one chip console to arcade adapters

[abzman2000]

I had an Idea for creating a universal chip (stock chip with slick code) to adapt console games' inputs to standard input (sit between the controls and the console), here is an example:  a ps1 has 8 pins in it's standard controller, I think it would be cool to take a stock chip and input all the lines (data, clock, ground, etc..) from the console and output up, down, left, right, r1, r2, r3, l1, l2, l3, select, start, analog, led, triangle, circle, square, x, analog x left, analog y left, analog x right, analog y right, and ground all from separate pins to prevent the need to hack any more controllers for custom controls (and it would save a lot of space), one could just buy a set of stock PICs or ATMELs or something and load code onto them.  I mean to do this with all wired consoles (and obscure accessories that need other pins) memory cards and such will be accounted for too in output lines for things like the dreamcast.  

thoughts? questions? has someone done this before (keep in mind, I mean to do it with one chip)? can it not be done for some reason?


ps2 serial protocol info: http://www.curiousinventor.com/guides/ps2
mirrored: http://docs.google.com/Doc?id=d6qm4g8_579crrg4zfc

[SavannahLion]

You mean one chip for every console or one chip for each console?

The former would border on insane and virtually pointless. In my experience, adapters tend to be "married" to whatever they're connected to. So it would be a waste of money to buy a controller that works for every console under the sun only to connect it to one console... ever.

A universal chip where you can load code for a specific individual console would be interesting though. There are a few projects out there that use general purpose chips (like a PIC) any one of these can be adapted to accomplish that kind of interface. But as far as I know, there's no global universal project that brings all of it under one umbrella. I could be wrong though, I lost interest in that sort of project when most of the interesting sites ended up having dead or useless links. Other projects were closed source for some reason and I have zero interest in following those types of projects.

Did you try a Google search? There should be a few projects that might be of interest.

[abzman2000]

what I plan would be open source and using a generic chip load the appropriate code onto it depending on the console you wanted to use it on although maybe I would try to make the code similar enough that if you soldered everything up, and had a working setup for the dreamcast you could desolder the cord, put on one for the saturn and re-flash the chip to make it work (the analog sticks, and the the buttons used similar pins, and I waold always have the input pins from the console use the same x number of pins, meaning they would be vacant when being used with a game system with less input pins.  another idea is to have the code so well documented that anyone could remap the inputs and outputs to any pin they liked, or even omit pins like for example R3 would not be mapped because it is not used, and the person wants to use a chip with fewer IO lines (cheaper maybe).  I want to make some really flexible code here. 

[SavannahLion]

I would be game then. There are a number of things I would really like to see on any kind of project of this type. I'll hold onto those thoughts until later though.

[abzman2000]

for me this is a summer project and probably very long  term as I have a LOT of threads to read on various lists of types of game suggestions, then I have designing my CP and art for my new cab, and all the rest involved in getting that running.  although I have always wanted to learn microcontroller programming this is still a side project. 

another thought: interfacing a bill acceptor with a microcontroller and one of those coin sorters that rotates, fill the "resivour" with quarters and trigger the motor accompanied by a light sensor to find out how many quarters dispensed.  (get rid of the sorter half, just leave the top half)

[abzman2000]

added ps2 info to first post

[SavannahLion]

Here, I'll put my foot in my mouth. There's another thread on this board where someone went and did just that.

I still believe that the board is going to be "married" to a particular set of hardware, but my definition of insane apparently isn't in the same dictionary as what everyone else uses.

<---

[abzman2000]

are you referring to the universal pcb?

[SavannahLion]

Yeah, isn't that pretty much what you had in mind?

[abzman2000]

yes, well mostly.  I would like to find a way to implement all the functions in the chip, rather than having other components, have output pins for rumble motors, and possibly even a flash chip to act as a memory card (not in the "one chip" but still cool...).  also I would have output pins for expansion port items, other than memory cards and rumble.  I have seen the upcb, and although I like how it does most things, there are some things I think that I could improve upon not necessarily mentioned here.   The flash chip I just thought of, but I thought it would be really cool (although hard to implement). 

[MonMotha]

There's nothing that says you can't do this on a single chip.  Worst case, you use a non-volatile FPGA and then you really can do just about whatever you want.

That said, this should fit on just about any medium-large microcontroller.  I have working code for PSX/PS2 (digital only) as well as USB for the Atmel AVR series.  I also have working PSX (but not yet PS2) code for Intel 8051.  I have some docs on NGC/N64 that I've been meaning to play around with some time.  NES/SNES are pretty simple, so that shouldn't be a problem.  Many "really old" consoles like the 2600 are about as simple as you can get.  Dreamcast is a bit strange but very well documented.  If there are docs on making the memory cards work for your console of interest, there's no reason you couldn't use the onboard EEPROM if it's big enough.  Many AVR devices even feature self-programmable flash memory if their EEPROM is too small for your uses, though then you have to deal with page programming.  If you still need more memory, largeish (1-8MB) serially interfaced flash devices are readily available for a couple bucks.

Usually, the major issue is getting documentation good enough to get started.  Inevitibly, you'll end up debugging things with a scope or similar as you compare your device to the "official" devices when the docs are unclear or wrong.  Most console manufacturers are pretty tight-lipped about their stuff, so most of this data is gathered via reverse engineering leaving lots of gaps in the specs.

[abzman2000]

that is really encouraging, I had begun to think I had created an un-reachable goal for myself.  oooh, what about storing save-games on an sd card, I think I have seen similar things done (I don't mean the gamecube sd adapter). 

[SavannahLion]

If you still need more memory, largeish (1-8MB) serially interfaced flash devices are readily available for a couple bucks.

If something like an SD card is used, you'll have my money down. I've intentionally avoided memory devices that are incompatible with what I have now. The ability to swap the same SD cards between the PDA, Cell, Laptop and Wii is a god send. I almost bought this music player that used an SD card for music storage before I realized it was a piece of crap at $45, no screen and no card.

[ShoLouie]

I think what your guys are talking will be amazing if you guys can pull it off.  I'm all about buys a bunch of these if you guys can pull it off.  You got my vote

[MonMotha]

Well, I can certainly do something like this.  Only issue is generally the newer consoles (Xbox, 360, PS3, Wii) due to a lack of documentation on how their controllers work.  Memory card support will also be questionable, but if somebody can dig up specs, there's no reason it can't be done.

Unfortunately, I don't have the time right now.

FYI, in small quantity (a dozen or thereabouts) hardware like this usually ends up costing 3 figures when all is said and done.

As far as using SD cards, that should be doable.  MMC has better public docs, but SD can be made to work with just about any micro, too.  You'd need a larger micro in terms of code size and RAM to handle the FAT filesystem on a memory card, but even really big micros aren't that expensive.