Noob to EPROM programming question

AtomicBomb

Member
Joined
Sep 11, 2011
Messages
402
Reaction score
9
Location
Boston, Massachusetts
I'm having some trouble with my Gyruss machine, and have narrowed down my problems to EPROM issues.

I'm still new to reprogramming EPROMs, but I think I know enough to make me dangerous. My programmer seems to be functioning and reading data properly.

So the problem I'm having is that I downloaded the Gyruss rom set from teh internets and there's a different number of files than I have EPROMS for. The Gyruss data set I downloaded (which seems to work just fine with MAME) has 9 files labeled gyrussk.(1-9), I have a gyrussk.(1-3)a, and I have gyrussk.pr(1-3), and I downloaded 2 different copies from different locations, both with exactly the same data.

My Gyruss eproms are labeled 1-13 (missing #4). Gyruss is a 2 board set and the sound and power board contains Roms 11-13. I THINK these refer to gyrussk.(1-3)a but I'm not sure. Could be gyrussk.pr(1-3). I opened gyrussk.pr(1-3) and they all contain VERY little data. I think 1 and 2 contain 16x16 blocks of data, where as 3 only contains 2x16.

Either way, I'm more concerned about the main set, and the fact that they're labeled 1-9 where as I have 1-10 with 4 omitted. Should I just be subbing in set 4 from the MAME rom into my #5 EPROM and so on?

Whats throwing me off is I started with EPROM #1, read the data off of it, and started comparing it to ROM 1 from the MAME data, but they were DRAMATICALLY different. I did the same with #2, and again, DRAMATICALLY different. Should I assume they're all bad? Should I re-burn the whole set? Is it okay to drop 1 chip worth of "MAME data" in with the rest of my supposedly good eproms?
 
Sounds like your Gyruss may be a bootleg. Does the PCB have a sticker that reads "TAITO (NEW ZEALAND) LTD"?

The MAME driver (http://mamedev.org/source/src/mame/drivers/gyruss.c.html) has some good info about the various Gyruss romsets. Look at lines 590 to the end. Pay particular attention to the text off to the right, explaining the labels on the bootleg's EPROMs.

The question is: why are you programming all new EPROMs? Does the game self test say they're bad??

I don't just program new EPROMs for no reason. I usually pull them, read them with the programmer, and submit them to a romident site (I use http://romident.coinopflorida.com/) to see what they are. If they identify, they're likely OK, and I know what they are. If they don't read properly, or don't identify at all, then I consider programming a replacement.
 
No, I don't think it's a bootleg. PCB say Konami all over it. Every processor is branded with the Konami badge. It looks identical to the one posted at Crazykong ( http://www.crazykong.com/pcbs/G - O/Gyruss.pcb.html ) and it came out of a dedicated cab. It even has the same Mitsubishi eprom window stickers with the same font stamped on them with 12 and 13 written by hand. Absolutely identical to the CK board.

I don't see a Taito sticker on it; why would it have one? It's a Konami game licensed by Centuri. Where does Taito factor into it? I'm actually really confused by the mamedev link you sent me. It keeps mentioning Taito, but why? Is the bootleg a Taito build? Also, on that one, they have roms 1-13 listed, but 10 is omitted. Mine has 1-13 with 4 omitted. What gives?

I'm looking to reprogram roms because I have no text at all in the game. Nothing on the Title screen, no score, no in-game messages. I'm also missing all of the star fields, including the one around the planet on the title screen.

I already corrected one issue with this board that was caused by a bad eprom socket (graphical anomalies related to enemy ships). In addition, the eprom I suspect problems from had it's window sticker almost entirely peeled off.
 
OK, it may not be a bootleg.

But the the advice to pull them, read them, and try using an ident utility remains the same. Start with the one you suspect.

I use my EPROM programmer more often to read than to write.
 
Assuming its not a bootleg there are a couple of reasons for what you are seeing, firstly games that were popular were released in different versions, so your version may not be the one that ended up in MAME. Not every version of every game was MAMEd, so just because you get an unknown read doesnt mean the ROM contains crap. Also on the early ROMs it is worth using the "slow read" option on your reader (assuming it has them), often modern readers try to pull data too fast for the very early ROMs, far faster than the board itself tries to read them.

Secondly if the game was popular then the PCB may have had a second production run and at that point the price of EPROMs may well have shifted. Games on v2.0 boards may well have been made when it was cheaper to use larger capacity EPROMs and half the number. Eg instead of using 6 27C32s, they used 3 27C64s and each larger ROM contains the contents of two smaller ones, at a board level this is incredibly easy to do, and a give away sign is if your board has gaps inbetween the chips where other chips could live, there might or might not be empty sockets in these locations. The ROMs cost an absolute fortune back when these boards were made so the price point at the time they were made dictated which chips got used.
 
Let's see a picture of your cpu/video board (the bottom one). Sounds like you are missing the character EPROM (which would be labeled "4"). From your description of the problem, that should be the only EPROM you're looking at anyway (errors in the others would not cause what you are seeing).
Also, the differences you're seeing in ROMs 1-3 are probably just because you have the Centuri version and you're comparing them with the Konami version.
 
