"Faking" a ps/2 mouse.

**Scouse Monkey** · 04-10-2005, 07:00 AM

i should imagine girder does as you want using an anologue joystick as all the x-y data will be there from the joy control. if not i am sure it wont be too hard to make.

EDIT: let me know how it goes as i want this feature too eventually

**hcker2000** · 04-10-2005, 07:06 AM

Yea I think it was frodo who whiped some thing up in like 5 min. It didn't do exactly like I wanted but that was because he dosn't have an analog joystick to try it out with. Wish he would let me know if he ever figured it out or not.

**Chris31** · 04-10-2005, 09:39 AM

Wow, look like a private convo here.

Mind if I join?

**Scouse Monkey** · 04-10-2005, 09:40 AM

Originally Posted by Chris31

Wow, look like a private convo here.

Mind if I join?

yup, if you have something useful to say!

(looks like you just did)

ARE YOU FOLLOWING ME NOW???

**Chris31** · 04-10-2005, 09:47 AM

Originally Posted by hcker2000

Ahh see thats how the Joymouse is and it is anoying. I want it so when I move the stick all the way to the left it moves the mouse fast left. When I move the stick a little to the left it moves the mouse slowly left.

Stick pointer as found in laptops?

**Chris31** · 04-10-2005, 09:48 AM

Originally Posted by Scouse Monkey

yup, if you have something useful to say!

(looks like you just did)

ARE YOU FOLLOWING ME NOW???

Oh no, not at all.

**mushin** · 04-10-2005, 12:51 PM

Originally Posted by Scouse Monkey

lI am sure there is probably a way to just connect a few pins with a resistor load or summit.

I don't think that's going to work... You can see the mouse initialization routine about 2/3 way down on this page , but looks like at the very least you'll need to transmit 0xAA and 0x00 (for self-test passed and mouse ID) on the data line (which in turn requires the clock line). A PIC could easily do it.

**Scouse Monkey** · 04-10-2005, 12:59 PM

Originally Posted by mushin

I don't think that's going to work... You can see the mouse initialization routine about 2/3 way down on this page , but looks like at the very least you'll need to transmit 0xAA and 0x00 (for self-test passed and mouse ID) on the data line (which in turn requires the clock line). A PIC could easily do it.

hacking apart an old mouse and aking the crap you don't need out sounds easier. How do KVMs stop the PC thinking the device has been unplugged? they must do it somehow.

**egeekial** · 04-10-2005, 01:31 PM

If you don't want to pay for Girder, you can always use WinLIRC. I've never used it, but I am using Lirc for Linux, and it works great.

**mushin** · 04-10-2005, 01:50 PM

Originally Posted by Scouse Monkey

hacking apart an old mouse and aking the crap you don't need out sounds easier. How do KVMs stop the PC thinking the device has been unplugged? they must do it somehow.

I agree, just using an old mouse is easier. You can probably even hack away a good deal of the PCB if space is really important.

Do KVMs keep the PC thinking it has a KB/mouse? At least for the the manual A/B switches, I thought they just physically change the connection. I suppose it could connect the 'unattached' PCs to an internal chip that acts like a kb/mouse.