ok this is pretty cool. i thought about this just a second ago.
i never had an sp1200 or any 12 bit sampler but for the longest time i tried to emulate the sound of 12 bit with the Hyperprism Sonic Decimator, a nice little direct x plugin. good results. however, it was a "last move" type of thing - ie after i tracked out.
let's say i made a bassline on the mpc with a sine wav/saw wav autochromatic program with qlink recoded low filter automation on for charactor as well as tune bends. we're talking about a 4 hour bassline here, something with a lot of nuance.
now let's say i recorded that into the daw and applied the sonic decimator to it, turning it into a 12 bit bassline. this action would apply AFTER the tune drop and AFTER the low filter - ie it would be a "static effect." it wouldn't get filtered out with the low filter or detuned with the tune automation.
however: here's what makes this jj bit degrader different.
it's all BEFORE. it's BEFORE the tune automation and BEFORE the low pass filter. ie, say you put a crispy edge to your sine/saw wave samples in the mpc so you can hear that fried, crystalline 12 bit sound. that sound mostly hits me in the high frequencies. now when you filter sweep it, you'll cut off that high frequency! it will come and go, which is quite different than the "static" overlaying effect if you had put it into your daw.
likewise with the tuning. as i posted before, downsampling a sound and THEN pitching it yields a different sounding downsample - it's "slower." when you downsample your sine/saw waves and then detune them with a qlink, you're going to have a "walking" downsample.
this applies more to tuned (16 levels or autochromatic) sample programs. if you have a multisampled bass pgm it won't be the same way since the sample conversion can be placed on EACH SAMPLE rather than just the origin sample.
all this promises a special charactor to the mpc1000's bit reducer. it's not REALLY putting the sound at a new bitrate/sample rate - it's still at 16 bit, just sounds different.
result: strangely automatable downsampling using lowpass filter and tuning. low filter = cutoff, tune = slow/speed the downsample.
i dont think this is something i've heard before. it's theoretically possible to set it up this way in reason or something but i've certainly never done it.
very excited to try this.
