OK.. here's a list of the known problems - please don't write to me about them!
1) Apparently, many problems have been encountered with QEMM... I really
don't know what's going wrong here at the moment (I haven't had any
experience with QEMM myself), but try using the following line in your
CONFIG.SYS file. If this still doesnt' work, I would recommend that you
steer clear of QEMM if possible for the meantime.
DEVICE=<path>\QEMM.SYS DMA=64, HANDLES=255
2) With the GUS, 16-bit samples > 256k will *NOT* be played correctly. This is
due to a 'quirk' of the GUS - something that I don't know how to easily fix.
This does *NOT* occur with 8-bit samples. (This is the same problem that
you'll find in FastTracker II also)
3) The "Active Channels" indication is *NOT* always 100% accurate (it's
even affected by stuff played in muted channels.....) Also, on a GUS or
AWE32, the program requires the hardware to provide the 'end of note
indication', whereas the mixing routines are calculated internally...
hence, if the hardware provides the indication a little later than expected,
another channel is allocated - what this means is that the value as played
on wavetable cards *MAY* differ from the value played on software-mixed
cards. This difference is minor (ie. 2 channels at most, unless you *TRY*
to setup a special situation)
4) There is no 'enforce Amiga limits' option in IT (Which was in ST3) so
pitches of notes exceeding the Amiga limit will not be suitably played.
5) If your system behaves unexpectedly in IT, it *MAY* be necessary to
specify the complete parameter list on the command line. This should NOT
be a problem in most cases, but it has been known to occur.
6) In Windows '95, if you shell to DOS when using SB on a high IRQ, then
playback may stop! I've got no idea as to why this happens, but it seems
to restore itself after loading another module (on a SB16)
7) If you Shell to DOS with insufficient memory available, WEIRD things may
occur...