you could simply use hex2obj - view the wavefront obj data in the test.obj file it creates.
The vertex data is also to be seen in the left lower listbox (go2 button pressed): (first submesh only)

Someone who can code such tools is able to get such info by a quick glance into the hex editor.

Anyways, here you go, vertices start at 0xC43 in this fml file: Keep in mind that the FVF size is odd here: 43 (it might be different for different files, 49 for example).

your thread is not forgotten so you don't need to bump, Sir. :D Thing is that I'm thinking about a simple approach to a quick animation test. Simple ways require a lot of thinking in many cases. Since we've a skeleton (to be loaded into Noesis for example) here you could think of creating an animati...

If its possible to make simple vertex edits and then inject it back to the model file,i'll be really happy. same problem as before: you'll need to find someone who's willing to code a tool for you. Also, what does that mean exactly: " simple vertex edits "? As long as the vertex count doesn't chang...

ugly format, ugly models. Who would waste life and time?
I made some ugly patches to get 000l.fml -
although I read only 14 of 26 (33?) assumed submeshes the model looks to be complete (heads are separate I guess)
after turning face culling on:

this is one submesh (of about 16) from c102_win32_trb : c102_win32_trb.jpg (uvpos=8 is wrong here) (FIs count at 0x71B2C, vertices count at 0x7463C) This is a H2O file for another SM: 0xDFD98 1215 Vb1 24 99 0xE0788 539 120400 0x0 255 btw: Noesis' debug log says: "This is not a supported TRB platform...

thanks for your efforts but I won't download about 100 MBs just to get some samples.

sdru contains a mesh, uvs have some extra lines:

that's not a 15 minutes job and I'm too busy currently. You might try 0x55D0 as a FIs' starting address (it doesn't give better results but your's is not the best choice, imho, the pic may explain, why). marge-FIs_skip_26_bytes.jpg The vertices count seems to be 385 here but at 0x5790 we'd have an i...

The Sourge Engine smd importer doesn't like extra blanks at lines' ends: delete the blanks behind "end" and "skeleton" in the following lines and you're done. 80 "Spot01_shadow.Target" 77 end skeleton time 0 You could try out Open Asset viewer which doesn't have the prob with the blanks (right side ...

why not use search function of this forum?

viewtopic.php?f=16&t=12374&p=101672&hil ... er#p101668
(post as of dec. 12th, 2014, 12:00 am)

talking about orc_warrior_base.dat? then, nope, I'm not sure.

But I didn't find any bone names in it. It's a waste of life and time to analyse 3D data without bone names.
So we don't even know the number of bones.

(Seems there's 105 matrices blocks.)

It does not make sense to care for animations as long as you don't have a model with suiting skeleton and working skinning.

possibly. But who would and why?

this model could help, because MMX uses the same models as Heroes 6, but I don't know if bones are named in the same way. Maybe this helps. how should we know if the bones are named in the same way if the don't have an MMV skeleton? H6 skeleton bonenames: nodes 0 "NotABone0" -1 1 "c:OrcWarrior_geo"...