Posted by Nathan Strum on 6, 2001 at 6:43 PM:
In Reply to: My 2 cents worth posted by Fultra on 4, 2001 at 11:08 AM:
: The problem that you guys are describing seems to be a combo of several things...all of which have some contributing factor. However, while the spacing on the encoder disc can have some affect on this problem - most of the problem is due to the mouse port (PS2 or Serial) and the driver.
: (From memory) What happens when you spin the mouse is that the information (3 bytes worth) is stored in the buffer and then sent at 1200 baud to the serial/ps2 port. If you spin quicker than the interface or mouse guts can store and send the data, you end up in the stalled situation. Not every piece of 'movement' info gets sent and the result is a small period of time where the mouse stalls and/or jumps left and right.
Wouldn't an encoder wheel with more spokes of the same diameter (say 64 instead of 32 for an easy number) be more prone to this problem, due to twice as much info being imparted to the circuitry per revolution? This is more of what I was referring to (although somewhat unclearly in retrospect) than just the physical width of the spokes. I meant to imply that the thinner the spokes, the more of them there are, and the worse the problem gets as you spin it faster.
At least that's what I think I meant. ;)
At any rate, the different mice I've gutted and tested seem to back this up. Mice with more spokes on the encoder wheels tend to stall much more easily. Oddly enough, they tend to stall even when using other encoder wheels. I suspect this has something to do with the sensors needing to be so close together in order to properly read the skinny spokes, that they don't read thicker ones as well (or are more fussy about the spokes being perfectly perdendicular to the sensor path). Or perhaps the circuitry just happens to be designed to read "x" amount of data, and anything over that just blows their tiny little minds.