In last weeks I was quite occupied by
studying piano voices and here's my conclusion so far:
First, if talking about
real grand pianos, not many people are capable to distinguish between them. If at all, then only if pianos are compared next to each other. I'm not talking about high rank professional piano players, of course. There are many high quality videos where the best grands are compared and it's quite a challenge trying to notice differences. Now imagine a blind test.
Some might say "I have CFX and Bösendorfer on my keyboard and the difference is obvious". Yes, it is... because someone (voice creator) made that difference obvious enough, so even untrained ear can hear it.
The next thing (as mentioned by some), is quality of audio system. We just can't expect that a small box with 5" woofer will sound the same as for example 8" speaker in much bigger box. It's not only about frequency response at low end. Explaining that would lead to off-topic, but I'm sure you get the point. In short, the same voice will sound better on better audio system -and will at the same time also reveal eventual flaws in voice.
The next important thing is, it depends on audio (wav) samples that are used in voice. I've downloaded quite many of them and my conclusion is, it's not that important if it's Yamaha CFX, Bösendorfer, Steinway, etc. More important is, how these samples were created: in what room (i.e. studio or concert hall), how were microphones positioned, how much post-processing was done on samples and finally, how much of effects was applied at voice creation.
What I'm saying is, we can't just say Bösendorfer is the best -it just means, compared to other voices, we like how this particular voice was created. That is, we need to find a voice (no matter what grand piano), which we think sounds most authentic.
I tried many samples and the best (in my opinion) I could find is
Garritan CFX. I don't say that's because it's CFX, but because how samples sound. Complete package contains over 120GB! of wav samples, taken with multiple microphones in various positions around and above piano, where each note is is recorded at various velocities. That means, we don't need to use artificial effects to make different sounding voice -we use samples that were recorded at certain microphone positions instead (i.e. where player is sitting).
I'm no voice creator expert (and YEM documentation is very poor in that), so I'm sure there are things that can be improved. Still, it sounds
much better than any built-in piano in my SX700 (just my opinion, of course).
Bogdan