Ibara emulated (read first!)

[clp]

as a contributor to the ketsui dump i think guru should stfu about his todo list and make it priority , the guys a nob .

[Elixir]

Why, yes. I was making the point that if you believe in sending a person a one thousand dollar arcade board for it to be dumped, only for no further progress to be made afterwards, you are beyond stupid and should leave this hobby.

So I guess the 30 or so people on the forum here who donated for the Ketsui dump should GTFO?

Most of them have probably already abandoned all hope.

[bcass]

The more informed ones haven't.

[Keade]

What are you talking about ?
According to Guru's dump list, Ketsui has already been processed (many months ago, if memory serves).
Which is why the board was bought and sent to Guru : to be dumped.

[bcass]

I'm going off the last few entries on the Ketsui dump project page:

http://community.livejournal.com/the_dump_ever/

[Keade]

It had even been pointed out on this very board months ago, I believe.

[bcass]

It seems you're right. I've just had a look on guru's dump status page, and Ketsui has indeed been dumped:

http://guru.mameworld.info/global_index.html

Strike-through games indicate that the game has been dumped.

Which, in turn, means that guru has honored his promise. One in the eye there for everyone slagging off guru then.

[Evrain]

It seems you're right. I've just had a look on guru's dump status page, and Ketsui has indeed been dumped:

http://guru.mameworld.info/global_index.html

Strike-through games indicate that the game has been dumped.

Which, in turn, means that guru has honored his promise. One in the eye there for everyone slagging off guru then.

Allow me to disagree.
As Guru points out somewhere deep within the pages of his site (which I agree, at times shows quite an egotistical and boastful attitude I find unnerving, and trust me I know what I'm talking about, I've been project manager for iDeaS as well as "knowing" several like-minded devs, Jabo from PJ64 among the "worst"), every board that is SENT to him, e.g. not purchased via the various MAME groups or by devs or himself, is dumped whenever he gets to it, and then it can be:
- kept for resale to help fund MAME
- sold to Guru for resale (or for him to play)
- returned to the owner
In any case, whoever sent it can request to RECEIVE THE ROM IMAGES, no matter the purpose: it's clearly stated that he dumps, they get the images, they can do whatever they want, including distributing them. So, whoever was the person that PHYSICALLY contacted guru and sent the Ketsui board, has every right to show him the page and request the dump. Problem is, knowing who this one is means CAVE can trace back the ROM "leak", and it's.... well, you know what happens next.
However, if one can contact him again, we can ask NICELY guru to please GTFO and send the files, since it's our right, or give back the board - there are other talented dumpers out there, some even gladly happy to help strike one against guru. But if he abides by his own principles, he will send the dump. Oh, and don't forget that the board runs a custom bios, so that's needed too.
That's step one. Who's the sender? take a step forward!

Step 2.
Haze once stated that, even with the full dump of the game, the problem lies in the ARM ASIC cpu the game uses: most newer games just employ it for copy protection, and that's it - and we know how to DEFEAT it - but CAVE is different: according to him, they effectively use the chip to execute the mathematics behind bullet movement. In short, the game would have no bullets, à la DDP2 (not that anyone cares about that piece of crap anyway) or not boot at all (much more of a chance). The chip needs decapping for proper dumping, that means $1200 and a few boards into the bin.
To be honest, after this "Ibara" exploit, I'd think that either Luca Elia had an Ibara board and did the above (minus actually pressing Guru for the ROM, of course), or it's proof that games dumped, even newer ones, ARE indeed passed among the core MAMEdevs - I once found traces of Metal Slug 4 in the neogeo driver, and it had been just released. But more than that, it would mean even the ARM bios has been dumped, and thus all we have to do is get the full package (ROMs + PGM BIOS + ARM internal code), ship it to the FBA devs and let IQ_132 do his magic. He already stated he would work on CAVE games, if he had the full dump (he only managed to get his hands on a partial Galuda dump).

Best course of action right now would be securing both the Guru dump, as is the right of every contributor, as well as the board.

P.s.: I'm all the way against piracy, mind you! Already bought MSF, Ketsui DL and hoping for Galuda 2. However, I'm afraid Ketsui is in a whole different league, both due to CAVE's apparent ineptitude at proper marketing and a sense of dread over the fast-growing childishness of quite a few devs (back in the day, it was only Reip and Razoola, now instead...)

