Official: CPS1 (CPS1.5) Multi by Darksoft/Apocalypse - Interest Check Thread
- Thread starter Derick2k
- Start date
J
jassin000
Yea this is OG hardware, now don't get me wrong I'm glad projects like Mister exist.
I just don't want my multi experience to involve one.
I just don't want my multi experience to involve one.
rockbottom
Enthusiast
Have you considered some kind of dynamic register patching instead of writing keys to a b21?
eg. lets say game wants ctrl reg at 0x2c but b21 is at 0x26, just redirect writes to 0x2c -> 0x26 via your logic. Should be quite easy assuming you have fpga somewhere on address bus between b and c boards.
Could handle layer enable masks in similar way too.
You could even then support other c boards but would run into lack of multiply protection support on the pre-21 chips (but your fpga could handle that perhaps).
Just a thought
eg. lets say game wants ctrl reg at 0x2c but b21 is at 0x26, just redirect writes to 0x2c -> 0x26 via your logic. Should be quite easy assuming you have fpga somewhere on address bus between b and c boards.
Could handle layer enable masks in similar way too.
You could even then support other c boards but would run into lack of multiply protection support on the pre-21 chips (but your fpga could handle that perhaps).
Just a thought
ShootTheCore
Legendary
I'm not really understanding the argument for "originality" on the hardware of an arcade Multi.
The Neo Geo uses custom serializer and address-mapping chips inside their cartridges, like the NEO-PCM2, NEO-ZMC2 and NEO-CMC. But I don't have to transplant a donor for those chips with the Neo Geo Multi.
Some Sega STV carts (like Radiant Silvergun) use a 315-5881 custom for security. But I don't have to transplant a donor for the STV Multi.
Why should the CPS1 Multi require any original donor chips - especially if code is (or soon will be) publicly available for an FPGA implementation of that hardware? The B21 customs are known to have a high rate of failure.
Also, if you really want to be "all original" wouldn't you be swapping appropriate C boards with the B custom for the specific game you want to play? I understand that the current CPS1 Multi is patching non-B21 games on the fly to use B21, but if there was an FPGA, the ROM could be left unaltered and the FPGA could instead be setup to simulate the appropriate B custom.
Last edited:
laevateinn
Student
I think it's best for preservation to do the thing that is least likely to lead to someone destroying existing hardware if it's not too difficult.
XtraSmiley
Legendary
Agree.
Not trying to beat you up, but wanted to address a few items:
-ud
I don't believe Jotego has 1:1 recreated the CPS1 customs ("A" and "B"). I think he's done what is necessary to get data where it needs to go with proper timing. I would love to be proven wrong, but I believe that to be the case as his core stands today.
Not to my knowledge. I've encountered a few failed units, but 100x worse is the "A" custom on the CPS1 A-board.
Truth.
Non-B21 ROMs do not require patching to work with B-21, hence why B-21s are sought after in the context of this project. Just need to send B-21 the proper config to act like desired B-XX chip. That's what InfiniKey CPS1 does (which is still in production limbo *bitter tears*).
-ud
A
Apocalypse
It's been done and tested years ago (2014).Have you considered some kind of dynamic register patching instead of writing keys to a b21?
eg. lets say game wants ctrl reg at 0x2c but b21 is at 0x26, just redirect writes to 0x2c -> 0x26 via your logic. Should be quite easy assuming you have fpga somewhere on address bus between b and c boards.
Could handle layer enable masks in similar way too.
You could even then support other c boards but would run into lack of multiply protection support on the pre-21 chips (but your fpga could handle that perhaps).
Just a thought
aje_fr in France even used that on the fly patching for his multi.
Multiply protection was made by FPGA indeed.
Kick harness/3p-4p were deported on the B-board for the C-boards lacking connectors.
All of this has already been discussed with @Darksoft when we started the CPS1 multi project.
Last edited by a moderator:
There is no perfect solution that relies on B-21 chips because there is a limited supply of the chips. Unless someone comes up with an FPGA B-XX replacement, the best solution is one that accepts any C-Board / B-XX chip, like aje_fr's CPS1 multi.
-ud
I wonder how the guy from China figured this issue out or if he just copied someone's work around? I picked one of his multis up out of curiosity. Its good, but the CPS changer SFZ is all wonky since the JAMMA pinout is way different from the CPS1. On this board the C board connectors for the kick harness have been transplanted to the B board. Picture of the board Here
That is exactly what the guy from China did. Read Apocalypse's post on that thread. Plus the last 3 or so posts in that thread and key in on the giggles.
arcadeaction
Enthusiast
Please put me down for a second kit also, thanks
For the moment, your best bet is to source a battery-backed B21 C-board, suicided or on battery. If it has the de-suicide mod you might need to revert the mod.
But by the time the multi is out, it might support any kind of working C board, we'll see.
But by the time the multi is out, it might support any kind of working C board, we'll see.
PascalP
Legendary
Yeah don’t go crazy until final needs for the multi is confirmed 
No. The final details haven’t been released. It’s best to wait before assuming anything.
Brentradio
Champion
