Mouse IBM PS/2 Mouse Program Diskette v1.00 IBM Mouse Program Diskette v1.2 [96F9262]

How do They Run? (PC Mag article about mouse interfaces)

Optical Wireless Mouse

Mouse Connector Pinouts
Mouse Connector Signals
Using Serial Mice on a PS/2?
IBM Mouse Types
PS/2 to Serial to PS/2 (Dumb KVM switches)
PS/2 Mouse Maker

Ed. Sorry, but I cannot totally verify the below. It's my best SWAG.

Location of Mouse (pointing device) connector.

The mouse port is always the PS/2 port FARTHEST from the power supply.

The PS/2 mouse port is IRQ 12 and I/O Port Addresses 60h and 64h.

Bus Mouse Problem!

A bus mouse uses a mini-DIN connector just like the PS/2 mouse, although they are totally incompatible. You can damage your motherboard by plugging a bus mouse into it.
Note: Microsoft calls it's bus mouse an Inport mouse. Be warned!

PS/2 Appears to Hang When Exiting Windows 3.1x

Some IBM PS/2 models may appear to stop responding (hang) when you exit Windows version 3.1. The computer eventually restarts; however, it remains frozen for up to a minute while the mouse port is reinitialized.

This problem is most severe on PS/2 models 56 and 57. It also occurs on Models 50z, 55sx, 70, 80, 90, and 95.

The problem can be corrected by adding the following line to the [386Enh] section of SYSTEM.INI:


This prevents Windows from attempting to reinitialize the PS/2-style mouse port prior to exiting to MS-DOS. This may cause problems when you run some mouse-aware, MS-DOS-based applications after exiting Windows.

Mouse Connector Pinout

The keyboard connector is a 6-pin miniature DIN connectors. The signals and voltages are the same for both connectors.

Pin Signal
1 Data
2 Reserved
3 Ground
4 + 5 V dc
5 Clock
6 Reserved


The keyboard and auxiliary device signals are driven by open-collector drivers pulled to 5Vdc through a pull-up resistor.

Sink current Max 20mA
Hi-level output V Min 5.0 Vdc minus pull-up
Low-level Output v Max 0.5 Vdc
High-level input v Min 2.0 Vdc
Low-level input v Max 0.8 Vdc

Serial Mouse Conversion?

The "Serial Mouse" uses a RS-232 style interface with -5 to -12 VDC as logical "1" and +5 to +12 VDC as logical "0". The PS/2 mouse interface is a TTL-style interface, which uses 0 - +2 VDC as logical "0" and +3 - +5 VDC as logical "1".

In addition - and to make things worser - the RS-232 is an asynchronous interface, the PS/2 interface is a synchronous, where the data is sent along with a clock signal. It uses a simplified 4-wires serial interface with +5VDC (for the transceiver), GND, keyboard / mouse clock and keyboard / mouse data. The data and clock line can be used from the keyboard / mouse controller and the attached device as well following a particular handshake, which defines which is the active "sender" and which is the "receiver". So: it is not *that* easy conversing serial mouse to PS/2 and vice versa.

The "dual mode" mice have an automatic logic detection and sort of adaptive interface electronic, which detects whether the mouse is attached to a serial port or a PS/2 port and set the output drivers accordingly.

As you can see from the above: it is not enough only *physically* changing the plugs.

PS/2 Mouse Versions

From Fred Mau:
   Best as I can tell, IBM had five distinct species of Rattus Armonkus in the PS/2 world: (Not to mention all the clones).

  • The original ugly wedge-shaped PS/2 mouse
  • Early oval-shaped mouse. White body and white buttons.
  • Later oval-shaped mouse. White body and brownish buttons.
  • A smaller black mouse used with the CL57, N51, TP 700 and 720 laptops. Lower voltage than a regular mouse, not interchangeable.
  • The current IBM mouse, OEM'ed by Logitech. A more ergonomic oval.