[bcass]

I don't especially care about any ongoing slanging matches people might have with guru. The fact is, it's been dumped. I've PMd the guy who sent the PCB to guru to let him know it's finally been dumped and to request the dump. If a decap is required, then it might be worth contributing the board to that project too. I don't think any of us who paid for it would object to that.

[Evrain]

I don't especially care about any ongoing slanging matches people might have with guru. The fact is, it's been dumped. I've PMd the guy who sent the PCB to guru to let him know it's finally been dumped and to request the dump. If a decap is required, then it might be worth contributing the board to that project too. I don't think any of us who paid for it would object to that.

You don't care, unfortunately many do, and that's ruining emulation as well as most homebrewing/hacking communityes
If the guy is still active, then we might very well be able to obtain a copy of the dump: however, I think it can be legally "kept" only by him and whoever else has the board - dumb copyright laws - as well as obviously the contributors. The legal problem still stands though.

As for decapping, that again needs professional intervention, and apparently Guru is one of the few that can offer it for less than $1200 per chip. For now, better re-secure the files and, most importantly, the board!

[bcass]

To be honest, every technically-oriented community (especially where programmers are concerned) are full of people with massive egos. That's not news, it's just a fact of life. Also, guru does decapping for $200 per chip IIRC.

[Keade]

The cost depends on the type of chip, I'm not too sure. According to Mr Decapitator's blog, the price can be >200$, depending on what kind of chip it is.
Also, that's just a detail, but guru doesn't do the actual decapping, it's done by this Dr Decapitator.

[bcass]

Also, Guru doesn't do the actual decapping, it's done by this Dr Decapitator.

Yes, I'm aware he's just the middle-man.

[pixelcorps]

I will contact guru this week.

If it's dumped, what purpose does the decap serve?

[orange]

get the full package (ROMs + PGM BIOS + ARM internal code), ship it to the FBA devs and let IQ_132 do his magic.

[elvis]

If it's dumped, what purpose does the decap serve?

Decrypting certain dumps, as I understand it.

From what I've heard, a lot of the bullet pattern data on Cave PCBs is encrypted, which is kind of integral for playing the game.

To be honest, every technically-oriented community (especially where programmers are concerned) are full of people with massive egos.

Quite frankly they deserve their big egos. People like that make people like the rest of us appear like retarded chimps.

Call me a suck up, but I happily bow and grovel before their greatness.

[Evrain]

I will contact guru this week.

If it's dumped, what purpose does the decap serve?

Every PGM cart, aside for the few ones released along with or shortly after the system, also carries an ARM ASIC cpu: on games such as Knights of Valor, it contains code for decryption of the content, quite nasty because every time the returned value changes slightly and it either messes up the GFX or it crashes the game altogether. Some games had theirs dumped (KOV among others) and added to the sets, others didn't (DDP 2), some had it "simulated" but crash very, very soon into gameplay (again, DDP2 as well as other IGS games before being cracked by ElSemi).
With CAVE games, it's a whole different league: along with having a customized PGM bios (which, though, I assume they were smart enough to dump and add to the current set) the code in the CPU performs the arithmetics behind bullet translations on screen, taking some work off the main 68000 CPU and also ensuring copy protection - according to Haze, it's their "bullet processor", and the definition suits it right. He mentioned its complexity, however I trust Haze, I don't think it was just a bogus - nothing a good decap can't take care of, mind you.

That's exactly what we need, in short: the BIOS files (unless already dumped, Guru should know?) to add to the already available set; the Ketsui main (encrypted) ROMs; the ARM code for decryption. Given we have all these files, it's just a matter of offering them to the FBA team (to be honest, adding games to a driver is easy, getting them to work a whole different league!) and we should see some results: ZiNC was born because MAMEdevs were reluctant to emulate a certain system, then deemed "too new", after all.
Oh, and also that should ensure us Guru delivered a full dump and not a... "partial" one "casually" missing a program rom.

