Arcade Rom Patcher (V14)

[GadgetFreak]

So if "someone" wanted to take a look at screen flipping in the code, not naming any name, ahem... @GadgetFreak ... it would be great.

I did take a brief look in Mame and the bad news is that the flip stuff is purely to do with cocktail style flipping and not mirror image flipping. Will take more of a look when I get a chance but not looking good at the moment.

Also just to say that I have just posted V14 with a few more additions

 

[SCUBAking]

I did take a brief look in Mame and the bad news is that the flip stuff is purely to do with cocktail style flipping and not mirror image flipping. Will take more of a look when I get a chance but not looking good at the moment.

Also just to say that I have just posted V14 with a few more additions

Soooooo, after all this I realized I dont even need the flip. With the jamma adapter done, I can pop OT board into RevX or T2 cab that is also flipped and using the mirror, so I think the new shooter cab will be RevX (game is okay, music soundtrack makes it fun), T2 (awesome, lots of time on that one as a kid) and now Thunderbolt.

As always, awesome work on the Patcher, thanks again for all the work you do to make it more fun for those of us who have no idea how to do that kind of stuff.

 

[THEENEMY]

Does anyone know if these patches would work on mister?

 

[GadgetFreak]

Does anyone know if these patches would work on mister?

Mostly I would say no as the majority of the high score saves require a ram chip to be changed out for a Non Volatile version and the mister emulation won't have this.
However, a few like Namco System 1 (Pacmania etc) and M92 (Major Title 2 etc), have an NV ram by default which I use to save high scores too. So if this NV ram has been incorporated and the data saved then yes these ones will work.

 

[THEENEMY]

Mostly I would say no as the majority of the high score saves require a ram chip to be changed out for a Non Volatile version and the mister emulation won't have this.
However, a few like Namco System 1 (Pacmania etc) and M92 (Major Title 2 etc), have an NV ram by default which I use to save high scores too. So if this NV ram has been incorporated and the data saved then yes these ones will work.

Just patched marchen maze and works great on mister .... thanks

 

[Georgian2]

Hello,

I patched the Teddy Boy blues romset (teddybb) using this great tool (V13). The game runs fine. If all the dips on bank A are ON (free play), the attract sound dosen't work. If they are all off, the attract sound works just fine. Anybody else got this problem?

Thanks for sharing this great tool.

 

[Georgian2]

I tried the Sega Ninja too. seganinju romset.

With all the dips on on bank A ON, I can one time press 1P start and start a game. After I die, there is no way to coin up or start the game again. It keeps doing the attract mode as normal but pressing 1P start has no effect., Credit 0 and free play text on screen.

If the dips are all OFF, the coin up works and game starts every time.

 

[GadgetFreak]

Until I finished my multi I had both these versions running perfectly on Sega System 1 boards.
I can only suggest you revert the board to its original roms and check the DIP switches fully. I imagine you have a stuck bit or a failure in the multiplexer on your board.
All of my patches have been checked and used on my boards or with just a handful, a friends board, but I know they all work as listed!

 

[Pr3tty F1y]

@GadgetFreak - The first thing I want to say is THANK YOU! The Arcade ROM Patcher is awesome!

Unfortunately, I have a Pang3/CPS1 issue and my skill is not awesome I have an original Capcom A and C board, but B board is undoubtedly a bootleg. At first glance, it looks identical to an original B board, but something is amiss as my Final Fight is not looking too great.

Game logic seems fine, sound seems fine, but the graphics decoding is a wreck. The original board did not have socketed roms and removing them was painful. However, I only lifted one trace on socket #1 and, as long as I traced the traces right (which I believe I have), I have continuity to the right place for that pin.

I've attached some images of what my Final Fight looks like. The graphics are identical regardless of using the CP1B1F.JED from the burn directory, the original Pang3! gal (which seems to be protected as reading it only spits out all 1's in XGPro), or no gal at all - all the exact same mess of graphics.

I did a lot of multimeter continuity testing, but I'm not entirely sure of what I'm looking for so I may not have covered all of my bases.

Between CP1BF and rom sockets 1 and 7, all continuity is the same. That is somewhat concerning to me, but what I'm saying is that if a pin on CP1BF had continuity to a pin on socket #1, it also had continuity to that same pin on socket 7. I couldn't find any instances where a pin on the gal socket talked to one rom socket but not the other. My findings were:

• CP1BF pin 10 had continuity to ROM pins 12 and 31 (Vss/Ground)
• CP1BF pin 12 had continuity to ROM pins 12 and 31 (Vss/Ground)
• CP1BF pin 15 had continuity to ROM pin 13 (G/Output Enable)
• CP1BF pin 16 had continuity to ROM pin 11 (E/Chip Enable)
• CP1BF pin 20 had continuity to ROM pin 22 (Vcc/Power)

Those all seem reasonable to my novice self, but - again - I'm also wondering why E/G on CP1BF pins 11/13 talk to both rom sockets as I figured the GAL would be bankswitching or something to turn on one rom of the other. However, I could also be very wrong on that and maybe that's not what the GAL is there to do

As far as continuity between rom sockets #1 and #7:

• Pins 1-13, 31-42, and (I think - I shouldhave written this one down) pin 22 all had continuity between the same pins in both sockets
• Only the data lines seemed to be different between to the two sockets

And, again, to my amateur assessmet that makes sense somewhat.

