Professor Layton vs Phoenix Wright: Ace Attorney models

[Virt]

So I was hit up with some PLvAA model files. I managed to extract the files within those, but I don't know where to go from there.
Everything that was extracted is filetypes like .atr, .cmn, .mbn, .mtr, .prm, .sil, .txp, and .xi
(Oops, there are some others in other folders, too.)
Does anyone know about these or are they yet to be cracked?

Spoiler

Man, why can't any of these games have normal filetypes?

[Shade]

It would be helpful if you posted some of the files you're referring to.

[Virt]

https://www.dropbox.com/sh/w8zcjp4xictpq...OgCPa?dl=0
Here's a folder full of stuff that was extracted from one file.

[Ploaj]

Super bump, but I started trying to figure out these files, and I'm pretty new to model data, so hopefully someone else with more experience may be able to figure it out from here.

The ".xi" files are images and can be extracted already.
I believe the ".prm" files are the mesh data. They are stored in a fairly basic way, but I can't figure out how to read them.

So here is Layton's body mesh file (000.prm), along with the textures and the already decoded data for the faces and vertices.
The prm basically just has the vertices and faces, so you can just ignore the PRM file.
https://www.dropbox.com/s/nsq8jtrjo5l50s...n.zip?dl=1

To help with reading the data, there are supposed to be 412 points and 1085 faces, according to the headers in the prm.
The face.bin and vertex.bin seem to match up to that amount, so it seems accurate.

Each face value is obviously just a short, but some are repeated in a very odd manor. It seems wrong to have a face made up of three of the same point. Maybe there's a special order to read it in?


The vertex data, I have no idea. There appears to be something like 20 floats per entry or something like that. The math adds up, but I couldn't figure out which ones are points.
The second set of 3 floats seem to be the closest to making something, but the face data is so messed up I have no idea.

Hopefully this will help with figuring out the format and be worth the bump.

[Carpaccio]

Isn't this the same engine as the other 3DS Professor Layton games? If so, would anyone be willing to send me the extracts from those so I can have a look at them as well? I've been wanting to rip from the Layton games for ages now...

[Curious Villager]

Oh, I've been hoping to see some of the Professor Layton models in more detail as I've been trying to figure out how Layton's model was designed in Azran Legacy and Luke in PL vs AA.

I'll try to have a further look into it though.

[Ploaj]

So I finally sort of figured out the mesh stuff.
My script seems to already work on a majority of the files, and once I finished up a few things, it should work on EVERY model in the game!

Next up is seeing if I can figure out the skeleton and how the materials are properly tied to the polygons (right now I have a very hacky way of doing it but it seems to work for now...).

[Carpaccio]

Excelllent! Glad there's progress for it!

[Virt]

WHOOOAAAAA!!! AWESOME!!
DUDE I can't believe this is finally happening!
I'm so excited you don't even know

[Ploaj]

So my program exports a SMD file because it's it's easy to write, so I tried using Noesis to convert it to a DAE.

In Noesis, the model looks fine

But when I tried to open it in 3ds studio, I got this.


Am I doing something wrong here?

Edit:
Here is the file

[Struggleton!]

So my program exports a SMD file because it's it's easy to write, so I tried using Noesis to convert it to a DAE.

In Noesis, the model looks fine

But when I tried to open it in 3ds studio, I got this.


Am I doing something wrong here?

Edit:
Here is the file

In Noesis, you forgot to flip UV's on export. It's very important, because Noesis automatically flips them regardless if they are correct. After flipping UVs, the UVs and Materials are still there, and with a bit of work you can reassign the bitmap to the material:


Great job, Ploaj

[Ploaj]

Flipping the UVs helped with the texturing issue, but I still don't understand what's going on with the... shading...?
It even does it in Blender.


May I ask what version of Noesis you used?
If I export as an OBJ and convert to DAE with Noesis everything is fine, but OBJ files can't hold the skeletal information.
I also noticed trying to do OBJ to SMD with Noesis gives a very buggy result. Maybe my Noesis is glitched or something.

[Struggleton!]

Flipping the UVs helped with the texturing issue, but I still don't understand what's going on with the... shading...?
It even does it in Blender.


May I ask what version of Noesis you used?
If I export as an OBJ and convert to DAE with Noesis everything is fine, but OBJ files can't hold the skeletal information.
I also noticed trying to do OBJ to SMD with Noesis gives a very buggy result. Maybe my Noesis is glitched or something.

I'm using the newest version of Noesis, 4.177. I can upload it, if needed That shading issue is odd; It only happens in the DAE file you uploaded, but not in the OBJ file. EDIT: I think it's because the DAE has it's Specular Level/Power and Glossiness up way high, causing that weird shading error.

Without Spec/Gloss


With Spec/Gloss

[Turbo Thunderbolt]

flipped faces all over the place from what I can tell

[Ploaj]

The problem seemed to come from the collada importer for 3ds max; there wasn't a problem per-say with the file itself.

I ended up writing my own DAE exporter... which ended up being a bit beneficial because I could control every aspect of the format down to the object and bone names...! It's basically a straight conversion from XC to DAE...!

The model submissions I did were auto-accepted, but I would appreciate it if someone would check them to make sure they work correctly!
3ds Max opens them fine for me, with textures no less, but I want to be sure they are okay.
http://www.models-resource.com/3ds/profe...eattorney/

With how much trouble I have had with models, this was a big accomplishment to me