Edit for those interested: it's an ARM7, the earliest could be trojaned, the newest can NOT. However, apparently the decryption scheme (well, the beginning of it at least: PGM games make multiple checks at certain setpoints or as internal timers go off) is the same or very similar to other games, as IQ_132 was able to decrypt ESPgaluda, which apparently was compiled on the Ketsui template and for which we do NOT have the ARM7 internal rom. See here for details

EDIT: just noticed that Ketsui, along with Galuda, are just strike-through, whilst DDP2 is marked with "needs trojaning", so there is a good chance he managed to get a full dump. That would make sacrificing the board unnecessary. It's worth asking.

[rtw]

Why does the ending screen indicate that it's the location test version being played? Is the dump from a proper Ibara PCB or did Guru get his hands on something dug out of the dumpster behind Cave's offices?

To quote Luca from the mameworld forum:

Code: Select all

That's actually the only mock-up among the screen shots (as hinted by the filename and PNG meta information).

It was taken from video memory with the graphics viewer built into the driver since, given the emulation speed,
I couldn't get that far otherwise (in this lifetime).

[rtw]

@Evrain

You seem to know quite a bit about the PGM system, do you know how to create the decryption table in order to decrypt the main 68000 program code ?

[Evrain]

@Evrain

You seem to know quite a bit about the PGM system, do you know how to create the decryption table in order to decrypt the main 68000 program code ?

Creating it is not a problem, as it uses the same XORing as just about every PGM game out there. The problems are:
1. is Guru going to give out the files (it would be naive not to know we're probably going to plug these into a FBA/MAME hacked build)? Given how Crazy Otto created such a mess, I'd say he's going to
2. is Guru going to give out the ENTIRE archive?
3. has he dumped the ARM7? I truly hope so, although having no external RAM, trojaning is not an option. Since it's struck through and not marked like DDP2, there is a chance

Sorry, straying off topic.
Apart from that, no need to create the decryption code: IQ_132 already did that - EDIT: disregard, as rtw pointed out it's just for Galuda.

Code: Select all

unsigned char espgal_tab[256] = { // IGS0007
        0x49, 0x47, 0x53, 0x30, 0x30, 0x30, 0x37, 0x52, 0x44, 0x31, 0x30, 0x33, 0x30, 0x39, 0x30, 0x39,
        0xa7, 0xf1, 0x0a, 0xca, 0x69, 0xb2, 0xce, 0x86, 0xec, 0x3d, 0xa2, 0x5a, 0x03, 0xe9, 0xbf, 0xba,
        0xf7, 0xd5, 0xec, 0x68, 0x03, 0x90, 0x15, 0xcc, 0x0d, 0x08, 0x2d, 0x76, 0xa5, 0xb5, 0x41, 0xf1,
        0x43, 0x06, 0xdd, 0xcb, 0xbd, 0x0c, 0xa4, 0xe2, 0x08, 0x65, 0x2a, 0xf0, 0x30, 0x6b, 0x15, 0x59,
        0x99, 0x9e, 0x75, 0x35, 0x77, 0x4f, 0x60, 0x99, 0x8c, 0x8f, 0xd2, 0x2b, 0x21, 0x57, 0xc3, 0xe5,
        0x48, 0xf9, 0x8a, 0x29, 0x50, 0xc6, 0x71, 0x06, 0x89, 0x01, 0x9a, 0xc9, 0x39, 0x04, 0x12, 0xc8,
        0xdf, 0xb1, 0x33, 0x6b, 0xa7, 0x1c, 0x3f, 0x7b, 0x2d, 0x76, 0x3a, 0xaf, 0x76, 0x3d, 0x08, 0x74,
        0x2c, 0xa2, 0xc8, 0xfd, 0x1a, 0x3a, 0x6f, 0x8b, 0xe8, 0xe9, 0xa9, 0xfe, 0x17, 0x0c, 0xed, 0x9d,
        0x40, 0xe6, 0xdf, 0x22, 0x89, 0x4d, 0xea, 0x09, 0x68, 0x96, 0x1e, 0x1a, 0x9c, 0xbd, 0x47, 0x35,
        0x68, 0xd9, 0x4f, 0x5e, 0x12, 0xbf, 0xd6, 0x09, 0x9d, 0xf6, 0x0f, 0xa7, 0xc2, 0xdb, 0xde, 0x70,
        0x35, 0x15, 0x2f, 0x73, 0x16, 0x3c, 0x9a, 0xdc, 0xb5, 0xc5, 0x35, 0x86, 0x8a, 0x31, 0xb8, 0xc1,
        0x74, 0x76, 0xd7, 0x65, 0x32, 0xad, 0xdc, 0x17, 0x1f, 0xfe, 0x85, 0xda, 0x32, 0xc9, 0x1d, 0xda,
        0x36, 0x16, 0xde, 0x76, 0x45, 0x3f, 0x85, 0x8c, 0x8b, 0xdc, 0x37, 0x08, 0x39, 0xef, 0x94, 0xaf,
        0xc8, 0x51, 0x19, 0x29, 0x70, 0x5d, 0xbb, 0x4e, 0xe8, 0xdb, 0xc2, 0xb2, 0x5f, 0x2e, 0xe3, 0x73,
        0xba, 0xc2, 0xa1, 0x42, 0x10, 0xb0, 0xe5, 0xb0, 0x64, 0xb4, 0xdc, 0xbb, 0xa1, 0x51, 0x12, 0x98,
        0xdc, 0x43, 0xcc, 0xc3, 0xc5, 0x25, 0xab, 0x45, 0x6e, 0x63, 0x7e, 0x45, 0x40, 0x63, 0x67, 0xd2
};

