all your samples share the same basic format. the '.jpg' extension has some run-length encoding and the '.tga' files look like jpg masks uint flags char sig[3] // FBN when flags == CDAh // raw image data uint enoffset byte jpgMask[ enoffset - FTell() ]; when flags == AEEh // rle byte flags[3] // ??,...

if you haven't wrote anything for blender, that's the next step - i guess it just uses different api calls to do the same thing (only now you can edit stuff too..) 2d formats - textures? i couldn't get into that either - most graphic formats are dds variants and just headless or using some preset op...

around this topic, his blog is really interesting - http://theinstructionlimit.com/category/fez
he was also interviewed for the indie game movie if you want to know more

there are many examples in the manual on for loops. to create the mesh object from the vertex/face index data you need to store them in an array before you can create the 3ds 'mesh' object: arr_v = #() for i = 1 to vertCount do ( append arr_v [readfloat(f), readfloat(f), readfloat(f)] ) note face i...

there are many examples in the manual on for loops. to create the mesh object from the vertex/face index data you need to store them in an array before you can create the 3ds 'mesh' object: arr_v = #() for i = 1 to vertCount do ( append arr_v [readfloat(f), readfloat(f), readfloat(f)] ) note face in...

you can challenge yourself by working in different software environments. you've done some python stuff with neosis? try using 3ds maxscript, or maya (melscript?). being by porting stuff.

you say branch out, is there anything specific you have in mind?

there is a null-terminated string table with some japanese encoded texts

i meant my script was useless. you can easily parse the filenames, but the data is encrypted or compressed somehow. there is all kinds of data in the last archive posted: sfx/wate_001.mp3 2541926 71117 64331 sfx/you_win.mp3 9688207 42904 38249 sfx/zangief_bonus.mp3 9726456 146349 139543 table_logo_v...

filename parser. useless. idstring "PX" goto 12 get files long get filetablesize long goto 40 for f = 0 < files get a long get b long get fancy_f char get strlen char getdstring str strlen padding 4 get unknown long print "%str%" next f print "Finished"

there was a filename table NOTE the output xml needs reconstructing. i have no incentive to write a reconstructor for that # QuickBMS script for Dark Strokes # Written by WRS idstring "RDFZ" get strlen long getdstring str strlen if str != "Zlib" print "Unsure how to handle f...

try [url=http://aluigi.altervista.org/mytoolz/offzip.zip]offzip]

attached is a c routine for decrypting the tables but that doesn't really help with the filetable

for debugging, the first expected spk file is 'paks\update0.spk'

Z3Ex.dll is now obsolete.

I am now finishing up a standalone new tool which can extract all existing client releases:

https://github.com/x1nixmzeng/z3ResEx#download

if you were curious about the decompression (skyflox outlines it here) - then I have completely reversed the RLE z3Rle.h

how on earth did you figure out that hashing stuff? is it a known algorithm?

you also have an interesting coding style

ot: while the gameplay looks terribly repetitive, that sloth boss was hilarious!