Out of curiosity, what's the planned method of release? I'm guessing some kind of patch program, or pre-patched .iso?
The .unt files are pretty simply once you pop them open, there's not a lot there:
Code:
000000000: 4e55544c 48000000 24000000 00000000 NUTLH...$.......
000000010: 00000000 8c4c0000 04000100 00000000 .....L..........
000000020: 00000000 01000000 704c0000 736e5f6c ........pL..sn_l
000000030: 78785f70 7370325f 64706974 656d3031 xx_psp2_dpitem01
000000040: 2e757672 00000000 00000000 00000000 .uvr............
For some reason I can't find any .uvr files from using essen's nbl tool. Could they be in the TMLL files? Can you include xx_psp2_dpitem01.uvr as an attachment?
View attachment 179738
And thanks for the textures! Was able to get a drop working with vertex colors and texture. Need to add another material for the ring effect on the outside.
Edit:
I went through and made a scan to look for Magic numbers in the file.fpb file to make sure I was grabbing everything, and these were the Magic number candidates and their frequencies.
Code:
{ NMLL: 11413,
STD: 50404,
NUIF: 104567,
UNT: 33982,
TMLL: 2066,
NNVR: 37225,
GBIX: 37290,
PVRT: 18422,
UNR: 6592,
RIFF: 9266,
REL: 24496,
UNJ: 23307,
UNA: 9649,
UNM: 6233,
UNV: 15743,
UVRT: 18911,
NXIF: 31951,
XNR: 16644 }
I'm generally really surprised, I thought it was just archives stacked inside the file. Really surprised to see GBIX, PVRT, REL, and UNV occurring in the file.fpb file directly. I think their psudo-blowfish function in the archives mixes the bytes up enough that these don't seem to be false positives. I might also check to see where these occur. If there is a length declared after each number then it would be easier to look for specific file types, grab the length, and only pull out the files that I need as opposed to trying to pull out everything end-to-end the way the tool currently works.