When I first saw the picture of this card, I wanted to say it had a bus mouse port, but YawetaG said the card indicated it was PS/2, so I assumed it was silkscreened somewhere I couldn't see. Now that I can see the close-ups, it's definitely for a bus mouse.

BTW, the Serial X and Y is not unusual at all. Serial ports are normally configured as COM1 thorugh COM4, and the serial ports just use variable names so that you can identify which jumper block to use to set the COM port.

Is a PCI card an option?

PCI cards with PS/2 keyboard and mouse ports exist, sometimes with USB ports on there as well.

feipoa wrote:

[...] I feel the best approach is to modify the motherboard's keyboard controller to allow for native ps/2 mice.

Are you referring to Rio444's way or yours?
Personally I stopped trying Rio444's pcb/tsr since the issues were too annoying. Basically it's pure luck: you can have a mobo working fine with the mod, or a mobo that does work with the mod or a mobo that works some times depending on the hardware config. 😵

Flare wrote:

Is a PCI card an option?

PCI cards with PS/2 keyboard and mouse ports exist, sometimes with USB ports on there as well.

PCI cards with PS/2 mouse/keyboard headers are all USB cards with a USB-to-PS/2 converter onboard, even if the card does not have USB ports. So to use this configuration in DOS, you'll need to find or code some DOS USB mouse driver. I've personally only tested this configuration in Windows, which does work. I think I got it working even in NT4 with some special drivers.

keropi wrote:

feipoa wrote:

[...] I feel the best approach is to modify the motherboard's keyboard controller to allow for native ps/2 mice.

Are you referring to Rio444's way or yours?
Personally I stopped trying Rio444's pcb/tsr since the issues were too annoying. Basically it's pure luck: you can have a mobo working fine with the mod, or a mobo that does work with the mod or a mobo that works some times depending on the hardware config. 😵

Rio444's approach is practically the same as mine. I think the issue with his is that the OshPark PCB's are too thick for the pins to make sufficient contact. I tried my keyboard controller adapter on 6 motherboards and they all worked fine, so I think the odds are pretty good of others getting this to work.

feipoa wrote:

Flare wrote:

Is a PCI card an option?

PCI cards with PS/2 keyboard and mouse ports exist, sometimes with USB ports on there as well.

PCI cards with PS/2 mouse/keyboard headers are all USB cards with a USB-to-PS/2 converter onboard, even if the card does not have USB ports. So to use this configuration in DOS, you'll need to find or code some DOS USB mouse driver. I've personally only tested this configuration in Windows, which does work. I think I got it working even in NT4 with some special drivers.

Makes me think: Some motherboards with USB have an option called "USB legacy support" build into the BIOS. Enabling it allows you to use USB keyboards (and maybe USB mice, I never tried) with software that expects a PS/2 or DIN connection.

feipoa wrote:

I think the issue with his is that the OshPark PCB's are too thick for the pins to make sufficient contact. I tried my keyboard controller adapter on 6 motherboards and they all worked fine, so I think the odds are pretty good of others getting this to work.

I don't think it's a connection issue - it does not explain why on my FX3000 motherboard it works perfect with an 8bit ATI VGA and hangs when using any CL or WDC ISA vgas (that are not jumpered to use any irqs) or the fact that my Unichip386DX motherboard does not work at all with a kbc that supports ps/2 whether it's modded or not.
IMHO there is more to this to make it a truly universal solution. Maybe the BIOSes need modding, maybe there is another way for the TSR to work with them, maybe something additional needs to be done to the mobos, too many maybes 🤣

I suppose you could wire it up the way I did it if you think that is any better. From my perspective, I didn't have any serious problem with this approach.

Maybe it's not a universal solution, but it's still a hell of a lot better than what we had before.

I haven't taken a close look at the most recent version of the PCB, but I'll assume it's somewhat similar to the crappy prototype that I wired up a number of years back. I tested mine on two two motherboards. It worked fine on my M-tech R418, however with my DTK PKM-0031Y, it caused some kind of conflict with the turbo functions (even though I couldn't properly test the module on this board, since it was equipped with AMI BIOS and the TSR had not yet been written). Assuming I had a copy of AMIBCP, I could have possibly disabled or remapped the turbo functions thus avoiding the conflict. I would be surprised if this issue didn't pop up in other boards.