Cartridge Selector for Home Consoles [Project]

General concept is from an old accesory that is hard to find now, there are some rumors it never even came out.

Just like an AV switch, but instead of video it selects cartridges. Always wanted one of these, and even more so now as the consoles and carts get older and more picky about when they want to work (this way you could get them in the position to work and just leave them there). Anyway was wondering if anyone had one of these, or pictures of the insides. I wanted to try to make one for some home cartridge based consoles out there using console pin connector, old IDE ribbon cables, a project box and some switches. Gonna need some replacement pin connectors for 8-bit/16-bit consoles. The main one I want to try first is a SNES but the cheapest I can find 62 pin connectors for is $20 each. Then I would get one of those HoneyBee connectors for the SNES that has all the pins (not the cheap ones that are missing the pins on the sides) and wire it to that, so I could attach it to any SNES/SFC/Korean SNES/PAL SNES. It would most likely be easier to start with one for the NES as 72-pin connectors (as I can get those for about $5 each). Also the SFC hotel units from japanese hotels in the 90's have a similair concept. Any ideas on what I would need to move forward besides the stuff I mentioned?

There was also a GBA game selector for the GBA SP that clipped on the bottom and even one for the DS. I think this could actually be a pretty popular item if ever mass produced. And one called a "Game Brain" for the 2600. Pictures below...

 

I would soo love to own a video jukebox. I would sit my model 1 genesis and model 1 CD on it, plus a 32X. It would be the ULTIMATE sega tower. hands down. Oh, and i'd put a game genie, sonic and knuckles, and sonic 3 in every cart slot

 

Tip: cartridges shouldn't be switched with switches.

1) You will be hard pressed to find switches large/complex enough to fully switch between even two carts. It'll end up requiring a ton of switches. It will look hideous. Pretty much any DIY game selector is guaranteed to be hideous.

2) The IC inputs in the cartridge will have clamp diodes which will forward conduct and power the chip, leeching current from the bus and possibly causing bus conflicts/other chaos. The proper solution is to power each cart and gate the cartridges with digital logic, while taking the cartridge timing constraints into account. This is not so trivial, especially if you don't have intimate knowledge of computer systems.

2) Putting a bunch of carts in parallel presents a severe capacitive load to the buses, especially when you connect them via IDC ribbon. You must buffer the buses, again while taking timing constraints into account.

 

Well how about something like this.... (It's an early morning crude picture in photoshop kind of day)



Instead of using a swtich, you would wire the ribbon to a piece of metal that would slide over the contacts of another piece of metal that has rounded pins (which was then itself wired to a 62 pin connector). This slider would being pressing up against the rounder pins and touching it with its flat metal contacts attached to the slider.

This would remove the need for a switch, you would need to turn the power of the console off of course. And if there was worry about wear on the contacts you could make it so the slider control only slides when it is pressed down, after aligning it with the cart you want, you let go and a spring presses the two contact boards together. You could also have plastic grooves to make sure it aligns right.

 

Switching game whilst the machine is on would probably damage the game and the machine using that method unless you had some sort of solonoid to lock the game in slode switch in place....

Older machines are not too difficult to build a cartridge switcher as all you needed to do was buffer the data lines, and use a 3 to 8 line (or 4 to 16) selector like the 74ls138 and a not gate to pull high or low one of the chip select lines to select the cartridge (the other would remain low . Most required you to switch off the machine which was a pit pointless. By using the PSU of the machine to the unit and using a relay and a simple reset circuit using a cap and a resistor you could have a reset which reset the machine by turning it on and off. The NES and SNES with their lock out chip would require a bit more logic but again could use the same basic theory as could the Master Systen and Megadrive.

 

I agree but, I have no idea why I would not shut of the console to switch games? The selector idea was just a way to make it so I didn't have to take out and put in new carts and get them to boot up (which can take some time with a heavily used SNES and/or finicky games). I am already getting up to switch the game with the selector, it would take very little effort to switch the console off when I do so, and then back on when I selected the game. In the end this method is saving me time as well as catridge and pin wear.

Having it so you didn't have to shut off the console is nice, but not really a big deal in my book.

Really appreciate the info you are bringing to the table!

 

? You can switch carts on most consoles without the system even crashing. Unless you cause a bus conflict from contact bounce it shouldn't be an issue. With the SNES there's a global open-collector reset signal so switching can be made seamless.

Disabling the data bus won't stop on-cart SRAM from being trashed so gating the read/write strobes is essential. You also need to buffer the address bus to prevent loading. It's also not so simple because the logic adds significant propagation which will be problematic with plenty of games using secondary address decoding which nearly use up their propagation quota already. I think to be safe for SNES for example all logic musn't exceed 30 ns. A PCB is probably necessary as will be low-capacitance layout...

