There's a curious trick you can do with the N163 to get it to produce tones much higher than intended with MML. The effect seems to be very aliased or gets screwy with the filtering and makes for vastly different results across emulators. I'm not certain it hardware supports it either.
It's the simplest thing really, I'm sure someone has to have figured it out as well. Since the SA command was added, we have much more control over tuning and sweeps on the channels. Make it go as high as it'll go and it'll eventually settle on an odd tone before cycling back to the lowest hz it can output. Even with note envelopes, it still cycles through the tones. However, something very curious happens when we apply a pitch envelope with a decreasing value. Once it hits the lowest level, it cycles back to an extremely shrill range with harmonics bouncing all over it sometimes. Sometimes it goes silent.
When I messed around with it further, I noticed on a waveform of many buffers, it would remove all but the first buffer when cycling back through the tones. If you let it continue sweeping to the bottom and cycling back to the top again, it would add the next buffer samples back into the waveform, adding another buffer every full sweep until the entire waveform is back and then cutting back to only one buffer upon the next cycle through the tones.. This is why it would go silent sometimes, when the first buffer(s) were just flat waveforms with no difference. It would stay silent until the sweep reached a buffer that had different samples, and the shrill would kick right back in.
The thought was then to make an envelope to hit a sweet spot in the tone and then bounce it between a couple spots, add some volume and noise channel flourish and voila! A tambourine or sleigh bell effect. This was used roughly halfway into the Shining Force II cover that was entered in the compo. Course it sounds like crap in the flash player, and it's all over the place everywhere else...
Digging through a testing mml, here's another fun testing example to compile;
Code: Select all
//////////////////////////////////////////////////////////////
#EX-NAMCO106 8
#PITCH-CORRECTION
PQRSTUVW @t20,20
@EP1 = {|-10}
@N10 = {3 10 10 10 0 10 10 10 0 0 0 10 10 15 15 10 10 0 0 10 10 0 0 0 10 10 0 0 10 10 0 0 0}
/(the buffers past the first 4 samples shouldn't matter on this waveform. It was just from an early test.)
@v1 = {1 1 1 2 2 2 3 3 3 4 4 4 5}
@v2 = {5 5 5 5 5 5 5 5 5 5 5 5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0}
PQRSTU o1 EP1 SA1 @@10 l16
Q K1
R K2
S K3
T K4
U K5
PQRSTU v0 c8. @v1 EP0 w1^1^1^2 EPOF w1^1
P @v2w1
Q ^@v2w1
R ^8@v2w1
S ^8^@v2w1
T ^4@v2w1
U ^4@v2w1
//////////////////////////////////////////////////////////////
I'm sorry as always that I probably don't have the right sound terminology, I just wished to share an interesting trick that I'm not sure is legitimately useful or not given what I've learned about having many sound channels active since the compo. Any explanation would be very helpful and interesting.
