>>I still find it difficult to believe that the mouse driver could have such a major and fundamental effect on the system, but I can't argue with success.
>
>If you think about where the mouse driver runs (based on that it will keep working when EVERYTHING else is locked up) it would make sense that it can trash everything else.

IME it is by no means always the case that the mouse continues to function; in fact, again IME I have lost the mouse but can still close apps and reboot with the appropriate keyboard combinations e.g. Alt-F4.

Under NT/W2K *most* hardware is isolated by the Hardware Abstraction Layer - see http://support.microsoft.com/support/kb/articles/Q99/5/88.ASP

It used to be that *all* hardware was abstracted, but starting with NT 4.0 MS relaxed this slightly and allows video drivers to bypass the HAL for increased performance. This is why NT4 and W2K are vulnerable to beta or buggy video drivers on new display adapters.

Mice and similar input devices are very low speed and have no need for this sort of speed enhancement. I would be *very* surprised if mouse drivers bypass the HAL; if so, it's to MS's and Logitech's discredit.