Ultima VII ending has two wrong rhythm events.
Because rhythm 28 is not implemented by MT-32/CM-32L, they become silent.
But Munt play that wrong event, so it sounds "bang".
I believe the out-of-range reverb switch value makes no difference here. Note, all versions of control ROM clamp all out-of-range values generally, so this could hardly be a problem.
I tried to reproduce the problem with MT-32 ROMs 1.07 and 2.04 as well as with CM-32L ROM 1.02 without any luck. Here's the SysEx I attempted:
F0 41 10 16 12 03 01 20 │ 40 50 07 07 3E F7
that is a normal drum R0 from preset but with the bogus value 07 in the reverb switch field. I suspect this requires a deeper look and investigation of the custom timbre too.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Please note, I am fairly unable to proceed without further details on how to reproduce this bug, so e.g. a SysEx dump of the suspicious timbre would be very much appreciated.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
"Because rhythm 28 is not implemented by MT-32/CM-32L, they become silent.
But Munt play that wrong event, so it sounds "bang"." -> wrong.
I've tracked down exclusive message before ending.
Rhythm 28 defined by this exclusive, with wrong parameter.
Address 03 01 20 -> 03 50 07 07
Last "07" is wrong. It must be "00" or "01" (Reverb Switch).
MT-32 sounds this patch, but LAPC-I never.
I believe the out-of-range reverb switch value makes no difference here. Note, all versions of control ROM clamp all out-of-range values generally, so this could hardly be a problem.
I tried to reproduce the problem with MT-32 ROMs 1.07 and 2.04 as well as with CM-32L ROM 1.02 without any luck. Here's the SysEx I attempted:
that is a normal drum R0 from preset but with the bogus value 07 in the reverb switch field. I suspect this requires a deeper look and investigation of the custom timbre too.
Please note, I am fairly unable to proceed without further details on how to reproduce this bug, so e.g. a SysEx dump of the suspicious timbre would be very much appreciated.