I've been watching this thread with some interest, mainly due to the polyphony issue, which to be honest 'doesn't exist'.

Korg sounds are made of 2 samples, while Y&R keyboards can be made up of 4. For example if your keyboard is 64 note poly, on a Korg that would drop to 32 with a 2 sample voice, while on a Y/R that would drop to 16. That's typically why you hear drop-outs on Yamaha's and the Motif in particular.

The above is also the reason why Korg is able to do the 8 track combi's and Yamaha does the 4 track performances.

In my experience, Korg boards average around 80-90 poly compared against Y/R.

I've owned an i30, Trinity and Triton and haven't experienced note drop-off's. Like AJ, I have done on the Motif. What you must also consider is that the Pa1X has sampling features, and it's possible to record the drum part (with mp3 option) and replace that in a style with the sample, thereby saving you perhaps 10 notes.

In response to other comments. The Pa1X, like the i30 before it 'might' turn out to be a much understated keyboard. The i30 wasn't to everyone's liking mainly because it didn't have a lot of 'flash' elements like Rolands V series, especially.

What I would say to all here is 'seriously give it a go', but please don't come back after a 1 hour demo at your music store and pronounce it 'dead on arrival'. The i30 was a composers dream (then), and while it might not be 100% suitable for your needs, maybe live, it may make up for this in other areas.

A poster commented about the lack of hype surrounding the Pa1X as compared to the Tyros. Please remember that Yamaha is a BIG company compared with Korg, and in the arranger market much more embedded, especially schools/teachers and the like. Dealers also make a bigger profit on Yamaha's, not surprising.

One other thing b4 I go and it's the Mega-voice thing. Quite honestly, i don't see the excitment. Mega's are nothing more than multi-layered voices, albeit programmed well. Playing them is a task in itself, and best left to programmers. What is surprising though is that we've seen this technology well-before the Mega-Voice term was coined.

The way the programming market seems to be going is using and manipulating samples (which are far more realistic than any mega voice can product). I think Korg sees this and that's one of the main reasons there's sampling features on the Pa1X.