The msh and mgn wont work together, and I probably wont be focusing on it for a while. Still need to hammer out this lighting issue and the rest of the SHT file as it relates to MSH files. Now as for helping you figure out from what I already know, that I can do. How I export the MSH models is that I pretty much ignore the model and focus entirely on the UV map. Now this could be why I have lighting issues and if so, will have to be tweaked in the future. But it gives the user the greatest flexibility when designing. Now what you are going to run into is that you can move uv maps around all you want, but as soon as you create or take away verts, you are going to have an issue. Thus why you are probably going to need an exporter, not just to pull the info out, but to recreate what you see into a proper MGN file. Think I spent like 6hrs studying a cube, trying to figure out how in the world 8 verts and 12 faces, and a texture went together. The good news is that fatducks importer does import mgn into 3d max, and if we knew what needed to be pulled out and how it was organized, we could do that and assemble it with Tre explorer until a proper exporter comes along. The bad news is that I know nothing about MGN right now, but thats how I started off 6 weeks ago with the MSH.
Aigik- I put on your suggestion. It was one line of code and probably one of the easiest things I have done, thats why 1.32 is up. Also if you put the textures into the uv editor and do not collapse the stack, it will pull them into the sht creator.
This will probably be the last update for a few days. I ripped the effects out of the SHT files and am going to try and correlate them with the texture codes. If it does pan out, I will add around 85 effects and filter them depending on which texture codes are chosen. This should limit crashing from choosing the wrong effects, but will take some time to code in. Lets see... Im at 1,831 lines of code right now. Any guesses what it will be after Im done?
Texture stretching- I will probably have to hand map the uv coords from a MSH model to figure out whats the deal. What I do know is that the original MSH files did not usually use the same UV coords for each channel. May or may not be related, but I will look into it.