Chrono Compendium
Kajar Laboratories - Fan Works and Submissions => Chrono Cross Modification => Topic started by: FaustWolf on October 13, 2007, 02:00:15 am
-
BREAKING NEWS EDIT: As of Sunday, October 14, 2007, we and the folks at the xentax forums have come to agreement that the .BIK archives are in fact "false positives." In other words, we have not yet discovered the archives Square stuffed its Chrono Cross files into. But the search continues in earnest for ways of getting at the insides of the Chrono Cross .iso.
SUPER EDIT - 10/13/07: Okay, I'm re-invading my first post to provide myself with some organization and to "cut to the chase" for Compendiumites who are wading through this the first time. My sincerest hope is that this whole .BIK occurrence is going to result in the "mapping" of the Chrono Cross CD file structure. For this lofty goal to occur, we'll need three things as I see it currently:
1. *COMPLETE* A very good hex editor. Hex Workshop is the utility of choice here, but a great alternative is Cygnus. Its search feature takes some getting used to, but is actually quite a bit more powerful than Hex Workshop's IMHO. If anyone else would like to download Cygnus and help out here, find the Hexit hex calculator as well, since Cygnus doesn't have as good a "goto" feature as Hex Workshop does. *Thanx Luminaire85!
2. *COMPLETE* A .TIM viewer that can rip .TIM files and output them in their native format. PSicture rips .TIMs and outputs them as .BMP files. Clean .TIM rips will provide us with hexadecimal references for exploring the Chrono Cross iso and the .BIKs -- provided the image data isn't compressed in its pre-ripped state. *Thanx Zeality!*
3. A PSX emulator with debugging capabilities a la SNES9x Debugger. Does such an emulator even exist? If it's out there somewhere, it might tell us which disc sectors the emulator is reading at any given point in time. Therefore, perhaps we'd be able to tell where on the disc/iso pre-rendered backgrounds and character models are being pulled from.
All of the above is intended to give the Chrono Community access to Chrono Cross' resources. We've got textures already, though Zeality and I strongly suspect we're missing some of those. Let's push onward and get model data!
This idea has only just developed over the past few days, and in many ways I still don't know what the heck I'm doing. I've dabbled in each of the necessary aspects of this blossoming project, but I've been here long enough to know that more gifted Compendiumites have far greater expertise in these sorts of things than I. If you're reading this and can provide advice or have ideas on how to proceed, by all means join in! Hopefully this will be a long-term Compendium project that will occupy Chrono fans for a while, and if we're successful in ripping models and the like, it will usher in a new age of Cross hacking or Cross-based fangames.
Original post and edits appear below:
Hello again everybody! I swear I'm not trying to up my post count by spamming this section of the forums! :lol:
I recently decided to throw my Chrono Cross iso at a program called "Filestripper" produced by the xentax community. I *don't* recommend anyone else do this because it takes a God-awful amount of time and will tie your PC up for hours. Anyway, I can see that I've gotten about 420MB worth of ".bik" files so far. I assume it's an archive of some sort, and I've got three such archives (still running Filestripper, but I don't expect to get any more such archives based on the way things are progressing).
Has anyone here encountered / produced .bik files before? Any ideas as to what they might contain? My cursory Google search suggests that .bik files usually contain music and/or video, but I sincerely doubt that Chrono Cross' soundtrack and its videos combined would take up 2/3 of an entire disc -- unless Filestripper has somehow bloated the file sizes by converting Chrono Cross' music and FMVs into some weird lossless compression format. I dunno.
I'll upload them via Rapidshare for interested parties once Filestripper finally finishes. It's taking its good ol' time, so that won't be until tomorrow morning in my neck of the woods.
Here's to hoping it's something interesting and that we can find out how to open them! I don't want to give anyone false optimism though. It could be 420MB of unusable junk data too.
EDIT: Just thought of something. Filestripper's been giving me a gazillion "Invalid MPEG frame found @ such-and-such an address" messages. Maybe the program's stuffing its error logs into .bik files and that's the source. I hope not; that would be Teh SuX0Rz. I guess we'll see.
