drstawl said:
>I have yet to hear a tone module produce nuance in color that a real instrument will. With only 127 possible volocity settings available for sequenced material, I doubt I will hear anything soon that will capture it!
Sound Cracker: I think that the Roland Sound Canvas does this occasionally. Expression is more than just the available resolution for a particular parameter. I'd agree that 16 bits worth of any parameter have a better chance of imitating the sampled instrument than 8, but in the thousands of SC-88 based tunes I've written I'd say that a few patches there DO have the capability of being elevated to the sound of real instruments by the sequence itself. Personally I think that 24 bit samples combined with better capture of the sample at different parameter levels would help out more than just increasing the # of different choices for (for example) velocity settings.
Funny that you bring this up after so much time doc. Me and Slack were discussing this vary thing ealier.
I stand behind my original statement. While it is a bit oversimplified, it is on track.
How many sampled sounds do you think they use for the average snare drum sound? 3? 4? Do you really think even a couple hundred could truely capture the nuance?
Then, we have the issue of "tracking" velocity via the trigger. I don't underestimate the power of muscle control. A well practiced drummer could literally play at hundreds of different levels on the same snare drum, each producing a nuance in sound that is totally distinctive to a listening ear. Until modules have THAT many samples available PER SOUND, and the hardware can "track" the input and accurately apply the proper sample to it, with latency under 5ms, tone modules are going to sound phoney. Imagine just the hardware requirements for something like this!!! A box like this ain't gonna be cheap.
I would suspect that each acoustic instrument sound would literally need thousands of samples available to be triggered to get in the realm of "realism". I could see something like an
acoustic guitar needing 10's of thousands of samples. Then there is going to be a need for intelligent algorythems that can create the nuances that hitting multiple notes at once will produce, thus, many more samples.
I don't see this happening for a very very long time.
Just about when I start thinking tone modules sound like the "real thing", I record the "real thing" and find out all over again how far they have to go before they can emulate it. It is one thing to fool the casual listener, quite another to fool the enthusiast or professional.
That Sound Cracker guy sure is a bull head about these things eh?
ed