void pgm_espgaluda_decrypt()
{
   unsigned short *src = (unsigned short *)PGM68KROM;

   int i;
   for(i = 0; i < nPGM68KROMLen/2; i++) {
      unsigned short x = src[i];

      if ((i & 0x040480) != 0x000080) x ^= 0x0001;
      if ((i & 0x084008) == 0x084008) x ^= 0x0002;
      if ((i & 0x000030) == 0x000010) x ^= 0x0004;
      if ((i & 0x000042) != 0x000042) x ^= 0x0008;
      if ((i & 0x048100) == 0x048000) x ^= 0x0010;
      if ((i & 0x022004) != 0x000004) x ^= 0x0020;
      if ((i & 0x011800) != 0x010000) x ^= 0x0040;
      if ((i & 0x000820) == 0x000820) x ^= 0x0080;

      x ^= espgal_tab[i & 0xff] << 8;

      src[i] = x;
   }
}

Unverified, but apparently it already decripted an ESPgaluda program ROM, so we can assume decrypting 68k's program rom is not as hard as it may seem.
rtw, I see you are a programmer as well as the webmaster of WoA (a site I loved and still love) and knowledgeable on this matter aswell (excellent accomplishment with DOJBL), so some tests could be run on the PROM in your possession.

[rtw]

Creating it is not a problem, as it uses the same XORing as just about every PGM game out there.

The table you posted is for Espgaluda, each PGM CAVE game has their own unique encryption table so this will not work for Ketsui.

[Evrain]

Creating it is not a problem, as it uses the same XORing as just about every PGM game out there.

The table you posted is for Espgaluda, each PGM CAVE game has their own unique encryption table so this will not work for Ketsui.

Realized the massive slip-up, sorry for that, excitement took over logic apparently. I still believe decryption of the 68k prom is not that much of an obstacle though (at most, I see no lack of documentation on that, do correct if needed, at least Galuda was decrypted without too much hassle), given the correct set.

At any rate, nothing can be done if not if/once the romset is available (apart from experimenting on the proms already available for both revisions)... I'm rather curious to hear Guru's reply, given that this is no ZN-1 or Model 3 there shouldn't be another "ImpactEmu" mess.

[rtw]

Realized the massive slip-up, sorry for that, excitement took over logic apparently.

I know the feeling

The Ketsui 68000 program dump is available, think world-of-arcades

The full Ketsui dump is not done even though it is indicated on the list. When the dump is done Dr Decap will still need to decap the protected chip.

[Evrain]

Realized the massive slip-up, sorry for that, excitement took over logic apparently.

I know the feeling

The Ketsui 68000 program dump is available, think world-of-arcades

The full Ketsui dump is not done even though it is indicated on the list. When the dump is done Dr Decap will still need to decap the protected chip.

