Sample animation should be here (shield bash file) no, is not, there's the py script for Homm6 gobj files and a gobj sample. ( shield_bash.dat is from MMX, isn't it?) Also we'd need the MMX skeleton before we can deal with animations. And as I wrote 4 weeks ago: "how should we know whether the...

sry to interrupt you (see you're using this thread as kinda memo pad :D ) I modified chiri's C# code to directly import your uploaded .dat file: 0b8f_14.jpg There were 23 objects contained, most of them didn't provide proper vertices for some unknown reasons (FIs ok, so far). had to expand this line...

Most confusing thread I've ever read. :D 1. it doesn't make sense to care for animation data as long as you don't have the skeleton, imho 2. give the .toanmb file some structure (search for "EDN_" for example) Offset 0 1 2 3 4 5 6 7 8 9 A B C D E F 0B20 30 80 00 00 CA 02 00 00 3A 01 00 00 ...

the script searches for ko-KR_0-mpk i which you didn't provide lines which (indirectly) causes the error: open FDDE mpki 1 from quickbms.txt : Open FOLDER NAME [FILENUM] [EXISTS] It opens a file for reading, practically it assigns a file number/id to an existent file that you want to use arguments: ...

I don't have that model but it looks like this formats uses relative face indices (FIs) for its submeshes (SMs). So you'll have to find out where the face indices for the next SM start (0x68BF4, other model !). Then you need to calculate the vertices start address for the next submesh: ActorModel_sc...

uvs are not in the FVF block - search for the first 000000FF after the vertex block; behind it the uv block starts:
(other model of Scarlett): btw: your friend is world famous Fatduck?

We know it ! On video i show right animation well, how was this created then? I using blender script to load in blender wrencher.s3d and b3d. The script from Mariusz in blender 2.49b, right? Then i export fbx from blender. Which blender fbx plugin did you use? (The one from Campbell Barton for Blen...

thx - since we don't know how the attack sequence should look like it would be a good idea to test your script with a walk_anim, wouldn't it? How did you chose the sequence of bones? In the wrencher.b3d it starts with hip and ends with fingers_l_lower , so totally different from that in your fbx file.

Nice script for testing skeletons! :) Sadly I can't load your Max scene because I'm limited to Max 2009 now. Could you show your bones hierarchy, please? Mine is like this but doesn't seem to work with your script correctly: 0 "blend_root" -1 1 "hip" 0 2 "tail_upper" 1 ...

I've a deja vue of such - which game it is? cyborg.jpg ok, it's Halo: http://forum.xentax.com/viewtopic.php?f=16&t=12696&p=104431#p104431 also is it possible to extract bones or skinned models? I would have tried but since I've wasted my time again with something I've already done I won't, s...

nasty format; small meshes, most of them with a vertex count of about 200 or less
(autocreation of FI's seems to be useless here)

If it's a PC game I'd suggest to try out a 3D ripper.

Question why using 7 floats numbers for one animation time step ? [...] Four floats for quaternion and three floats for position ?! yep, I'd say so. Where it might be 3 floats for position and then 4 for quaternion. no time to check it atm. There should be functions in the web to check whether 4 fl...

In anm file i see 27 bones and 14 14 frames? Agreed. There's a count (27) at 0x0D but where do you see 27 bones (blocks)? 26*13*16 = 5 408 bytes. a brave calculation, but I fear, it doesn't represent the structure of that anim file. what I found is this, 18 blocks starting at: 0x21, 0x1E9, 0x309, 0...

Although I didn't completely understand when you said: "It's just a matter of building 1-y for the uvs". uv Coords 0.0 0.0 -> 0.0 1.0, 1.0 1.0 ->1.0 0.0 Some people say it's mirroring at the y axis, but it's an additional adding of 1.0 to y, so in the end y -> 1-y but I can't figure out h...