This may be a 'needle in a haystack' request, but how feasible would it be to edit the original bms script by aluigi to locate a specific file type that currently isn't being extracted by the current script?
I'm after the .rpe audio bank files. By default, the game folders have the 'Hub' Scene rpe files uncompressed, however all of the main level rpe files are in the pak files.
I attach a copy of one of the uncompressed rpe files from Buffy, and also from the Indiana Jones game (PC version)
The rpe files themselves can easily be opened in GameAudioPlayer or similar....if only I could get my hands on them
PRP1986 wrote:This may be a 'needle in a haystack' request, but how feasible would it be to edit the original bms script by aluigi to locate a specific file type that currently isn't being extracted by the current script?
I'm after the .rpe audio bank files. By default, the game folders have the 'Hub' Scene rpe files uncompressed, however all of the main level rpe files are in the pak files.
I attach a copy of one of the uncompressed rpe files from Buffy, and also from the Indiana Jones game (PC version)
The rpe files themselves can easily be opened in GameAudioPlayer or similar....if only I could get my hands on them
they are, though, in xbox adpcm, and GAP will play them with the xbox adpcm codec installed , for pc use they need to be converted to regular wav though
***EDIT** had the wrong rpe's up, had to edit to replace the DL with the correct buffy ones)
in fact here is the rest of the files created by cxbx that would be stored on the xbox HD once the game is run once dialog goes right along beside the rest of these folders just like it does in indy https://www.dropbox.com/s/9b44boxw3s9i8 ... 4.rar?dl=0
Acewell wrote:here is Noesis python script to open the mtx texture samples tex_BuffytheVampireSlayer_Xbox_mtx.zip supports dxt1, dxt5 and rgba8888, top level mip only
try as I may, I can not get python to install noesis...needs javabridge and it will not compile/install... have 32bit java 8 (And sdk) and 64bit java 10 (And sdk, no 32bit variant to install) and no matter what I get "Failed to find the Java Development Kit. Please download and install the Oracle JDK 1.6 or later") RuntimeError: Failed to find the Java Development Kit. Please download and install the Oracle JDK 1.6 or later
***EDIT*** Ok fixed that error, had to edit the script, it was looking in the wrong registry key \\software\JavaSoft\Java Development Kit should have been looking in \\software\JavaSoft\JDK
Last edited by medievil on Sat Apr 28, 2018 5:19 am, edited 1 time in total.
nothing to install/compile, download Noesis and extract it some place then move the python script to Noesis\plugins\python folder, then you can launch Noesis and open mtx files.
Acewell wrote:nothing to install/compile, download Noesis and extract it some place then move the python script to Noesis\plugins\python folder, then you can launch Noesis and open mtx files.
this has nothing to do with Java by the way, i've no idea what you got going on there.
alrighty... not the same noesis for python I found through google...lol at least I got it all lined out....python installed, and working..lol was wondering what data mining had to do with it..lol http://noesis-python.readthedocs.io/en/latest/
Is it possible with noesis to batch convert a whole folder (since they are individual textures) and save them as dds files?if so then the whole texture thing will be solved cause I can then groupe them into their proper multitex files with the mtx packer bms script.. that will be a step closer to getting this to run on the INDY slayer PC engine (next step, extract the audio from the RPE files, convert them to standard wave and repack them into RPE)
***EDIT*** Nevermind, just found the batch option, works fast!!
found a bug, either in the buffy pak bms or the main quickbms if you skip or overwrite same files(cause the pak files contain duplicates, apparently on xbox they just load the pak file in memory for the level) you will eventually error out: - error in src\extra\xalloc.c line 618: xdbg_malloc()
ummm..so all you have to do is change one byte in the xbox adpcm wav.. byte 14 from 69 to 11 and they play in vlc but MPC, WMP and groove all do not (with WMP Maleware byte even sees it as an exploit..lol)
ok huuuuge issue... @ Acewell The script to reimport into mtx was one you posted (Worked on) over on the xantex forum (I posted it as code earlier in the thread) does seem to work...it will not create an mtx file from the dds files (Stripped header/extension or not)
here is a pc indy mtx, for a refresher...I know it has been a while since you worked on that script...looks like sections with pointers to where the texture is located, probably a counter up top showing how many are in the file...but not being familiar with all of this I get lost easily...any breakdown of how to create the file is most appreciated!
medievil wrote:The script to reimport into mtx was one you posted over on the Xentax forum does seem to work...
yep, it "does seem to work" based on the direct feedback from the guys posting there. the purpose of reimport is not to create, but to find and replace data with modified data. are you following the reimport instructions correctly? i test all scripts before posting them.
medievil wrote:The script to reimport into mtx was one you posted over on the Xentax forum does seem to work...
yep, it "does seem to work" based on the direct feedback from the guys posting there. the purpose of reimport is not to create, but to find and replace data with modified data. are you following the reimport instructions correctly? i test all scripts before posting them.
yea figured that out, problem is the xbox buffy doesn't have the textures all in one mtx, so no base multitex to reimport them to ... I just assumed when I ran across the script that it created not substituted...already been looking into it, that multitex header is ummm confusing... current plan is to sub in some random stuff so I can find the pointer to the file start in each sub header...
this is actually the second sub header...the first two in the last group of 4 (88 25 Standard hex byte swap location is 2588) are the start pointer for the actual file, I can move there in the hex editor and see the next file start The first two (64 15) are the header start is is consistent across all the subheaders except the very first one, BUT on other mtx files, the numbers are different
second subheader from a different mtx.... location pointer is consistant.....and seems that the first subheader is consistant across mtx files, it starts with 80
here at location 7d and 81 are the width/height if you look at the other headers posted it is consistent they all have size numbers there.. in this case it is 512x256(00 02 00 00 is 512) (00 01 00 00 = 256) in the previous post at 7c and 80 you have (40 00 00 00 = 64)
ok, also not evident from whats posted, but the 4th byte in each sub header is the file number if you look at each one I posted that is the second subheader, you see 01(first subheader is 00) not found dxt format though no 41 or 45 (2 most common)
hmmm all the actual texture files also begin with 4 bytes not sure what they mean, they are consistent for that mtx... for instance one i just checked is 01 00 00 00 another 00 01 00 00 and still another 00 02 00 00