Best as I can tell, all the PS/2 mice (except the minnie mouse for early thinkpads) are interchangeable, with one notable exception: If you have an 8516 Touchscreen monitor, the touchscreen cable plugs into the mouse port and the mouse plugs into the monitor, but it will ONLY work with the original ugly mouse or the early oval mouse with white buttons. It won't work with the later oval mouse with brown buttons or anything newer. I have no idea why, but apparently something changed in the mice. It's something to keep in mind if you ever happen across an 8516.

From Jim Shorney:
   Side note: I seem to have an oval variation you didn't mention: white top, white buttons, brownish bottom, 33G5430/33G5410/FRU33G5420.
   BTW, have you seen the IBM memo on mouse ball replacement? Hilarious. (Ed. I think he means THIS.)

From Carroll Bloyd:
   And there's the track-ball/mouse combination (P/N 1397040) sold for use with the L40 SX laptop. Big ugly thing--the mouse side of the device looks like a larger version of the original PS/2 mouse.
   And, you might want to add the three-button mouse OEM'ed from Logitech and supplied with the 9595-PQx models. They seem to be rare--I've only seen a few--I wonder if they were used on the RS/6000 also?

From David Doerr:
   There was, indeed, a three-button mouse from Logitech used on the RS/6000. I have several, but they don't seem to work on my 9590...

Resolution Increase

From Aron Eisenpress:
   I'd like to point out one substantive change in the IBM PS/2 mice, which happened around the time of the two-tone-topped ones: the resolution increased from 200dpi to 400dpi (and noticeable if you swap mice around as the speed of the cursor will change!)...

PS/2 mouse to serial cable and back to PS/2 port (or Dumb KVM switches) (from Tim Clarke)

Hi Bill,
   I think your problem is in the PS/2->Serial and Serial->PS/2 adapters. You *must* ensure that they *are* as described and *not* 2 x PS/2->Serial or 2 x Serial->PS/2 used back-to-back. Otherwise, the pinouts will not get 'passed' correctly.
   The thing to be aware of is that, with a manual switch, the keyboard and mouse 'lose' power during a switch and 'reinitialise' when the switch closure to the 'to' computer is completed (i.e. 5V power returns). Also, you can get spurious bytes sent during the transitions (plaintive beeps from the BIOS).
   As I said, DOS (due to BIOS) and the OS/2 (due to it's drivers) recover most of the time (OS/2 seems to recover the mouse best if you use Desktop "Lockup" before switching away). Clones (AT-class BIOS) and Windows O/Ses are generally horrible because they haven't implemented the equivalent functionality and cause you to 'lose' the mouse and, sometimes keyboard function (except for manic beeping for key presses) after a 'switch-away, switch-back' sequence. It's not so much the BIOS in Clones, that seems to be compatible. It's the usual crap programming by the Microsoft "gurus".
   It may be down to the O/S drivers. E.g. All PS/2 BIOS/ABIOS seem to handle an 'unexpected keyboard reconnection' (i.e. 0xFF bytes and BAT-completion message) and reinitialise the 'Scan Code Set' and 'Shift State' indicators. Mouse stuff seems to be much more sensitive and OS/2 is the only server environment that I *know* can (usually) handle this. Netware 3.1x and 4.1x don't. Windows and WfWg don't. I doubt whether W9x does. NT *might* but I don't know. DOS is happy so long as you're at the command prompt with no mouse-driven s/w actively using the mouse.

Current Mouse Maker

Unicomp lists the IBM Two Button mouse. It has the brown buttons and the white body. I have a few of this specific model. Nice button switches, tracks well. At $7 it's a damn good bargain. Mine has a long cord (nice for use with 80/85/95 systems!)

96F9275 IBM Two Button mouse Regular $15 Sale $7

Content created and/or collected by:
Louis F. Ohland, Peter H. Wendt, David L. Beem, William R. Walsh, Tatsuo Sunagawa, Tomáš Slavotínek, Jim Shorney, Tim N. Clarke, Kevin Bowling, and many others.

Ardent Tool of Capitalism is maintained by Tomáš Slavotínek.
Last update: 08 May 2024 - Changelog | About | Legal & Contact