EDIT: Finally had to abort the program before it finished due to the need to get some work done today. I've uploaded the .BIK files in separate rars, which people may grab at Rapidshare below:
http://rapidshare.com/files/62248175/Track_010AAF37EB.rar.html
http://rapidshare.com/files/62269095/Track_0113E96864.rar.html
http://rapidshare.com/files/62272980/Track_0100563BCC.rar.html
Filstripper's readme says it can rip .BINK files, which come from Rad Game Tools (http://www.radgametools.com/default.htm). Though quite frankly I'm unsure as to whether or not this has anything to do with .BIK files (I most definitely captured .BIKs and not .BINKs, at least that's what my computer says).
RadGameTools says they've contracted with 1100 customers since 1999, and Squaresoft isn't on their partial list of companies. Knowledgeable Compendiumites, when was Chrono Cross released in Japan, and how long was its development cycle? Is it possible Square would have contracted with a company like this to provide...whatever it is this company provides? Rad Game Tools isn't in Chrono Cross' list of credits in the instruction booklet. Probably a false lead.
EDIT: Aha! BIK, BINK, what's the difference? Rad Game Tools is power and I've got the -- oh, nevermind. Anyway, Rad Game Tools does use BIK files. Could Square have contracted with this company? I've got a BIK compressor from Rad Game Tools, but I don't know if I'll be able to find a decompressor. Stay tuned.
-
I'm not sure which is worse - posting the longest thread ever, or double-posting. I'll just double-post.
Still haven't found a decompressor folks, but guess what's in these .BIK archives? TEXTURES!! Whoo-hoo! Apparently PSicture can get at them. It's most likely that they are exactly the same ones Zeality and I have already, but I'm going to upload all the textures I can find in these .BIKs.
-
Triple post! Sorry, but I'm excited over all this. 8)
Here's the textures PSicture ripped from all three .BIK archives I was blessed with last night. This is Disc 1 stuff only, btw.
http://rapidshare.com/files/62259892/BIKTextures.rar.html
I'd like to point out that the .BMPs ripped by PSicture only account for 130MB out of the 420 MB contained in the .BIKs, and the presence of the textures within the .BIKs strongly suggests that what's inside is useful.
What else could be in these .BIK archives? Music? Movies? Character model pieces? I'll throw a .BIK at the xentax forums and see if anybody there bites, but that will take months, years, or forever since those intrepid programmers do their decompressing magic for free and are inundated with hundreds of requests like this.
Hopefully we can find a way to decompress these .BIKs ourselves before the End of Time. I'll finish uploading all three .BIK archives for intrepid explorers to examine. FORWARD, KNIGHTS OF THE SQUARE TABLE! RIP THESE BIKS TO SHREDS!! :twisted:
-
I'm gonna go ahead and say that this is probably the most active compendium has been in months. This is just amazing stuff to work with.
-
All three .BIKs
(I'll refer to them as my "blessings" from now on, I think) are available on Rapidshare for interested parties now. See first post.
Yeah justin, 'tis certainly an era of exploration for both Trigger and Cross. What's so monumental about the .BIKs is that I believe these may be the archives Square compressed Chrono Cross' data into, sort of like the .lzs archives you see when you pop Final Fantasy VII into your CD tray. *IF* Filestripper ripped these archives straight from the iso without any junk getting into it, *THEN* we should be able to get an idea of the Chrono Cross CD's file structure. For example, we might be able to toss one of the .BIKs in a hex editor and see what its file header looks like, then locate that file header in the iso. Then we can say, ".BIK 1 starts at address xxxx and ends at address zzzz, .BIK 2 starts at address yyyy, and all the data in between zzzz and yyyy belong to files or archives we aren't familiar with yet."
From what I've seen you've way more experience than I have in the hexadecimal scene, justin. Can you tell me if I'm thinking straight? :lol:
EDIT: I'm linking to the pertinent thread on the xentax boards. Hopefully any feedback I get there will lead to synergy and fellow Compendiumites will have ideas for how to proceed. Posting on the boards requires a forum membership (which might not be free - I don't rightly remember). If anyone can't see the xentax posts from the link below, tell me and I'll recap everything that occurs there.
http://forum.xentax.com/viewtopic.php?t=2815
If you have any comments or ideas regarding what's posted over at xentax, please discuss in this thread, I guess.
