What's new

xodaraP

Legendary
Joined
Sep 21, 2016
Messages
4,750
Reaction score
3,793
Location
Gold Coast, Australia
Hi everyone,
I got this yesterday in a lot of faulty boards a friend bought - a GunNail bootleg. Since the game is so rare I thought I'd grab it even though it was a bootleg and hopefully fix it up. I can't find anything online to suggest that a GunNail bootleg has been seen in the past - nor any record in MAME that there is a GunNail bootleg set.

The board itself is enormous - it's almost as wide as 2 CPS1 boards sitting together and almost entirely made up of discrete logic with a few PALs and a TI FPGA

Here's a photo of it, once I get to it in the repair queue I'll dump the ROMs and if they aren't currently in MAME, I'll make sure they're added.

Hopefully it can be repaired and I can add it to the collection

IMG_3338.JPG


If you have any questions or would like photos of something specific - let me know :)
 
Last edited:
Ok so I quickly pulled the 2 main program ROMs off and put them into the programmer to dump them and see what comes up.

1 of the 2 main program ROMs is completely dead. It's got a lot of corrosion on the pins and the socket. I can replace the socket but the ROM comes up as unmatched device (30 pins) - it's a 32 pin 27C020, and since there is no bootleg version of GunNail currently in MAME that I can find, that's it for now until hopefully it surfaces at a later date.

I'm going to submit the other ROM that did dump properly and see if it's something known that hasn't yet been added or if it is a totally new dump. Even if it is a bootleg it still deserves to be preserved I think.
 
That is a tough break. Would have the bootleggers hack the rom? Thinking about it if the board layout and components have changed then there must be some changes to the rom?
 
This board is a beast! Great find!

Never seen before! and seen a lot of bootlegs...
Seen very similar NMK bootlegs tho Missing the 2nd oki chip or hacked raiden sound.
Gun nail uses the NMK 004 so the sound ROM could be anything! is very interesting if you can dump that ;)

Without the rom dump you're playing a waiting game for a next one to appear and land in good hands.
Try a few different programmers, if the ID is the holdup run without see if the dump is empty. Chip ID can be dodgy sometimes especially on bootlegs.
Keep trying on different programmers and hope for the best.
 
That is a tough break. Would have the bootleggers hack the rom? Thinking about it if the board layout and components have changed then there must be some changes to the rom?
The original game is protected so they would have had to modify the ROM to remove the protection. The hardware has been well bootlegged though, Thunder Dragon and USAAF Mustang bootlegs are common. This is a completely different layout to those though

The layout is vastly different to an original, there's an FPGA that I assume takes over duty for the main custom and several PALs that do the same for the other customs - with everything else being recreated in logic
This board is a beast! Great find!

Never seen before! and seen a lot of bootlegs...
Seen very similar NMK bootlegs tho Missing the 2nd oki chip or hacked raiden sound.
Gun nail uses the NMK 004 so the sound ROM could be anything! is very interesting if you can dump that ;)

Without the rom dump you're playing a waiting game for a next one to appear and land in good hands.
Try a few different programmers, if the ID is the holdup run without see if the dump is empty. Chip ID can be dodgy sometimes especially on bootlegs.
Keep trying on different programmers and hope for the best.
I can certainly dump the sound ROM, it's soldered though so it will have to wait until I work through some other repairs and I'll remove and dump it

The ROM that's not working is an AM27C020 and the other one read fine, this one is heavily corroded and made all kinds of horrible noises coming out of the socket (I was as careful as I could be) and reads as 30 pin so I assume 2 pins are internally disconnected :(

I did press ignore in the programmer and had it dump what it could and it did that fine, I didn't read it yet to see if it's just garbage or all zeroes
 
You may be lucky and have part of the file in the dump. You can try the OG roms on there but expect sound issues MINIMUM if it fires on at all.
Thunder dragon 2 bootleg is/was very close to the original, just the sound changed.

I still gotta dump the mask roms from the TD2 bootleg (and about 20 other boards), so know how it is! Don't stress over it :)
 
I would expect given the hardware differences it won't even fire up but I guess it's worth trying with the OG ROMs and seeing what it does

I assumed the main CPU code was protected, the one ROM that does dump ok doesn't match in ROMident

I'll definitely have a look and see what comes up in the dump or if it's junk
 
it looks like this bootleg attempts to actually recreate the sound section accurately, so this is probably a better bootleg than that Batsugun one that popped up a while ago.
 
That's even better, I really want to get it going. I might put the ROMs aside and load a fresh pair with the non bootleg CPU code and see what it does. I'll need to replace the sockets too. Hoping the corroded one doesn't take the traces with it :(
 
it looks like this bootleg attempts to actually recreate the sound section accurately, so this is probably a better bootleg than that Batsugun one that popped up a while ago.
It's close, but missing an OKI and it's sample rom so will be missing 1/2 the effects (normally the drums)
 
it looks like this bootleg attempts to actually recreate the sound section accurately, so this is probably a better bootleg than that Batsugun one that popped up a while ago.
It's close, but missing an OKI and it's sample rom so will be missing 1/2 the effects (normally the drums)
Ah, I forgot the original has two OKIs. That's so strange, for the bootleggers to come so far and then nix the last OKI.
 
