What's new
By:

Conversion of single layer Taito Cup Finals PCB to something else?

Some more progress. I have swapped the 3 sprite roms and 3 tile roms. Cut the old chips out, socketed the spaces and burned new roms.

Now i have this :)

gun lock conversion part 2 - 2.jpg

gun lock conversion part 2 - 1.jpg

gun lock conversion part 2 - 3.jpg

Will get some more pics later, but looks like all graphics are now working.

I still have no sound, but i am waiting for some 27c160 roms to arrive before swapping the ic2 & ic3 roms. So i am hoping this will solve it, as i think these are related to the sound section also. The 68000 processor isnt currently doing anything, no address line pulsing.

I have also swapped 3 pal ics, that were different versions on cup finals. One of them was near the 68000 sound processor (IC14. controls the !E and !G linse to the sound roms. cup finals - D29-13, Gun Lock - D66-09) so i thought that might spring the sound to life, but no cigar.

I have also added some wires to route A17 (pin 30) from the sound roms to the 68K cpu The trace is missing on this board. Also JP1 (which i think controls rom-A17 tied to +5v or to the 68k processor on the normal single layer ray force board) does not link to the pin30 of the sound roms.

Happy with progress though :)
 
digitron said:
Sooooo does/would this work with all F3 roms??
Click to expand...
To my recollection, all ROMs have max 4x program code ICs and max 2x audio code ICs (including Package System ROMs), I think the trickiest bit might be the graphics and sound data ICs, which the package system I believe had more slots for in the cartridge. I think theoretically it should work for a number of games from the package system set, hopefully mypinballs can run some experiments? :) However, I bet there are a few where there just aren't enough IC slots for on the old pre-PackageSystem PCBs. Since we've seen Package System cartridges be made for full PCB games, I wouldn't be surprised if things also work the other way around (barring IC count/capacity limitations).

Most people usually want to go the other way around though, from PCB -> ROM cartridge. :P So it's very cool to see some experimentation going ROM cartridge -> PCB. Thanks again for this! Very awesome :)
 
Apologies, I completely forgot to take pics of my Gunlock board, @mypinballs. Do you think they would still help you?
 
Dariux said:
digitron said:
Sooooo does/would this work with all F3 roms??
Click to expand...
To my recollection, all ROMs have max 4x program code ICs and max 2x audio code ICs (including Package System ROMs), I think the trickiest bit might be the graphics and sound data ICs, which the package system I believe had more slots for in the cartridge. I think theoretically it should work for a number of games from the package system set, hopefully mypinballs can run some experiments? :) However, I bet there are a few where there just aren't enough IC slots for on the old pre-PackageSystem PCBs. Since we've seen Package System cartridges be made for full PCB games, I wouldn't be surprised if things also work the other way around (barring IC count/capacity limitations).

Most people usually want to go the other way around though, from PCB -> ROM cartridge. :P So it's very cool to see some experimentation going ROM cartridge -> PCB. Thanks again for this! Very awesome :)
Click to expand...
On the large roms front, there are 3 extra roms on my white single layer pcb that aren't used with ray force/gun lock as far as i can tell, which were populated for cup finals. i have let these alone so far. They are marked up as locations IC64, IC65, IC66
 
Update. So i have now changed the big sound roms as my 27c160 roms arrived, but i'm still not getting any sound. I was kind of expecting this to be honest as before when converting things you normally get some sound but the wrong samples and samples begin run wrongly.

On this i get nothing other than a click sound as the reset goes high on the 68000. The weird thing is the 68000 isnt doing anything at all, not activity on the address lines or data lines. It was working fine on cup finals so i must be missing something here. Something else must need altering or changing to get it to come to life. I changed the 3 pal ics aswell using the files at jammaarcade.net to here - https://www.jammarcade.net/wiki/index.php?title=Ray_Force.

Cant decide whether the processor should start up first or the pals start first. Normally i'd expect some address lines to be fed to the pals meaning the processor is the first step.

Anyone have some pointers on what i should be look at now. What is 'in charge' of the sound 68000 processor??

gun lock sound roms - 1.jpg
 
Last edited:
Hammy said:
I think the F3 hardware has dual port ram for the sound? ( a bit like m72 )
Same / no sound with the old GAL's? They don't always need swapping ;)
Click to expand...
Thanks for the message. It seems like with the replacement gal in place at ic14 (part number D66-09) there's no activity on the 68000, but if you replace it with the original one then you get some activity briefly before nothing again after the reset line goes high. This looks like it tries to start and then crashes which is more what i'd expect if some address or data line is messed up, so that is encouraging.

I am wondering if IC14 has different or modified input and outputs on the official white ray force single layer board. The thing that worries me is the JP1 jumper (which i believe is to set 1mb or 2mb sound roms) is not the same on the cup finals boards. cup finals does not have any trace going to pin 30 of the sound roms. I added it manually to connect it to the 68k-A18 cpu pin but its weird that its not there.

On the cup finals boards, It looks like JP1-Pin 3 goes to the 68k-A18, JP1-Pin1 is 5v and JP1-pin 2 is to pin31 of the roms?? whereas JP2-pin 2 should obviously be pin 30 and that is what it looks like on the real ray force board? I dont know for certain until i cut the trace between pin 1 and pin 2 on the board. Could it be a mistake on an earlier board revision. Is cup finals before ray force??

