Marc Ecko's - Getting Up: 3d-Models (.msh file)

[Henchman800]

Hey guys, it's me again
Thanks to you guys, and especially to Szkaradek123 for helping me out to rip backyard wrestlings 3d models.
I was able to successfully mod a few characters into THUG PRO and i got most of them waiting to be accepted at vg-resource, so anybody can download them if they want to


Now i wanna do the same with marc eckos getting up character models (PC-Version)! Luckily the texture format (.st) has been figured out already:
viewtopic.php?t=15491
If you all remember albinoleopard and his ps23dformat wiki, you can probaply guess from where i got my spark to try to rip these models
It seems that he figured out start and ending of vertecies and faces! (Copy of the page with lose file documentation in attatchment. Html)
Now i have it a shot with model researcher myself and got a nice pointcloud that showed a character in tpose. So the 3d model file seems to be the same on ps2 and PC, since i used the ps23dformat Information on the pc-files (tranes model files in attatchment aswell).
To be honest i didnt really have a clue what i was doing in model researcher but i was pretty proud of my little progress!
So again i kindly want to ask you dedicated forum members to help me convert the 3d files into something readable so i can mod a little and provide the original files to everybody that wants to have them
Thank you for taking your time reading this.

P.s.: i found some russian blog that seemed to have some getting up models to download:

http://candykond-sergio.blogspot.com/20 ... p.html?m=0
http://candykond-sergio.blogspot.com/20 ... s.html?m=1
Sadly the file is not available for download anymore :-/

Ohh and if you guys wanna get your hands in the games Graffiti-Art, i uploaded it here:
https://www.textures-resource.com/plays ... ure/10280/

Thank you again for listening, if you made it till here

[Henchman800]

sorry for double Post, but it wouldnt let me attatch two files Sendung all this from my phone.
So heres the .msh file

[shakotay2]

using hex2obj (view links in my sig):
.

Trane_Act2-msh.png

It's oddities like such in the face indices list which make the sub meshes hard to detect:

Code: Select all

...
f 705 707 706
f 706 707 2
f 707 1 2
f 2 1 708
f 1 726 708
f 708 726 1
...

[Henchman800]

Sorry for taking so long to respond but i was super busy with work...we are doing the german dub-version of the NFL-games.
Anyway thank you very much for your reply and effort, shakotay2!

As you can see i was able to get a pleasing result with your help and this sweet hex2obj tool. The uvs had to be scaled by -1 on the y-axis and the mesh itself on the x-axis by -1 aswell.
Thanks for helping me till here!

My next question is:
Is it possible to co-search submeshes in hex2obj, too? Or do i have to look for all submeshes and export them as a single object and later combine them in blender?
I want to make a script, that exports all characters

[shakotay2]

My next question is:
Is it possible to co-search submeshes in hex2obj, too? Or do i have to look for all submeshes and export them as a single object and later combine them in blender?

That's the question here: how to find the 2nd submesh (SM). May require some trial 'n error. Getting them as one object produces a spoiled mesh.
From this list I'd guess the 2nd SM should start with f 708 1 726 or f 726 708 708 or something like this.
...
f 2 1 708
f 1 726 708
f 708 726 1
...

I want to make a script, that exports all characters

First you need solve the above mentioned question.

Here's another SM (destruction mesh, whatever) of Trane_Act2:

0x1AB4D 3387
Vb1
16 99
0x14DAB 1254
021000
0x0 255

[Henchman800]

Another night dubbing NFL.....

I ran into a problem. Everytime i try to use your settings for the second submesh and try to save it, hex2obj crashes on me :-/

[shakotay2]

thx for reporting - my bad, seems I bugged something into the last released exe.

well, missed an important line:
pFIs = (BYTE*) pFBuf ;

fixed version hex2obj_0.24e1.zip:

[Henchman800]

I gotta thank you man

So i gave the second submesh a try, to find out that this was the odd mesh i Got out of ninja ripper like 2 years ago...

I wonder what it is....
And interestingly enought its missing the Head and left arm plus fingers like the other model....but why would it make sence to Split a character in 2 submeshes like that? I also noticed that the .msh file for another character (Medium boss) is much smaller.

