UPDATE: I managed to fix the problem with the original files (yeah, you got me started, LOL)... I right clicked some of the settings like in the Materials folder for the Cathie character you have settings like "Cathie All Maps"... you can right click it and select "Browse to File Location" then right click the DUF files in there and you can right click them and... at least with me, i can "Edit with Notepad++", or if you don't have a text editor to open it with, just select "Open With..." and open the DUF with Notepad or whatever text editor you prefer as they're just text files. Search for "//Runtime" and replace the "//Runtime" with "/Runtime" (only ONE "/") as the double "//" is the problem, leave the rest of the path intact of course. You can then resave the file (you will have to do this with all the setting files that have this problem) and you'll be fine. May want to back it all up for future use/installation.
Very nice! I was just watching some good videos on doing underwater scenes today. You create them much like you do any other scene. Hair tends to fly out all over in water when still, and flow backwards when you move. Skin doesn't look "wet" underwater (or hair). You can use wind effects for clothes and hair when under water. Quite interesting. Anyhow, nice job.
I've been testing out Stable Diffusion again.... been like almost a year since I last tried it and it's and it's improved by leaps and bounds..
There's a new extension that basically made me want to try using SD again mainly for Img2Img using my DAZ renders as a base...
You can learn about the new extension(controlnet) here
You must be registered to see the links
I decided to take my old Summertime Saga Roxxy Render and run it through img2img
What the new extension (controlnet) does is create an outline and force the output to conform to it to the shape and/or pose(it does more than that with different models)
With that, I tried the same prompts with different models:
What I will try next is make a bunch of renders of a character of my choosing and create a LORA model and img2img with halfassed renders of the same character to make stylized version that can pose in any way that I want..