Kenny, if MB ever equalled 1, it was the first thing I changed two years
ago. This is a performance issue, not an MB issue; and that's why I sent
you benchmark numbers. Here they are again. The error msg is not so
"benign"! Each time Xy puts it up the whole system takes a huge hit. First
with, then without, msg \688 (the timed routine simply issues 1000
keystrokes):
XyWin XP 14.65 secs with default UI=1,1,0,0,0,0,0,0,0,2,0,0
XyWin WG 14.81 secs
XyWin WZ 22.19 secs
XyWin XP 3.85 secs with default UI=0,0,0,0,0,1,0,0,0,0,0,1
XyWin WG 3.96 secs
XyWin WZ 5.91 secs
Surely your programmer understands that the ONLY way for a user to get into
this position is very deliberately (I don't think one can do it through the
menus, not UI=1,1,0,0,0,0,0,0,0,2,0,0); so obviously users can find their
own way back out of the black hole if that's what they want. In any case,
it is inaccurate to state that \688 only goes up if the PRline isn't
otherwise in use! It goes up on *every* *single* *keystroke*, and thereby
wipes out any existing PRompt. The whole thing is simply a mistake, is all
it is. Let's dump it and move on to testing your OS/2 version. How's it
coming? You sly sly fox you, our faith turns out to be warranted at last!
------------
Robert J. Holmgren
holmgrn@ibm.net
------------