So if anyone has any thoughts as to what may be wrong, let me know! I'd appreciate them. I pulled the chips and dumped them again and they're matching the CRC32's of my roms. Pictures of everything I have below:

 

[GadgetFreak]

It's been a long while since I have looked at CPS1 and at the moment I have too many things on to get the time to pull out my boards and look closer.
But here is the Opal equation file for the JED file the patcher supplies, helpfully I used reasonable pin names so it may help you a bit more:
;PANG 3 B Board (27C322 Gfx Roms)
;SPRITES 0000-43ff B0
;SCROLL1 4400-4bff B1
;SCROLL3 4c00-5fff B0
;SCROLL2 6000-7fff B1

chip P3-FIGHT GAL16V8

A22=1 A21=2 A20=3 A19=4 A18=5 A17=6 i7=7 IN74=8 ROE=9 GND=10
JP3=11 GND1=12 OA19=13 OA20=14 OE=15 CE1=16 CE2=17 CE3=18 CE4=19 VCC=20

equations

CE4 = VCC
CE3 = VCC
CE2 = VCC
CE1 = GND
/OE = /A22 * /A21 * /A20 * /A19 * /A18
+ /A22 * /A21 * A20 * /A19 * /A18 * A17
+ /A22 * A21 * /A20 * /A19 * /A18 * A17
+ /A22 * A21 * A20 * /A19 * /A18 * A17
OA20 = /A22 * /A21 * A20 * /A19 * /A18 * A17
+ /A22 * A21 * /A20 * /A19 * /A18 * A17
OA19 = ROE
GND1 = GND

 

[Pr3tty F1y]

It's been a long while since I have looked at CPS1 and at the moment I have too many things on to get the time to pull out my boards and look closer.
But here is the Opal equation file for the JED file the patcher supplies, helpfully I used reasonable pin names so it may help you a bit more:
...

Thank you very much for sharing that. I tracked down a copy of OPALjr and have it running under DOSBox, so I'm set on that front.

I'm going to have to poke around and educate myself as it's still beyond me as to why the boolean equations are created the way that they are, but after a lot more mulitmeter testing, I'm at least seeing where all of the GAL pins are going (although I'm not sure what they're actually all doing :-D)

 

[GadgetFreak]

The equation I posted is the source file that matches the JED file I generate with the patcher so you don't need to compile using Opal. I posted it as a reference for what pins did what on the GAL.
Are you certain that the programmer is writing the Gal correctly? A friend of mine always gets strange results with those cheap TL866 type programmers when trying to program 16V8D's.
The other thing is the 27322, it should be programmed in byte mode not word mode. Also if your programmer doesn't support 42 pin devices I would presume you are using an adaptor that allows you to select banks which can make it harder to be certain you have programmed the device with the data in the right place.

 

[Pr3tty F1y]

The equation I posted is the source file that matches the JED file I generate with the patcher so you don't need to compile using Opal. I posted it as a reference for what pins did what on the GAL.
Are you certain that the programmer is writing the Gal correctly? A friend of mine always gets strange results with those cheap TL866 type programmers when trying to program 16V8D's.
The other thing is the 27322, it should be programmed in byte mode not word mode. Also if your programmer doesn't support 42 pin devices I would presume you are using an adaptor that allows you to select banks which can make it harder to be certain you have programmed the device with the data in the right place.

I was only referencing Opaljr in the event that my bootleg B board somehow had the address lines scrambled. However, I think I would be reduced to guess work in trying to fix the equations as I haven't quite wrapped my mind around how they're created and what they're actually controlling.

I'm using an XGecu T56, so I don't need an adapter for the 27C322's. Per the XGPro guide, using the default 8-bit data setting (which I'm using) will program in single-byte mode. When I dumped the 27C322's the hashes match the patched files. I just reburned some blank 27C322's and I'm still getting the same scrambled graphics.

