It's easy to rip them using a hex editor and Hex2obj, since they are just a group of .obj (not wavefront) models, but the process is so time consuming and tedious it took me two days to rip SOME of the file (not even quarter of it).

Anyway i can speed this up ?

track03_01.zip

this is what i mean by .obj btw

You'll need to be a coder to speed it up, imho.

Search for 000001000200 (found 824 times) , get the vertex count as explained in the other thread, get the start of belonging vertex block.

addresOfVertices= addressOfFIs - vcount x 36 -20

So for first picture:  18b887= 0x18be3b - 36 * 40 (decimal) - 20 (dec.)

 

The mesh format itself appears to be simple:

Edited April 21, 20242 yr by shak-otay

On 4/18/2024 at 4:52 AM, MaxBOOST said:

It's easy to rip them using a hex editor and Hex2obj, since they are just a group of .obj (not wavefront) models, but the process is so time consuming and tedious it took me two days to rip SOME of the file (not even quarter of it).

Anyway i can speed this 

so as not to parse the format and write an unpacker (since .obj uses a crude hack). you can simply set up a script for all *.obj inside *.clu, although this has already been done and fixed by the plugin. (update, add *.clu support) fmt_obj.py

Edited April 21, 20242 yr by Durik256

THANK YOU BOTH SO MUCH !
you have no idea how much time you saved for me !😅