My take on 'elastic' audio vs. MIDI styles has been said before, but essentially boils down to there are usually FAR fewer chord choices from the 'elastic audio' patterns.

An arranger's MIDI engine can derive ANY chord it can recognize, including 'slash' chords, suspensions and inversions, and make the MIDI engine play those chords on ALL the sounds. But 'elastic' audio can ONLY change the tempo or pitch of a recording. It can't change the chord, the suspension, or the inversion, only switch to another recording.

So you have to have, for instance, a recording of EVERY possible guitar chord type, in ALL it's inversions, suspensions AND 'slash' variants for EVERY single variation of every single style that uses a guitar (just about ALL of them). First of all, that's an insane amount of data. But forget that, let's just assume you HAVE a terabyte sized drive or two attached to your arranger...

The REAL problem will be... who's going to MAKE these 'elastic' styles, and sit there and laboriously record the guitar patterns (and THAT'S just for starters!) in dozens upon dozens of chord variants, for EACH of the style's variations, for EACH of the styles? And then make them affordable to buy...

This is one area where the MS is going to be unable to simply COPY some other manufacturer's already proven body of work (and the copyright laws are FAR more rigid for recorded music, which is what these loops are, than for instrument samples) and they are actually going to have to make these themselves, or hand the work to third party developers, that will have a hard time realizing enough profit from what the market will bear for new styles to justify the ENORMOUS time to develop them.

OR.... you will all just have to get used to the fact that you have a vastly smaller pool of chord choices to use in your music.

Not exactly the great leap forward in arranger technology we are all looking for, IMO.

And before Dom gets all worked up about me, again... this criticism applies to ALL arrangers that are looking to 'elastic audio' as the next, greatest thing. Audya and all...

I just feel that, compared to developing more realistic sound sets, and better arranger engines to derive more realistic performances from those sounds, looking to recordings (other than drums and percussion, that don't NEED multiple chord choices) to give us more realism is a choice that only those that play the most basic chords in the first place would happily make.

It's an easy choice for a manufacturer, FAR easier than developing something like a cross between Yamaha's Mega voices and Korg's guitar Mode, for instance. Simply use already existing technology to time-stretch and pitch transpose audio (these have been around for years), and just HOPE that no-one notices you can't play a major ninth with a sharp five any longer...

JMO, yada yada yada...