Before anyone suggests the 3DS Max plugin, yes, I know it exists but I do not have the time or hard drive space to find a way to get that working so I'm using my sometimes reliable Advanced Mesh Reaper.

I've been at these just to get a few prop models and I've been stumped on one type of model, I was able to figure out the mesh for obj_cheesegorgonzola but other models like obj_autochefhead are stumping me (all the verts are strangely in a blocky formation without having a shape to it), I've tried changing the stride, data type and even switching from little to big endian and nothing seems to work.


I've attached both example files so someone could possibly figure out what's going on with this.

WGGA_d3dmesh_examples.zip

7 hours ago, Hell Inspector said:

but other models like obj_autochefhead are stumping me (all the verts are strangely in a blocky formation without having a shape to it),

I see a shape, but can't make sense of it atm:

Edited yesterday at 01:18 PM1 day by shak-otay

11 hours ago, shak-otay said:

I see a shape, but can't make sense of it atm:

Pretty sure that's the UV map, here's a screenshot of the model in-game, it kind of matches up with the shape in the screenshot I originally posted but those verts seemed off.

According to https://github.com/WeaselOnaStick/TelltaleGames_D3DMesh_Importer/blob/25f2e4739d4deaeefb7cd783fcc9b7abce5ba3f7/Original%20Scripts/TelltaleGames_D3DMesh.ms#L1941
if number of vertices is > 768 they are compressed - you can see bellow reading 256 of min and max values and then for each vertex 4-byte value is unpacked to actual vertex position

2 hours ago, other1 said:

According to https://github.com/WeaselOnaStick/TelltaleGames_D3DMesh_Importer/blob/25f2e4739d4deaeefb7cd783fcc9b7abce5ba3f7/Original%20Scripts/TelltaleGames_D3DMesh.ms#L1941
if number of vertices is > 768 they are compressed - you can see bellow reading 256 of min and max values and then for each vertex 4-byte value is unpacked to actual vertex position

I see, do you know if there's a certain process for working with compressed models?

That link is exactly part where vertex buffer is handled and it's readable enough to reproduce it

Ofcourse if all you have is software for quick model research you won't be able to get results - you need to code a Noesis, Blender or some other script/program to get your models