I'm also thinking that the GAL programmed properly. I dumped my GAL16V8 and then used JED2EQN to flip it back to the equations. After renaming the pins to match your naming, the only difference that I'm seeing is that it (for whatever reason) flipped A21 and A22 in the order of the equations. I'm thinking that should not matter, but I may be wrong on that:

CE4 = VCC
CE3 = VCC
CE2 = VCC
CE1 = GND
/OE = /A21 * /A22 * /A20 * /A19 * /A18
+ /A21 * /A22 * A20 * /A19 * /A18 * A17
+ A21 * /A22 * /A20 * /A19 * /A18 * A17
+ A21 * /A22 * A20 * /A19 * /A18 * A17
OA20 = /A21 * /A22 * A20 * /A19 * /A18 * A17
+ A21 * /A22 * /A20 * /A19 * /A18 * A17
OA19 = ROE
GND1 = GND

With all of that being said, while my bootleg B board does seem to match the real deal as far as layout (and is so paper thin that holding it up to a light makes it quite translucent), I have made a startling and unfortunate discovery. It's lacking MOST of the electrolytic and SMD capacitors that the real Pang!3 board has. I quite frankly have no idea how Pang!3 was running on it if those components were required (as Pang!3 did run fine). Comparing against MAME, I think the romset was from pang3b3, but the decription of that set in the source indicated that it was on a single board. So I'm not sure where this custom B board is coming from.

I actually just picked up a Mitchell silk-screened Pang!3 B-Board on eBay. I probably won't get it for a couple of weeks - but once I get what I hope is a legitimate board, I'll begin the process of identifying the missing components. I'll reply back if I can get the legit board working. For reference, here's my board:



Here's a pic of the officaial B board that I will hopefuly be receiving and there are A LOT more capacitors

 

[Pr3tty F1y]

@GadgetFreak - I figured out my problem. As usual, it was me.... Upon getting the Mitchell B Board from an original Pang!3, I found the error of my ways. It was the wiring on sockets 16 and 17. I started at the bottom and joined 20 through 13, rather than 19 through 12.... 😕 Dumb, but unsurprising to me

After making the change, I now have two working B Boards 😁 So the missing capacitors on the bootleg B Board seem to have no or little effect. Final Fight runs great.

After swapping around the couple of C Boards I have and between my CPS Dash A board and my long A board, it seems that I have some sort of palette issues going on with my long A board that I'll have to investigate fixing (as seen below with this first enemy that shows up blue instead of gray and has a flicking red glow below), but that doesn't seem to be anything related to the Arcade ROM Patcher as my CPS Dash A board from SFII CE set renders the game without any issues. Thank you so much for creating Arcade ROM Patcher. It's awesome

 

[Scott C]

Just became aware of your super awesome patching tool. Thanks!

Reviewing the games, I see several I could very much like to patch. A few questions for you:
- Strider - Do your patches consider the other patches available to fix sound, graphics, and scene transitions? Trying to determine if I use your patches or those, or if they will work together.
- Quartet - Have you considered a Quartet 2 patch? It's the exact same hardware, with the only difference in gameplay and title screen being due to the ROMs (four fixed players vs. two dynamic players - IE. 4-Player dedicated cabinet vs. 2-player conversion cabinet).
- Flicky - As I am understanding, what you patch are the Flicky ROMs for the Sega System 1 hardware (Flicky (Sega System 1, 5 versions) [FreePlay, HighScoreSave, CustomA], https://system16.com/hardware.php?id=693), but not the Midway Six-Pack hardware. I am trying to help a buddy convert his Midway Up-N-Down PCB to Flicky. Despite lots of searching and multiple people saying those unencrypted ROMs are available, I have not found them.
- Great Sluggers - Namco, NB-1 hardware - Any thoughts about patching this game. I have one that needs repair, but would like to see a HighScoreSave, etc. for it.

Have a wonderful day.

Scott C. (of KLOV, RGVAC, etc.)

 

[ken]

This is a great patch!

This message is a translation function, so there may be some errors.

One thing I noticed is that in "Dragon Spirit", which is compatible with Namco System 1, the game hangs up the moment you pick up item "S".

It's a game with great visuals and music, so I'm looking forward to the next version update.

 

[Pr3tty F1y]

I presume you need an adaptor for an 8Kx8 ram with dual select to be used with an M82 perhaps.
The following board was made for the STK12C68 chips that require a capacitor (sometimes 2), but if you leave the capacitors out will happily work with the FM1608's.
The socket outline is where the NV chip goes and you would put header pins into 2 of the 3 lines of holes depending on if it is a narrow or wide ram chip being replaced.

Here is a zip of the gerber files that will be accepted by JLPCB. They were doing a special of $2 for 5 pcb's and were giving vouchers for free postage when you sign up.

STK-X64K_2020-03-17.zip

GadgetFreak - Can you re-upload the attachment to this post? It's broken on my end and I'm just seeing a link to "uploads/3419/STK-X64K_2020-03-17.zip" and I've had no luck with trial and error figuring out where the file may be living

 

[SCUBAking]

DELETE. My Bad

 

[GadgetFreak]

GadgetFreak - Can you re-upload the attachment to this post? It's broken on my end and I'm just seeing a link to "uploads/3419/STK-X64K_2020-03-17.zip" and I've had no luck with trial and error figuring out where the file may be living

Here it is again. I would imagine it got lost when the forum was upgraded.

 

[Pr3tty F1y]

Here it is again. I would imagine it got lost when the forum was upgraded.

Thank you!