High CFG means it will ignore your prompt, very low CFG means it will follow it more exactly. Too low and the AI is too constrained and you will get poor quality. For SDXL about 7 is a good place to start, for Turbos 2 and below.
