I'm having issues with my 161 multicart on my MV1B, thought this'd be a good place to ask. I'll start a new thread if this is too distracting.
Anyway as above, I'm having issues. Its an MV1B and a unibios 3.1, neo bios masta mod.
With the unibios, I get a code corruption message. With standard US MVS bios I get z80 errors.
Did you need to tweak any of the multicarts to get them to boot on a 1 slot?
The 120 in 1 seems to have 2 pads for the caps already.
What sort of SMD 47pF caps should I use for the 120 in 1 cart?
There are a few different types on eBay - are these 0805 like the 100nF bypass caps?
Cheers!
My understanding is that 5v pull-ups on 3v output logic levels should in theory pull a higher logic level. The theory is there's ~5v until the line is pulled low by the output, thus giving a difference in logic level? Am I missing something there? But yes its a cheap and dirty way to deal with the issue and dedicated level shifters would be best. I could be wrong but I am sure many years back when I did something similar I looked at the logic levels and they were affected by using 5v pull ups on 3v outputs. Any information you can provide here is appreciated btw - I am basically getting back into the 2 years I spent in electronics after 20+ years, so I do make mistakes in some of the stuff I ramble on about.
EDIT: I didn't add pulls and pull downs at the same time obviously! I added pull downs later to see if that helped with the PCM issue but it made no difference either. It worked the same with or without pull ups or pull downs - with the exact same timing glitches.
It might be interesting do something like you've described with the bus tranceivers there but I honestly don't think it's worth it - I imagine it being a tonne of work for little reward really. Sure the MVS and or cart might live longer but I think it's not causing that much of a problem. Since I've done the work to 5 or 6 of the multicarts I have there are no stability or sound issues what so ever - other than dodgy ROM images they used to program them with.
EDIT2: I've just refreshed myself on the theory behind this and you are correct I think, the pull up wouldnt solve the issue at all if it had been a logic level issue. Just reading some material that suggests that when nothing is connected it would be 5v logic level, but when being driven it would drop back 3v. It has been a very long time since I did theory and it still doesnt make sense to me how a 3v output could pull it low, but I guess it's all just about PD.
In this case and the case of many, many integrated circuits, the supply voltage (3.3v or 5V) is used to provide the high logic level. Some ICs, made to work on an open collector bus (or configured for such) will go high-z (practically not driving it at all) for a logic high, and sink current to ground for a logic zero. This way, multiple chips can pull the line to 0 and let it "float" back up when nothing else is dragging it down. It is driving it to 3.3v because that's what the chip has to source with. Sometimes a 3.3v chip can be run outside of its tolerances at 5V and provide a working 5V bus at the expense of the chip's lifespan. You can see that in action in many new SNES clone consoles, where 3.3V ram is run at 5V.
the real solution is to put bus transceivers in there, but doing all that trace scratching and patching would be a real pain in the ass. It would be best done at the manufacture stage for these cartridges, but the people designing them can barely put enough care and money into a well built cart that plays all the games properly. They certainly don't care about the power consumption or long term damaging effects of the mismatched voltages.
Theoretically, an adapter could be made to sit between the MVS and the cartridge, doing bus level translation. The various chip select signals should provide enough information regarding the bus direction. At the very least it would be good to do on the video bus, as that is connected to irreplaceable SNK custom chips.
Has anyone posted a fix for the newer revision 161-n-1 boards? Mine is 201213 but I haven't seen any guides or pictures of a fixed board.