-
Craaap, I won't be free to look at this until late tonight. Anyway, yes...this is revolutionary!
-
Let's hope this all leads somewhere. I've just discovered that the .BMPs we've been capturing as textures over the past few days are, in fact, .TIM files thanks to Darkfox at the xentax forums. Bah, I should have known that already -- PSicture just outputs to .BMPs after capturing .TIMs and other files.
So I guess my question now is, does anybody know of a .TIM ripper that will actually output .TIM images? It would prove useful because that way we could examine the hex signature of the .TIM files and compare it to the hex code inside the .BIK containers as well as inside the iso itself.
-
TIMViewer does it. I had to use TIMViewer instead of PSicture on the demo because PSicture, for some reason, refused to read any of the textures past a certain point. Thankfully, TIMViewer didn't have such a problem, but it meant I had to rip to .TIM first and then batch convert to BMP.
-
*Woot-woot!* Now I've got 9619 TIM files from Disc 1 - that's 9619 giant hex strings that can be used as "mile-markers" for the Chrono Cross internal file structure map if my theories are correct. Since you've ended up with only 5000 or so unique images I think, it will be interesting to see the dynamics involved with duplicates (if we could get our hands on a PSX emulator that provides live feedback regarding its various internal processes). I mean, if Serge's texture is duplicated in sector range "x" and sector range "y", does the game instruct the emulator to look only at "x" at all times? But that's just a passing interest, I guess, compared to finding out where the model and pre-rendered background data is.
I figure we can map Disc 1 first, work out all the kinks in that process, then apply it to Disc 2.
-
BREAKING NEWS
Booya~! I can now say with approximately 90% certainty that Serge's (first) TIM texture is located between offsets
002B6340 ~ 002BA55F (these values were acquired before I enforced "quality control" on myself and are no longer valid).
in the first .BIK posted at Rapidshare above.
I have proof! Pics are big for clarity, so click on the image URLs to see what I'm seeing.
http://img205.imageshack.us/img205/8215/hexproof288beginningvc4.jpg
The lower panel shows the beginning few bytes in TIM 288, Serge's first .TIM texture if you rip from the full iso with TIMviewer as per Zeality's suggestion. The upper panel shows the same bytes starting at address 002B6340 in the first .BIK file. They're identical, though the .TIM hexadecimal seems to be shifted by one, er, "slot" (2 characters). It's easier to see the identical-ness if you refer to the ascii strings to the right of the indicated hexadecimal blocks.
http://img140.imageshack.us/img140/8028/hexproof288enduw8.jpg
Jumping ahead by 00004210 hexadecimal bytes(?) - the file size of .TIM 288 expressed in hex - we apparently come to the end of the file in the .BIK.
The reason I say I'm only 90% certain that this is Serge's texture .TIM 288 is that it may be the case that I'm only showing the generic header and footer of a .TIM file - maybe any character's texture would have the same beginning and ending, and I could be looking at Norris in hex for all I know. But still, this is turning out to be *very* promising.
It also has an important ramification - the .TIM data in the .BIKs are not compressed; if the .TIMs were compressed, their data would be scrambled by the compression routine I imagine. I hesitate to say that none of the data in the .BIKs are compressed; is it possible that some data in a container or archive could be compressed while other data is not?
I don't know where Serge's .TIM 288 is in the full iso yet, because my puny early version of Hex Workshop is neither man nor woman enough to handle a 600MB file with a Microsoft Word document open at the same time. I'll try again once I'm done studying for the evening and report.
Now a whole other can of worms is opening up. How should we keep track of where the various .TIM files are in the iso and .BIKs? My proposition is to create a separate notepad / text file for each separate .TIM with the appropriate info, as per the following: http://img187.imageshack.us/img187/7131/tim288recordoo5.jpg
But I'm not sure how feasible that would be. I mean, that's potentially 9000 text files being hosted at the Compendium if we track all TIM file locations including duplicates, and the one for .TIM 288 is 70KB as it is.