So a quick update on this - I have replaced the bad socket (I'll post up a photo later, it's quite bad) and attempted to dump the ROM that was in the socket ignoring that it registers only 30 pins out of 32 hoping I'd get something out of it. I got nothing but FF and 00 with the occasional bit of code thrown in, it's unfortunately very, very dead :(

I'll have to get some 27C020 EPROMs and try the non bootleg code, see if that helps. In the meantime I have attached the code from the other EPROM that did dump, not sure if it's garbage data or not but I looked in the buffer data in the programmer software and there's definitely plenty going on.

If someone wouldn't mind having a look at this for me and letting me know if it's just pure garbage or if this is in fact half of something that would work with the other half, I'd be interested to know. Device is an AM27C020

Edit: In the hex editor it actually looks really similar to the original code. However when doing a data compare with the programmer it tells a different story, it's not massively different but different enough. Kinda cool it is that close.

Edit again: I've decided to also attach the comparison file (shows the sections of data that differ from original) - I'm assuming this is either an encryption thing or the modifications to the sound system against the original

Filename of original MAME ROM used for comparison: 3e.bin from set GunNail (28th May, 1992)
 

Attachments

  • gunnail_27c020_6d.7z
    38.1 KB · Views: 406
  • gnbootcomp27c020_6d.txt
    6.4 KB · Views: 320
Last edited:
Further update. I burnt the original version ROMs and I managed to get it to give a work RAM error when I was trying to annoy the fault. I desoldered one of the mainCPU RAMs to test it as well (it tested ok in spite of being rusty) and socketed it.

There are so many different types of RAM on this board I don't even know where to start for work RAM, I assume it's the 8 62256 on the opposite side, however in the MAME driver this is right next to the sound custom, so is this sound RAM?

Edit to further update: I pulled the FPGA and booted the board without it, and it reacts very differently. I assumed the FPGA just had to do with sound but clearly it's doing more than that. I'm guessing the differences in the bootleg ROMs will be to address this FPGA in place of whatever chip it is replacing (I assume the NMK004)

Without being able to get hold of a working set of the ROMs - I'll have to shelve this for now which kinda sucks. The other issue is there may well be other board faults and without a known working ROM set for the board, there's no point in me trying to troubleshoot further.
 
Last edited:
Further update. I burnt the original version ROMs and I managed to get it to give a work RAM error when I was trying to annoy the fault. I desoldered one of the mainCPU RAMs to test it as well (it tested ok in spite of being rusty) and socketed it.
They likely moved the work ram address.
The work rams are the 2x 62256 by the program roms.

Is the attached rom the 'bad' rom? If not, upload the bad rom too and i'll see if i can work out what's going on.
 
Last edited:
The attached is the good one, I'll upload the bad one soon. It's almost entirely blank.

So the work RAM is the 2X 62256 right next to the CPU?
 
I had a quick look.
The changed locations don't hit on boot

0000327A 4E 67
0000327D 4E 33

Could be 4e71, 4e73 or 4e75. etc


0000A00A 60 67

Standard branch (during gameplay)

So from the data provided the original set will boot. These checks start ingame.
Does not account for any changes to the 2nd rom.

You may as well change the 2 work rams at this point.
 
Last edited:
That ROM I think is the second part of the CPU ROM space, hopefully the first part of the ROM is still the same or close enough to original to boot

I socketed one of the work RAMs but tested it off board and swapped it out with no effect so I guess I'll try the other one.

There's 2 ROMs for main CPU: 3e.bin and 3o.bin - the 3e equivalent on my board is fine, the 3o equivalent is the dead one. I've replaced the ROM, socket and RAM for that side.

It just cycles through what I assume is a RAM test rather than booting. The original board doesn't seem to do any such tests either which is weird.
 
Last edited:
I just noticed this is another one of those terrifying boots that has no decoupling capacitors. Once you get this working, I'd recommend peppering it with 0.1uF ceramic caps all over the board, first prioritizing RAM chips, the FPGAs/CPLDs, sound and video DACs, and then logic as much as you can tolerate soldering.
 
I just noticed this is another one of those terrifying boots that has no decoupling capacitors. Once you get this working, I'd recommend peppering it with 0.1uF ceramic caps all over the board, first prioritizing RAM chips, the FPGAs/CPLDs, sound and video DACs, and then logic as much as you can tolerate soldering.
I did notice that it had very few passive components in general but didn't notice the lack of 0.1s at all. Crazy.
 
Hi,
I might be able to help a little bit here since I have the same board.
The board booted to a static screen of garbage when I got it last week. After fixing a trace close to the 68000, it is now working with some noticeable faults:
(1) 2/3 of the background is missing
(2) the B button of P1 is not working
(3) the ship speed is way too high
(4) the sound is clearly not close to the original pcb (but this is probably related to the points mentioned above).

I will try to see if I can fix points (1) and (2) when I have some time, but my repair skills are limited...
I do not know if there is anything to be done to fix (3) and (4).

Anyway, I dumped the two main ROMs, which my programmer did not report as faulty.
I can of course provide them if it can help to fix your board xodaraP.


d68ukcu.jpg




RgsyCWw.jpg
 
Last edited:
Back
Top