Super Galdelic Hour (PS2) *.vug / *.vu1 / *.vu0 / *.mot

[R7CrazyCanucks]

UPDATE: I found the solution to fixing the direction of the inverted faces without losing any of the normals! I selected the parts of the model that have backwards winding because of the tri strip triangles, separated them from the whole mesh, duplicated them and keeping them in the exact same spot as the separated mesh. flipped the direction of the separated mesh, add the "Data Transfer" modifier to the separated mesh, checked "Face Corner Data" and "Custom Normals", set the source object to the duplicated mesh, apply, and then delete the duplicated mesh and join them back together. Doing this properly fixes the direction of the original vertex and face normals (blue lines) without tampering with the vertex-per-face normals! (magenta lines) In addition, I can also use the "rapi.rpgSetOption(noesis.RPGOPT_TRIWINDBACKWARD, 1)" function in the Noesis script to help me find where the inverted faces are located at.

And after putting all of the mesh together, adding the materials, fixing the normals and UVs of the shiny parts on Toko's outfit, fixing the inverted faces and removing all the doubles in Blender, here we have her full body mesh restored!

Now it's really going to be doable for me now! Now all that's left is to figure out the VU0 model format the character's faces in the f#.vug files and the vug files for Sista and that's it!

Edited October 7 by R7CrazyCanucks

[R7CrazyCanucks]

Would anybody like to help me out with cracking the VU0 model format? I managed to grab all of the character VU0 mesh out from their VUG files, and it's for their faces. The very last missing piece of the puzzle for me in ripping these character models.

The most I can get is just the vertices, normals and UVs. The faces data is the main issue. It's handled very differently compared to VU1 and trying to read this face data (tri-strip) is very inconsistent and different for each VU0.

And weirdly enough, using the PCSX2 3D Screenshot branch, the character faces only comes out flat compared to the VU1 mesh.

VUG-VU0s.zip

[roocker666]

It seems like at 0x24 is vert count(4F8=1272), then at 0x40 is a pointer to vertices(0x70780). After that a line of zeros and then other block of vertices, maybe other facel?(I don't know why are 2 faces). But vert count is taking that line of zeros. I tried to ignore that line and then including that line but is the same result. The only differenece is that large triangle that goes to the floor. So it both cases is a mess, lol. Oh, probably after pointer to vertices, are pointers to Normals(0x44), Uvs(0x48) and vertex colors(0x4C). I don't understand, this is so werird...

[R7CrazyCanucks]

The vertex count seems to read one extra null vertex which is why that big triangle is appearing there. The faces data always seems to be counting up rather than it being set to certain vertices. I'm not sure if there's any internal game code using Ghidra that reads this face data properly. Surely if the PS2 can read it, there has to be a possibility for it to work. Worst comes the worst, I'll might have to find a way to get a proper 3D screenshot of the face and just recreate the face model by hand while keeping its normals and UVs.

[R7CrazyCanucks]

Does anybody have a clue as to what all this stuff inside one of the VU0 files could be used for starting at address 0x6D5A0? This is just before all the vertices, normals and UVs.

[R7CrazyCanucks]

I sure hope that the way the faces are appended is not all game code. I've been trying to reconstruct the face by hand, but it's tough to fix the lips and eyelids.