What do we think, crew?
Oh, and I'm not all that well-schooled in ROMspeak, so I'm probably off in my terminology above. Feel free to correct me anybody, if the need arises during this discourse.
-
A few thoughts, in bulleted form:
- Looking at the *.tim file format (http://wiki.qhimm.com/PSX/TIM_file), it looks like only the first few bytes are identical between files, if that. Ironically it might be easier to identify TIM structures if more bytes were the same at the start because we could code an algorithm to do it. But I agree that it looks like you have found the right offsets for those files.
- You might try the free version of Cygnus Hex Editor (http://www.softcircuits.com/cygnus/fe/). I doubt it's the best one out there, but Wikipedia says it can handle up to 2 GB files. I had no problem opening the *.bik files in it.
- I also tried two of the *.bik files in RAD Video Tools 1.8w, and both failed with an error message saying "This Bink file uses an old or beta Bink file format - please re-Bink the original video file." This page (http://www.radgametools.com/bnkhist.htm) states that way back before March 1999 they changed the Bink format. Seeing as how Chrono Cross came out in 1999, it wouldn't be surprising if they were still using the old Bink beta format. :(
- This page (http://www.radgametools.com/binkgames.htm) states that Square Enix licensed Bink for, among other things, FF8 for Windows. Interesting.
- It's strange that they appear to have obfuscated their game data using an audio/video codec.
-
Holy cow, Luminaire, thanks for confirming the BINK/Square relationship!! Amazing info, and Zeality can even include that little tidbit in the site's next update perhaps. Fascinating industry detail there. I wonder if there's an old BINK decompressor floating on the Net somewhere...
If we could track down Rad Video Tools ver. 0.8, we could just be done with this whole matter in a matter of days, probably. Tantalizing!
I'll check out Cygnus too. My old copy of Hex Workshop lasts 89 more days, so I'll be good through Christmas break and a little beyond.
And -- writing algorithms!?
Spekkio: We've hauled in a Marlin here, kids! /Spekkio
Thanks for walking in here, Luminaire!
-
425.893.4300
We can just call them, tell them we have a really, really old BIK file we dug up on an old hard drive and no longer have the source video, and are wondering whether 0.8 is available anywhere so we can decompress it.
-
I found 0.8i and tried playing; got the same message.
But I can't find a general decompression option, just something to convert to AVI or other media formats. So of course it won't play a file archive.
Is there something outside "video tools" using BIK?
Googling brings up no downloads earlier than 0.8i.
-
I suppose it wouldn't hurt to send the Rad folks an email asking them if they can lend out a beta version of whatever program could decompress .BIK files. Though I'm a little worried about them asking exactly what it is we're up to. C&D time! Though Square must know what the Qhimm community's been up to with Final Fantasy VII all this time, and they haven't had any problems yet.
@Luminaire: CYGNUS ROCKS! Once I figured out how to effectively use its search feature, hexadecimal exploration has been going much more smoothly than it had with Hex Workshop. Thanx for the suggestion!
-
Okay, time for an update brought to you by Cygnus.
First, .TIM 288 and its clones appear in .BIK #1 within the following offsets:
0002F0FD ~ 0003331D
000C6D69 ~ 000CAF80
002B6341 ~ 002BA560
.TIM 288 appears in the .iso a total of 9 times:
00B57000 ~ 00B5B220
013488E8 ~ 0134CB08
026B728C ~ 026BB4AC
027260EC ~ 0272A30C
0902C2DC ~ 090304FC
0A374CE0 ~ 0A378F00
0AB228E8 ~ 0AB26B08
0ABBA554 ~ 0ABBE774
0ADA9B2C ~ 0ADADD4C
Whew! And that's just one of these suckers. Typing all that out brings back fond memories of the Ultimate Sephiroth Code. We're getting Sephiroth into Chrono Cross once we've figured out how to rip everything a few years from now...right? :shock:
Oh, but I haven't uploaded all the .TIMs yet, have I? That will happen tomorrow morning.