Here's a pic of the board. Like I said, GY 1 and 2 window stickers were peeling and are pretty scratched up. Not sure if it's enough for them to be corrupted, and I was getting what appeared to be "good data" (not all FFs) off of them, but I don't know what "good data" is supposed to be and I didn't go through every line. I get what you're saying that this could have been a later production run, and these are all 2764's; I just don't know where to get the data that I need if it is indeed corrupt.

http://sphotos.xx.fbcdn.net/hphotos-snc6/s720x720/283786_4062486517559_1592982994_n.jpg
 
Don't worry about scratched window stickers....
Your problem is the EPROM at location H1. Program a chip with the image "gyrussk.4" and put it there.
 
I thought that was just a seriously smudged 6.

Looked to me like 4 was missing from my set and maybe in the first production run would have gone in H14. You're saying instead that 4 should go in H1 and I should be missing a GY 6 from my rom set?
 
I thought that was just a seriously smudged 6.

Looked to me like 4 was missing from my set and maybe in the first production run would have gone in H14. You're saying instead that 4 should go in H1 and I should be missing a GY 6 from my rom set?

You're correct. My apologies. H1 is ROM 6. (there is a pic of it on my score save page: http://www.scoresaves.com/TPGyrHS.html)
J14 is normally empty (it is the test ROM socket), so that is not the location of ROM 4.
I will have to pull out a Gyruss board and look tonight to see where ROM 4 goes (as I can't find a good pic).
 
For the record, the confusion here is all due to poor ROM naming in MAME. The names/numbers in MAME do no correspond to the EPROM labels on the board. My original statement was actually correct:

Your problem is the EPROM at location H1. Program a chip with the image "gyrussk.4" and put it there.

Although the ROM at H1 is labeled "GY 6", it is known as "gyrussk.4" in MAME.

Remember when the ROM image archive ZIPs used to contain simple text docs that the dumper would include that detailed ROM locations and other board details? These useful docs have all been stripped out by the MAME ROM managers/cleaners since they are not "needed".... :rolleyes:
Some of that info did make it into the comments of some of the MAME driver code, but much of it has been lost.
 
Remember when the ROM image archive ZIPs used to contain simple text docs that the dumper would include that detailed ROM locations and other board details? These useful docs have all been stripped out by the MAME ROM managers/cleaners since they are not "needed".... :rolleyes:
Some of that info did make it into the comments of some of the MAME driver code, but much of it has been lost.

Of course its "not needed". We use MAME for a completely different purpose, to repair PCB's, than what it has become to the rest of the internet. IE: for everyone to bootleg classic arcade games under the guise of "preservation".

I have a complete set of the latest MAME build with ROMs... I use it for two purposes :

1 - Repair broken PCBs by burning new EPROMs or watching how a game is supposed to run (sometimes).

2 - To try out a game I'm not familiar with before I buy a new PCB.
 
Of course its "not needed". We use MAME for a completely different purpose, to repair PCB's, than what it has become to the rest of the internet. IE: for everyone to bootleg classic arcade games under the guise of "preservation".

I have a complete set of the latest MAME build with ROMs... I use it for two purposes :

1 - Repair broken PCBs by burning new EPROMs or watching how a game is supposed to run (sometimes).

2 - To try out a game I'm not familiar with before I buy a new PCB.

Another thing that kinda rubs me the wrong way is when people refer to them as "MAME ROMs", as if they somehow belong to MAME and wouldn't exist otherwise. The fact is, archives of classic video game ROMs existed well before there was any emulation. Although we can certainly thank MAME for the boom in ROM dumping and the ease of finding ROM images today. MAME is great tool for me. It's just a little sad when you see the preservation aspect clearly neglected....
 
That's an issue with people distributing the ROMs removing "unnecessary" files, and has nothing to do with the developers. The MAME team have no control over what is or isn't in the zips of ROM files. The only thing they have control over is the source. I keep a copy of the source for reference, and I often find *very* helpful info in its comments, including notes from the ROM dumper, and ASCII art of PCB layouts including component IDs. In general, over the dozen+ years I've been messing with MAME, the trend has generally been toward more accurate representation of the original functionality, rather than just "make it playable" (often to the chagrin of those who just want it playable).

Could the info in the source be better? Certainly, in many cases. But overall, I consider MAME (again, particularly the source) an *extremely* valuable tool for PCB testing/diagnosis/repair. Oh, and it's handy for ROM hacking as well ;)
 
J14 is normally empty (it is the test ROM socket)

Glad you chimed in here, as the guy who wrote the HSS kit should know a little about the hardware ;)

Speaking of the J14 test ROM socket... do you know if there is a Konami test ROM image floating around for use in this socket?? The lack of any test mode on these Konami games is notable...
 
That's an issue with people distributing the ROMs removing "unnecessary" files, and has nothing to do with the developers.

Yes, of course, it is clearly not the fault of the emulation developers. Developers of tools like clrmamepro, however, are to blame.
 
Speaking of the J14 test ROM socket... do you know if there is a Konami test ROM image floating around for use in this socket?? The lack of any test mode on these Konami games is notable...

None have ever surfaced, and I doubt they ever will. I think they were for factory use only (not field service).
Some 80's Konami games do have a test mode built in. Hold down both Start buttons on boot.
 
Back
Top Bottom