Python mesh converter for mounts and siege engines

[KnightErrant]

Are you using the old python meshconverter or GOAT on these?
Just wondering if it makes a difference. Now that it isn't the bump
file or the overlay file, I'm leaning back to grasping at the tangent space
basis being incorrectly computed by the old converter.

[AlphaDelta]

I'm using the python mesh converter (mount to mesh).

GOAT gives me this error when I try to convert the culverin from ms3d to mesh:

Read ms3d group comment section
Read ms3d material comment section, number of comments = 0
Read ms3d joint comment section
Read ms3d model comment
Read ms3d second subversion number
Read ms3d weight section

This ms3d file has self.isregular = True
Exception in Tkinter callback
Traceback (most recent call last):
  File "C:\Python25\lib\lib-tk\Tkinter.py", line 1403, in __call__
    return self.func(*args)
  File "F:\Games\Medieval II Total War\mod_work\tools\GOAT\GameObjectApplication
Toolbox.py", line 2144, in ms3d2meshConverterButton
    ms3d2meshConverter( ms3dfilename )
  File "F:\Games\Medieval II Total War\mod_work\tools\GOAT\GameObjectApplication
Toolbox.py", line 293, in ms3d2meshConverter
    meshobj.lodname          = ms3dobj.lodname
AttributeError: 'ms3dObject' object has no attribute 'lodname'

[KnightErrant]

Sorry, should have said. Can't mix and match between the two
because the data is stored differently in different comments.
If you use GOAT to bring over the culverin and then convert it back,
do you see lighting anomalies like you do with the old meshconverter?

[AlphaDelta]

Ahh the vanilla culverin.

I converted it to mesh, back again, and tried it. Unforuntately it crashed the game.

[KnightErrant]

Arrgh, AD, you're totally bumming me out. The culverin goes to the top
of tomorrow's TODO list. I was reading up on how the tangent space
basis is used to see if it does influence the luminosity and my best source
says it can somewhat, but I'm not sure this convinces me that this is
the problem. But, if the culverin crashes the game, that has to be fixed
and on the plus side may provide a yes or no answer on the tangent space
basis for luminosity problems. Diagnostic question, I'm assuming we are
only looking at what we'd normally call a lod0 model. Can you tell if
everything looks normal when you zoom out so the higher lods come in?
They should look normal because they've never been touched.
The reason I'm curious about that is that siege_engines are the only
models (besides a few mounts) that use the full tangent basis for all lods.
Maybe that points to something interesting.

(I'm asking non-obvious questions lately because a very clever modeller
found a huge mistake in one of my assumptions in GOAT. I'm learning to
disbelieve ALL of my preconceived notions.)

[AlphaDelta]

KE,

I now have 3 artillery meshes that all have this reversed lighting issue, I can pack them all up and send them over to you. Perhaps some cross-examination might help to reveal the issue?

Diagnostic question, I'm assuming we are
only looking at what we'd normally call a lod0 model. Can you tell if
everything looks normal when you zoom out so the higher lods come in?

Right, I have only worked on LOD 0. LOD 1 and 2 have perfect lighting.

This is my edited serpentine LOD 0 vs LOD 2 vanilla serpentine with the sun behind the camera.



(I switched between the 2 LOD's using the slider in options)

Cheers

[Ataegina]

i've edited serpentine, and only worked the mesh leaving the textures untouched but when i use my reworked model i have massive glintering without having ever touched the textures !

Spoiler Alert, click show to read: 

what to do to fix?