Spleen_01: 53 KB
Trane_Act2: 120 KB

(Spleen's .msh-file is in attatchment)

(Found this via Google "getting up spleen")
i Got the faceindecies start right with 0x94B9 it seems...nur i cant get the right vertecies start...

[shakotay2]

you can't start at 0x94b9 because of

f 1 2 3
f 2 4 3
f 3 4 5
f 4 6 5
f 5 6 19
f 6 20 19
f 19 20 21
f 20 22 21
f 21 22 29
f 22 30 29
f 29 30 31
f 30 32 31
f 31 32 39
f 32 40 39
f 39 40 41
f 40 42 41
f 41 42 43
f 42 44 43
f 43 44 45
f 44 46 45
f 45 46 1301 <<<<
f 46 1 1301 <<<<

correct start address of face indices is at 0x94ed:
f 1 2 3
f 2 4 3
f 5 6 7
f 6 8 7
...

spleen_01.png

(got a superfluous face again)

[Henchman800]

Thanks for the quick response man. Really appreciate your help on this

I dont get it.....so my faceindecies was one Position too far...same with vertecies: my best guess was 128 but it was 127...


I put the resulting Spleen.obj into blender:
So although spleens file is way smaller he still carries another submesh with the missing fingers and head.
What could be the season in splitting every character in 2 parts like this? Why not head separate from Rest of Body instead?

Thank you very much again for your work and time with this

EDIT:
could it be that tranes second submesh starts directly where the first one ends?

[shakotay2]

EDIT:
could it be that tranes second submesh starts directly where the first one ends?

You need to try it out. There's face indices counts before the submeshes (SM) start. For spleen 1300 precedes the fst one (so it's not 1305 as in my previous post).
For him I got a point cloud only for the 2nd SM.
.

spleen_01-mesh_hex.jpg

(correction: in the above picture end of SM is before the 01)

[Henchman800]

Awesome man thx!

I played around with the settings a little that you provided. 26 seems to be the right face count, since 27 boosts up the vertex count by so much.

But you obviously got the right amount of vertecies in the Shown point cloud (GREAT JOB! by the way). So does it mean that we might have the wrong start of faces?


IMPORTANT NODE:
it seems that Indiana Jones and tomb raider (slayer engine versions) use the same Format .msh
So we can probaply solve the issue for all these games

[shakotay2]

Next face indices' (FIs) block at 0x61e7 + 361x36 (dec.) = 0x94ab
So there's a "wide range of guessing".

The assumed start of face indices usually is NOT the start here as you may have noticed.
You need to look at the hex dump in my post as of Thu Oct 03, 2019 8:51 am to understand where the FIs' start to be located.

Look into the created test.obj files for the assumed start of FIs' block:

f 1 4 5
f 4 6 5
f 5 6 7
f 6 8 7
f 7 8 9
f 8 10 9
f 9 10 11
f 10 12 11
f 11 12 13
f 12 14 13
f 13 14 15
f 14 16 15
f 15 16 17
f 16 18 17
f 17 18 32
f 18 34 32
f 32 34 35
f 34 36 35
f 35 36 37
f 36 38 37
f 37 38 583 <<<
f 38 1 583
f 583 1 689
f 1 690 689
f 689 690 691
f 690 692 691
f 691 692 693
f 692 694 693
f 693 694 695
...

Looks as if 583 is the FI count here. So you need to find that 46 02 (hex of (583-1)) in the hexeditor.
Problem is that the following FIs are bigger than 362. So sad...

But there's other data where it fits:
.

spleen_0xa6d2.png

[shakotay2]

For him I got a point cloud only for the 2nd SM.

well, the trick is to keep the start of vertices of the first submesh:
.

spleen_01_SM02.png

I assume there's still 75 vertices left (688 + 285 +75)

edit: well, I think I've got the scheme now how it's working... (FI count is 144, though)

[shakotay2]

Trane, 2nd SM and bottle belt:
.

Trane_act2-2ndSM.png