Thanks for the understanding, it's always a shame to crash against a wall you know is there, but you just weren't thinking about out of excitement Anyway yes, I noticed both PROMS on the site and was thinking of contacting personally IQ_132 once I get rid of a few academic incumbences: we could at least obtain the decryption table for future reference (most IGS027A games are "plug and play" with the current driver, but God only knows what CAVE did when they mashed together the main board and cartridge).

So you contacted Guru and the board is still sitting there, if I understand correctly? That sure is quite a LONG time for a PGM pcb...

[gray117]

Admittedly, such boards usaully get circulated around a bit, but imho its rather more odd, in a general consumer sense, to spend such amounts on a single physical game for a single person/collector, than for a group to spend that amount so as to be intent on getting hold of an expensive game to essentially have it copied.

Now, whilst a lot of people are mostly interested in having a way to play ketsui (and similarly limited/rare games), there is also the aspect of preservation which - even if its secondary - is still an important and recognised aspect of this process.

Clearly in their naivety a lot of the ketsui group expected to soon be able to play the game. That simply wasn't going to happen firstly for technical reasons and finally for potential legal ones. Many people simply have no appreciation as to what emulation means aside from copying the game data, similarly a lot of people have no idea as to what this hardware is.

All in all the ketsui dump group were probably somewhat upset at not playing that game, but in all liklihood were a little more savvy than most emulator users and appreciated their investment at least gave ketsui the best possible chance of getting at least dumped, as soon as possible...

I'm not sure what more you can EXPECT, although you might HOPE for more, when people are doing this service for the love of the gaming in general, and I suspect out of appreciation for the work that goes into making even the 'simpliest' of games.

As for Guru; he will do all that he can to get hold of games ... And much of the time I suspect he simply expects people to donate games to him. In fairness he tries to acknowledge these donations as much as possible. However, he can offend, disappoint and not be clear in his communications ... whilst that may make him seem like a dick to some people, it must be remembered he is clearly just doing what he does is a very determined manner and that there's probably a million and one people to talk to and a million and one things he has to do ... I'm sure he'd want to be his best to everyone in an ideal world, but fck it he's got games to dump

As for the emulator for the hardware - they can eventually be optimised greatly especially with quad (or more core) setups where induvidual chip speeds can be sequenced/staggered across more cores, but this requires time and work... especially if you then try and split theses across gpus which has recently become a consideration for some. For now this bodge, and news, is really just more a coder looking for a bit of credit as to his and others' hard work so far in the first phases of mostly 'preserving the data', before, at some date, they, or someone else, hopefully take on the task of 'preserving the game'.

So well done them

[bcass]

Clearly in their naivety a lot of the ketsui group expected to soon be able to play the game.

Being one of that group, I can tell you that you are wrong in your assumption. I don't think any of us expected it to be emulated soon. What was important was getting the ball rolling, which we succeeded in doing.

[emphatic]

Just think about all the crap that Guru has to take daily from people nagging him to "dump this", and "let me play that". There are many people that believe that just because they've donated funds to the MAME dumping project automatically will be the first to play whatever game they want. I would be* an a-hole too if I've had my mailbox filled with complaints from people who just nag and doesn't do any significantly important stuff themselves.

* I know this is an opportunity for some "rad" or "fun" guy to twist this into a funny quote that I'm already an a-hole, so just don't.

[gray117]

Clearly in their naivety a lot of the ketsui group expected to soon be able to play the game.

Being one of that group, I can tell you that you are wrong in your assumption. I don't think any of us expected it to be emulated soon. What was important was getting the ball rolling, which we succeeded in doing.

Apologies, I stand corrected - that impression was probably more from commentators rather than group contributors proper. And just to be clear I do applaud the enthusiasm and effort in getting such things started

[Dave_K.]

I also donated funds for the Ketsui dump and I already own the PCB!
(of course that means I didn't go as far as donating my PCB, but I think you get what I mean in terms of low expectations)

[pixelcorps]

Seems like I will not be recieving any roms from Guru.

Bcass informed me I was entitled to a copy of the dump according to guru's site / rules, so sent a polite mail asking for them as I sent the board,

he says that after dumping the recent boards, and some complaints or something that occured on Mameworld.. no one can be trusted with any Cave ROMs until such time as they can be emulated without Cave objecting.