I'm starting to think the best way of mapping out .TIM file locations in the iso will be to get a huge Microsoft Excel file going. People can post their findings on the board and I can add them to an Excel file that could be hosted at the Compendium, perhaps?
Next up, besides the .TIMs upload, is to find out where the .BIK files are in the .iso. I can already see a .BIK header -- it literally reads "BIK" toward the beginning.
-
Hey, Zeality - I've been reading a bit of Terminus Traduction's User Guide, and Nemesis writes on page 2 (pg 4 on the Adobe navigator) that their team has been able to access a table contents at the beginning of the Chrono Cross ISO!? Am I reading that right? If so, that might cut out a lot of the "mapping" work and we could get right down to extracting archives and finding ways to open them. But my understanding of the user guide is hazy at the moment; I will take a more thorough look at it later.
Do you know if Terminus Traductions was working with the American version? I wonder if the files on other versions of Chrono Cross could be easier to get at...
EDIT: Daaamn, I've come up with an even easier way to map where the .TIM files are! Involves only the .iso, TIMViewer, and the Hexit Hex calculator! It'll be much more accurate than what I've been doing, too; less worries about quality control now. Should have thought of it earlier, silly me. Mwahahahaha~! I'm going to do an illustrated tutorial and post it in a separate Kajar Labs thread (if that's okay), and hopefully we can get it stickied. I'll get that done during the week, and then we can start on some serious datamapping.
I promised I'd upload the .TIMs, and here they are without duplicates if anyone wants to experiment with anything involving .TIMs and doesn't have them already:
http://rapidshare.com/files/62505597/TIMs.rar.html
Also, I haven't gotten around to determining exactly where the .BIKs are in the .iso. I'll get that done later today.
We are on the verge of some serious stuff here, folks! I haven't felt this alive in ages!
-
Ooh, just thought of something ominous. How different are the Chrono Cross versions released in various countries? Like, will the European versions have data stored at totally different locations than the American version? I own the Ohio version of Cross myself. 8)
-
There is no European version of Chrono Cross that was ever released. And there'll probably never be.
Same thing with Chrono Trigger. There are only the NA versions and the Japanese ones.
And probably our versions are a little better than theirs, if I may say so.
-
Thanks for that, dankun. That gives me some ease of mind, actually -- though I'm sorry for European folks. Silly Squeenix.
-
I'll see if Chrono'99 can e-mail Terminus again. They gave us the ripped script in '05 before releasing their tools.
-
CONFIRMED -- the .BIKs are indeed in the ROM in their extracted state, i.e., FileStripper didn't take data and slap a .BIK format onto it all by itself, which was a concern I had. EDIT: Or at least FileStripper didn't change the data any. Is it possible FileStripper could have added a .BIK attachment to ripped data without changing its hex code in any way?
The first .BIK appears between offsets 0AAF37EB ~ 12B040B5 in the iso file. I'll get a "macro" map of the iso going as well as the other .BIK offsets when I get a chance.
Also, the .BIK topic has attracted some interest at the xentax forums. I'm currently having a forum discourse with Savage, who I believe speaks English as a second language. He might be telling me that the files aren't .BIKs after all, but in light of the above I tend to disagree. Though I'm not exactly sure what he's saying -- could someone take a peek at the topic and give me a second opinion on what he's saying? IF you can view the forums without having to register, of course. If not, I'll do the best I can over there.
http://forum.xentax.com/viewtopic.php?t=2815
-
Your reply should suffice to get clarification. Chrono'99 will e-mail Terminus tomorrow once I fix something in the Cross script.
-
I'd say Savage is arguing that since he can compress the file you sent him considerably using RAR, he suspects the original data is not compressed and therefore not of the Bink format. He thinks that Filestripper saw the hex code for BIK and, since Filestripper is looking for Bink data, immediately started writing a .bik file. Since (1) it makes no sense for images/model data/etc. to be compressed with Bink and (2) you've found uncompressed TIM data and possibly uncompressed TMD data within the iso/bik, I'd say he brings up a good point.