Is there anything that would be useful to know from the rom card system around this area? How are different games configured for 1 and 2mb sound roms and what is the equivalent IC14? Is this the same for everything on this system??

Does anyone have a white ray force board that could trace what the i/o pins of ic14 connect to. I can do the same for the cup finals board. IC14 has 2 output pins that control both the sound rom !E and !G lines. So i know this chip is related somehow.
 
@mypinballs I just had a quick look and audio CPU ROMs are twice as big (256KB vs 128KB) and are interleaved. What you show us (IC2/IC3 in 27C160 devices) are sample ROMs. If code halts then problem is more likely with IC10/IC23 and possibly the highest address line is missing.

[EDIT]
Ok it seems you already did the A17 mod. Were pins 30 tied to Vcc before the mod? Are other address lines going directly to the 68000 without any buffers in between?

Oh, and since Cup Finals uses 128KB ROMs (1Mbit), have you checked they are JEDEC devices?
 
Last edited by a moderator:
Darksoft said:
What do you guys need to get the ball rolling?
Click to expand...
Spare time haha
I hate desoldering roms that are already dumped when there's stuff in line that actually needs dumping ;)

As you know there's much better shooters on the table at the moment that need finishing ;)
I'd also say taito F3 is already been nailed putting this lower on the ladder again.

I do think this single layer board will run everything but each game will need a tweak somewhere.

At this stage i'd try another title such as top ranking star or arabian magic and see what's going on with the other known single layer titles on this setup.
 
Apocalypse said:
@mypinballs I just had a quick look and audio CPU ROMs are twice as big (256KB vs 128KB) and are interleaved. What you show us (IC2/IC3 in 27C160 devices) are sample ROMs. If code halts then problem is more likely with IC10/IC23 and possibly the highest address line is missing.

[EDIT]
Ok it seems you already did the A17 mod. Were pins 30 tied to Vcc before the mod? Are other address lines going directly to the 68000 without any buffers in between?

Oh, and since Cup Finals uses 128KB ROMs (1Mbit), have you checked they are JEDEC devices?
Click to expand...
Thanks for the reply. I think my earlier picture may have bene alittle confusing. I was just posting a pic of the sample roms as those were the last 42 pin chips i had desoldered and swapped from the conversion point of view. Yes i have been looking at ics IC10 and IC23 and the gal at IC14.

Pin 30 on the sound roms was not tied to anything before i manually added the connection to the 68k cpu

I will see what i can work out for connections to the sound roms and also ic14. It would be useful to know if there are differences from a real gun lock white pcb at this ic14 location.
 
Going to keep chipping away at this, as i'm not in a mega rush to solve it and i need distractions at the moment. Better to be head in electronics and problem solving rather than thinking about the world and what's happening.

So i'll keep sharing stuff on here and info as i find and work out things. I feel like i've made good progress on this and dont fancy giving up now. Hopefully this info will be useful and can be included in bigger resources.

Plugging the original sound roms with the original D29-13 pal restores the music, although its not playing in the correct places (expected). No samples or effects though (also expected at this point)

So, here is my original ic14 pal ic from what is connected on the cup finals pcb. (Not 100% completed yet, but its what i have so,)

View attachment U14 PAL Cup FInals Single Layer PCB.pdf

I have also managed to read the D29-13 pal from the cup finals board original and have a jed file for this. Checking the checksum with this and the D66-09 files gives

D29-13 - 9BE1h
D66-09 - 9BF3h

so there not a massive difference but there is something different. I wanted to make sure these ics were not just the same but different part numbers etc etc

So i think i'll see what the mame equation dump tool has to say for these 2 files and then will report back.
 
Another small update. I have had my head in pal jed files and jed to eqn and eqn to jed convertors for the last week!

I can report some progress in that i now have about half the sounds working :) Yay!

But i'm still not there completely so more work to do.

The first problem i had was to extract the equations from both IC14 pals to compare. It turns out that mame's jed util doesnt yet support the pal type that i needed - GAL20V8. Typical! SO i found the national semiconductor tools and proceeded to learn how to use these. I have to say these are great and something i will certainly be using in the future for all sort of things.

So after managing to get both equations files i worked out the pal change is indeed to support the high rom sizes and basically removes the A17 line check from some enables for the roms and rams.

Next issue was that the only pals i have (and can buy as equivalents) for GAL20V8's are ATF22V10's so i had to convert the jed file to be in the correct format for the new larger type gal. This was the main problem from earlier (other than the fact the board has a mistake on it for the A17 address line sound roms jumper JP1 and needs adding manually.)

So putting this newly created gal in gave me some sounds, but not all. I can hear all the sounds from the first part of the attract mode, but there is no music and some sounds in game play are still missing.

Going into the test mode, gives me sound up to about address 0x39h but nothing after that. I'm trying to work out what the significance is of 0x39h (57 decimal) (00111001 binary) but not found anything yet. Something must be slightly different address wise somewhere, but the highest A19 line on the sound sample roms, is pulsed when you test sounds, so that part is ok

Feel like i'm pretty close to fully working board now.

Music seems to start around id 0x5Ch in the test. Does anyone know where the music data is stored, main sound roms, or sample roms?
 
You must log in or register to reply here.
Back
Top