VOGONS


First post, by GEOCE

User metadata
Rank Newbie
Rank
Newbie

I have a problem with the mouse in the photo.
I have a 486 with Windows 95 and a Pentium with Windows 98. On both it works perfectly. Then I have a 386SX to 25 mhz with MS-DOS and Windows 3.11. I’ve tried all the drivers available on the Internet and they all behave the same. I have to click for the mouse to move the pointer. It’s as if it needs one of the three buttons pressed in order to activate.
I have another mouse that works perfectly in that port.
Can an expert on the subject help me with their wisdom?

Reply 1 of 8, by GEOCE

User metadata
Rank Newbie
Rank
Newbie

Please, someone help me. Clicking any button activates the trackball and the cursor works, but as soon as I stop the mouse, I have to click again for it to work. I don't understand why this happens.

Reply 2 of 8, by Deunan

User metadata
Rank l33t
Rank
l33t

DOS (and Win 3.x to some extent) requires a correct mouse driver. CuteMouse is usually good enough but you might need to set some parameters for it. Back in the day I had a "square" Genius mouse with 3 buttons, GM-6 or something very close to it. It had a 5.25" floppy with a driver and some CGA programs, I remember that I had to use that driver because smaller ones would usually default to Microsoft mouse mode and would not work at all. And the Genius driver did support 3 buttons.

So try CTMOUSE /S to force Mouse Systems mode, if that doesn't work you might need to look for an older Genius driver pack, before PS2 and wheels were a thing.

Reply 3 of 8, by GEOCE

User metadata
Rank Newbie
Rank
Newbie
Deunan wrote on 2025-09-26, 08:58:

DOS (and Win 3.x to some extent) requires a correct mouse driver. CuteMouse is usually good enough but you might need to set some parameters for it. Back in the day I had a "square" Genius mouse with 3 buttons, GM-6 or something very close to it. It had a 5.25" floppy with a driver and some CGA programs, I remember that I had to use that driver because smaller ones would usually default to Microsoft mouse mode and would not work at all. And the Genius driver did support 3 buttons.

So try CTMOUSE /S to force Mouse Systems mode, if that doesn't work you might need to look for an older Genius driver pack, before PS2 and wheels were a thing.

I’ve tried 10 different drivers… in MS-DOS with Windows 3.1 and 3.11.
The mouse works, but only when I click. I move the mouse and the pointer doesn’t move, but if I click any of the three buttons while moving, the pointer moves until I stop the mouse. Then, to make it move again, I have to press one of the buttons again.
However, this doesn’t happen on a 486 with Windows 95, nor on a Pentium with Windows 98.
I have another two-button mouse that works perfectly on the 386 with MS-DOS and Windows 3.11.
So I don’t understand what could be causing this.

Reply 4 of 8, by Jo22

User metadata
Rank l33t++
Rank
l33t++

I've had a Genius mouse, too in the 1990s. On my 286. It was an an Genius Easy Mouse, I think.
I've used gmouse.com on DOS; the standard Microsoft driver on Windows 3.10.

Vogons Drivers has v9 of the Genius driver, for example.
https://vogonsdrivers.com/getfile.php?fileid=1784

"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel

//My video channel//

Reply 5 of 8, by CharlieFoxtrot

User metadata
Rank Oldbie
Rank
Oldbie

Does the mouse work in DOS mode with your windows 9x boxes?

Reply 6 of 8, by GEOCE

User metadata
Rank Newbie
Rank
Newbie
CharlieFoxtrot wrote on 2025-10-06, 17:42:

Does the mouse work in DOS mode with your windows 9x boxes?

That's a great idea, I'll try it.
But I think it's the hard drive's Local Bus controller that isn't compatible.

Reply 7 of 8, by Jo22

User metadata
Rank l33t++
Rank
l33t++
GEOCE wrote on 2025-10-06, 14:49:

So I don’t understand what could be causing this.

Could be anything. Maybe it's related to the mouse protocol, even.
If a button is pressed, the lenght of the data stream maybe is different.
Which in turn could matter because of the serial port buffer.

Ideally, since the 1990s, an 16550 UART FiFo should be used (16550AFN etc is pin compatible, so multi-i/o cards can be retrofitted).
The i8250 and 16450 have a buffer issue and cause a high interrupt rate, making them unsuitable for modems and weak computers.
Windows for Workgroups and up support the 16550 FiFo, there's a FiFo= setting in system.ini.
For DOS, various utilities exist to manually enable it.

"Time, it seems, doesn't flow. For some it's fast, for some it's slow.
In what to one race is no time at all, another race can rise and fall..." - The Minstrel

//My video channel//

Reply 8 of 8, by Deunan

User metadata
Rank l33t
Rank
l33t

Jo22 suggestions made me remember some odd mouse behaviour I had in the past. The problem wasn't the driver but rather the mouse. Well mouse iteself was OK, the flaky part was the RS232 output driver on the I/O card. It was just barely able to provide enough current to run a newer, more energy efficient mouse but would cause an older one to behave erratically. And another curious thing was toggling some control bits on the COM2 port (DTR I belive) actually helped drive the COM1 to output higher voltages. That's why I don't like I/O cards with custom, combined RS232 drivers - these chips are prone to problems and hard to replace.

Anyway, try another I/O card, preferably one with 1488/1489 class RS232 drivers in DIP (or SMD) packages. Try both serial ports because one can be bad. Make sure your PSU -12V output is at least -11V. If nothing else helps it might be some cable damage or a bad electrolytic caps in the mouse itself - it might work with systems that provide better voltages but not others.