Bah, now that I think about it it makes perfect sense.
-
That clarifies things for me, Luminaire. If FileStripper did chuck data into .BIK files by itself, it didn't change any of the file data as far as I can see in hexadecimal though. To complicate things further, I'd like to point out as well that the .BIK files don't seem to have a consistent header -- the ascii reads "BIK" and a bunch of random gobbledegook after that.
Though on the other hand, there's lots of instances where "BIK" appeared in ascii within the iso and FileStripper *didn't* shove the following data into a fresh .BIK file.
I'll show you the beginning and ending .BIK hex and see if you can make more sense of what's going on than I can:
BIK1: (http://img141.imageshack.us/img141/7899/bik1el9.gif) (http://imageshack.us)
BIK2: (http://img361.imageshack.us/img361/2117/bik2sj1.gif) (http://imageshack.us)
BIK3: (http://img361.imageshack.us/img361/1387/bik3cn3.gif) (http://imageshack.us)
FileStripper is literally trained to look out for .BIK files -- well, Bink & Smacker files, which I assume includes .BIKs. This could be taken either way though -- maybe FileStripper got "trigger happy" when it saw the BIK code and said, "This is a .BIK!" when it really wasn't. Ah, so confusing.
Should we forget about the whole .BIK matter and focus on finding .TMD and other data such as pre-rendered backgrounds in the iso, you think? It seems things are progressing beyond the need to open the .BIKs anyway since the data inside is uncompressed -- I hope it stays that way. We'll have to see if we can extract another .TMD or other file from somewhere else in the iso to be sure.
Thanks again for your wisdom, Luminaire!
Oh, Zeality -- maybe Terminus Traductions would know the offsets of the game text and "accent" files within the iso? If so, that would be good to know in case the Compendium really needs to map out the entire iso to get at the good stuff. Did they use the American version of Chrono Cross for translation? Not that I know whether it would make any difference.
And another question for ya -- I imagine people have done PSF and perhaps FMV movie rips (.str?) from Chrono Cross. Do you know of anyone who might know where that data is located? If not, I'd still be ecstatic if I could get my hands on whatever file type the movies and music were in their "natural" state, since we could easily locate its position from the hex signature. If we could get that stuff mapped out it would eliminate quite a large chunk of the iso that we have to worry about.
-
Yeah it's a bit unfortunate but it's more likley that they are a wrapper file that square has used to bunch files (doesn't look compressed though) together. A 'false positive' is the probable explanantion, just bad luck the the data in the headers looks the same.
EDIT: Still it's likely that these files contain most of the resources we are looking for. I'll start having a look at them at some stage this week (about to start exams at uni), it's tough when we don't know what were looking for but I guess it'll have to start some where. Who knows, we could get exceptionally lucky and stumble across other file types.
-
Once I get the positions of the "supposed" .BIKs (no longer my "blessings," grr!) maybe we'll be able to sit down, take a general look at the iso structure with the .BIKs in place, and see whether or not it makes any sense for Square to have used .BIKs as a noncompressing container for the files.
-
PSMplay and PSXMC both can scan a raw ISO for streaming sounds and movies. PSMPlay is better (because PSXMC doesn't let you rip the native files; it only lets you convert), but PSMPlay is also hard as hell to figure out (at least it has been for me).
IIRC PSMPlay can scan for textures, too, but I'm pretty sure it's just TIMs all over again.
-
Thanks for the tip, Zeality. I'll look up PSMPlay then and see how I fare.
The xentax folks have pretty much confirmed what Luminaire has suggested, i.e., that the .BIK files are false positives. Drat! The .BIKs I have are now only useful as snippets of the iso that can be more easily handled by hex editors than the entire iso. Useful in and of itself, but not the breakthrough I was hoping for after all.
Be sure to have Chrono '99 ask about the Table Of Contents Terminus Traductions found/created. That will be a major help, methinks.
[attachment deleted by admin]
-
Hello. Do you still have old version of RAD video tools? Could you upload it for me? I'm searching the web for old versions and I can't find anything. :)
-
VLC 2.2 introduces support for BIK files