I can achieve 160 notes of P\'phny but there is a but

When I load the giga piano on it\'s own - fine, 160 no prob\'s


When I load my performance file which contains almost the entire AO library as well as the giga piano, the same test (arms across the keyboard with the sustain pedal down) results in a peak of 136 voices if I\'m lucky.

I have: 1ghz AMD cpu, 768 megs of ram, Ultra160 scsi drive and a Delta66 sound card.

Now I know it can\'t be a data rate problem because of the test I perform with and without the additional gigs. So, I’m guessing it’s memory related. By the way, the .gsp file loads 388 instruments and uses (according to GS) 63% of the memory.

I have tried many tweaks. Many different cache settings (by hand and using cacheman), graphics accel’ is all the way down, Double Buffer = 0 (whatever that does?) virtual memory = disabled. If you can think of anything else I’ve probably done it – but let me know anyway… …worth a try!

One strange thing that accompanies this problem is that when I load the .gsp file I can’t get a sound out of GS. That is until I change a hardware setting within GS (i.e. dither type, Midi in port, outputs enabled), then change it back again and then press apply, affectively changing nothing. This is not midi related because GS is displaying midi activity, however, there is NO voices or voice peak activity.

I feel sure that this is linked to my polyphony problem, although I’m open to any suggestions about that.

One reservation I do make with my Giga dedicated system is that I may have bought a ‘bad’ motherboard. It’s made by jetway and uses the VIA VT8363 KT-133 chipset.

Well all, I’m stumped on this. What do you make of it?

Any help would be very much appreciated