good things. is this based on the ps2 or cube data tho? if you do the cube, i'm out. fine to me. have at it.

There's also a GameCube version, here's some samples from that: https://puu.sh/BOcUM.zip I would say 'forget about that'. :lol: that seems packed. i don't like figuring out compression algos. and... isn't the cube a big endi? i'm not gonna try to read that. float maybe but pointers get confusing. u...

well. to bad it's ps2 only. for pc you'd have a load of choices which mdl/mdg extractor you wanna use.

it sure looks like sotc format. 0x[count] and 0x6[datatype] flags on the strip data.

well. the dat container is really bad to read. no clear start and clear pointers. you see the script uses binary search to find the signatures. it does not hit the NMO directly either. neither extracts files as files. means i had to wing that file with the bird_A string and using some signature to s...

(if i'd ever meet this guy i gotta punch him for writing this cryptic and incomplete mess that nobody could finish easy. what a stupid dummy). yeah, punching people who create solutions. If that isn't stupid? :roll: The author is a funny guy, indeed, guess, a very young one, when he wrote the scrip...

well. yes, the model is already xtracted. nothing we can do with that data. you coulda just uploaded the bird_A.nmo and textures. that's how it rolls here. and while i'm getting a rom and further search for file extractors (that may not exist anymore) you could check out this link and see if that so...

i'm not exactly active on this d3 thing. just picked up the file puzzle on zenhax. and well... those samples seem to be bc6h format. i've never seen the pattern really, but i can spot half float data. will update when i got it. i won't update the d3texconv tho. i can't read this decode code. seems n...

what's this love for the dead beef? upload a sample then. done that other one already. i might just add to it.

got any model (with the textures ofc) samples and the script? i could lean myself out the window and assume you may just have to scale the uvs. the ps2 models i got to extract (using those weirdo 3ds bms scripts) used integer uv coordinates not floating point. they are just damn large. i think they ...

explain? take the file then. it's still incomplete. as posted i just read all the 3x3 and transform in the matrix prelist (so it's not messing up the matrices doing the recursion at the same time). then recursed (and multiplied) all into the seperate postlist. done that trick. still got no clue howt...

well. i felt like it. easy progress. the weight data seems easy, but it sure doesn't wanna work easy. i need a break now.

9 floats for 3x3 local rotation matrix, 3 floats for local position vector. snip @episoder Your approach is correct, as you need to reach the world space coordinates, but it can be achieved more easily using matrices. Load 12 floats as NoeMat43. Multiply bone matrix with the parent in recursion unt...

yo. it was a weird thing to find anyway. i figured it out when i messed with the boneids. i added those and the head added to the neck was at the right place. but it didn't work any further. i had to try to recurse it. basicly what it does is adding what seems local bone transforms based on their pa...

i got the head looking good. and a nice start at the body. parent recursion. i guess the rotation matters.



wip script if you wanna mess with it.

yo. same here. but i got somewhere removing all the matrix stuff. using an identity mat to pin the locations. your script did output weird stuff. and i forgot to save the offset after the bone array. :mrgreen: the head bones were sorta pinned by the L and R X values. for sure the array location with...