SPRlover wrote:[...]but I've been searching through the whole internet and I really didn't found any answer. :'(

You didn't find any answer?
I can't believe this - it's totally simple using Gimp (or IrfanView): btw: didn't check for palette data so the ingame colors might be different

I found only at the beginning of the UV data, I cannot calculate the next position of the elements of the UV, can you help me? 0x235c0 looks like a correct beginning; the other uv start addresses might require some fiddeling. 0x28858 looks like a block end - the delta is 0x5298 = 2643x8 dec. sum of...

I used shorts divided by 65535 instead of half floats and the walking anim looks better (a little bit like a spider , but it walks).

Hello guys, I'm brand new here, so I apologise if this is in the wrong section. Yep, it is. Totally. This is a forum for model research , not for model sharing. Luckily I have a Captain from my MakeH2O project on my harddrive. Guess he's not from Disney Infinity, dunno: http://www.uploadmb.com/dw.p...

Can't check it because I don't have paa files.

You could check the python script for blender from Marisuz
viewtopic.php?f=16&t=11849&hilit=.paa
and compare the results with yours.

hex2obj was invented to help people help themselves.

And AceWell alreday showed you how to handle .fml files.

So please ask here: viewtopic.php?f=16&t=13978&p=116120&hilit=+fml#p116120

SpaceDog wrote:Is that possible?

Not with the current version.
(If you uploaded a model sample I could have a look at.)

For hkx files with this header (which I was referring to in my last post): Offset 0 1 2 3 4 5 6 7 8 9 A B C D E F 00000 57 E0 E0 57 10 C0 C0 10 00 00 00 00 0B 00 00 00 WààW.ÀÀ......... 00010 08 01 00 01 03 00 00 00 02 00 00 00 00 00 00 00 ................ 00020 00 00 00 00 4B 00 00 00 68 6B 5F 32 30...

Dunno about the tools but the mesh format (in .chunk) is rather simple:

Use big endian. Vertex start for example at 0x8E4, count: 123 gives some kind of point cloud.
Find a suiting face index block - the one starting at 0xB8 with 1041 FIs doesn't seem to fit (results in a vertex count of 278).

Does anyone know any solution for this or another extractor that might work? Mean a fully featured extractor, don't you? :D Can't serve with that. Also there are not too many coders who care for maps. (It's not a pure map file, is it? Found bone names in it, for example.) Anyway I took a look using...

void build_TriList (WORD VertexCount) { int faceDir, startDir= -1 ; WORD a=0, b=1, cnt=0, f1,f2,f3 ; f1 = a + 1 ; f2 = b + 1 ; faceDir = startDir ; do { cnt++ ; f3 = cnt ; faceDir *= -1 ; if ((f1!=f2)&&(f2!=f3)&&(f3!=f1)) { if (faceDir > 0) fprintf( stream, "f %d %d %d\n",...

Autocreated face indices such as f 1 2 3, f 4 5 6 don't produce good results: (using tri strips it looks even worse)

(I could try something like f 1 2 4
f 3 5 6
or something like that but to be honest, I'm lacking in some motivation for such.)

yeah, this forum is full of posts like this one: "please someone can help me?" But what do you guys do once you've being helped? Why not use the infos you were given to proceed for yourself? This is what the helpers (me at least) expect in a research forum. Try to gain some basic coding sk...

use forum search, please:
viewtopic.php?f=16&t=12804&hilit=stadium

Have a look at my last post there.