To not wear out your power switch?

If a console has /RESET, use it.

 

So you are saying I could not just turn off the SNES completely, slide the thing over, make it set into the groove for the selected games and turn it back on? There is no PCB in the dummy cart. All it is, is pretty much a riser card. That is it.

I am hitting the power switch to turn off the console to switch games anyway.

I am not sure we are on the same page here. No PCB. These are just wires extending the 62 pin connector. That is it. Only one cart is "touching" the SNES at once. I am turning the system off. As far as the SNES is concerned, there is just a cart shoved inside of it.

Dummy cart insides (this example is missing the side pins but you get the point)

 

I remember my old dentist had something similar to this. He had a Genesis Model 2 with some sort of selector next to it in a stand like you'd find at the store that you press a button on the front of and it would cycle through I think 8 cartridges with a red light next to each to indicate your selection.

 

the slider method could work well, there may be an increase in resistance due to the extra wiring but dont see it stopping the carts from working, best way to try it is to built a prototype and try it out. logic etc only needed for digital switching.

manual switching means only 1 cart will be connected at 1 time, similar to the switch for the guy on here doing that huge multi system project.

it probably wont be too pretty but the theory is sound

 

I understand what you want to do clear as day (I don't understand the appeal, but that's another matter...), you can't just extend a computer bus passively like that and expect it to work, blame physics. You could get lucky and find that it works with handful of ROM-only boards, but there is no question it will not work with all games reliably. Just a short extension from an import adapter like that degrades signal integrity. To get the best compatibility you must buffer the busses so they may sufficiently drive the ribbon cable's capacitance. Buffers also isolate the cartridge from the rest of the internal busses (everything is wired in parallel) so that the ribbon cable capacitance doesn't load them down as well. Since one'd obviously use digital buffers, and you need a very high speed circuit because certain SNES games are almost too slow for the CPU already, you're probably going to need a PCB to control the signal integrity of the buffers... If you're going to use digital logic and produce a PCB, it doesn't make sense to fabricate your own custom switching mechanism because it's far cheaper and probably more reliable mechanically to digitally select the carts, the way all existing cartridge selectors will have done it. Even the Atari selector probably used buffers in the cartridge to drive the address bus back to the cartridge ports, then the switches either selected one of 6 buffers to drive the data bus, or they served as select inputs on a multiplexer which drove the data bus from the selector back to the CPU. (The Atari selector is almost certainly uni-directional/not compatible with a number of games.)

synrgy87: there will be a DEFINITE (tiny tiny tiny) increase in resistance, but resistance is not an issue, it's signal integrity from the parasitic capacitance and inductance that form impedance, which arrange themselves into filters and resonant circuits. If the selector didn't have a ribbon cable the parasitics would be negligible.

 

have ribbon cables not been used for cart slot relocation and extension in other mods? like multi system and unification mods? surely this is not much different?

edit:a relared thread elsewhere

http://www.atariage.com/forums/topic/131959-building-a-sega-genesis-jukebox-game-switcher/

didnt seem to go anywhere

 

Of course they have, generally by people who don't know what they're doing. Just because something appears to work doesn't mean it does, and even if it does work it doesn't mean it works well, or that it's always going to work, or that it leaves enough tolerance to work under additional constraints (such as slow ROMs, ROMs with difficult switching thresholds, finicky expansion cartridges, games with SRAM etc). "Modders" and paid engineers alike at all levels routinely get away with so so many bad design choices, not just debatably bad, but straight-up factually bad. This leaves their designs teetering on the edge of broken, but as they say ignorance is bliss... The reality is that if you want to do anything other than sloppy hacks, you have to actually learn the science--something that a relative few people have mastered over a lifetime.

Anyways, back to the topic. If a very time consuming, expensive and yes, sloppy hack to switch between only 3 carts, where compatibility is not probable much less guaranteed, is worth building.. I can't really argue with that.

 

personally i like the "screw it why not try it might work and if it does.. bonus" just my natural optimisim

anyway im not undertaking the project myself, its just interesting and would be a pretty cool thing to have / show off if it worked or even half worked


oh and i dont doubt any of your concerns, logic or knowledge.

 

Calpis - Thanks a bunch for the info (and dealing with my lack there of in a well thought out manner). If I ever get around to getting this concept off the ground I will let you know how it goes. Might actually make an interesting project for that group funded site Kickstarter, especially a more advanced PCB version. I would be willing to slap down $150 for one myself...