I'm not sure how "modeled" it is, but the description makes it sound like it doesn't use samples, and has a "live" piano in it. Uses the same keybed as 700gx. Pretty impressive. I'd have to hear the modeled piano, but if roland gives it a passing score, I have a feeling it's going to pretty darn good.
If this thing is a good physmod, then pianoteq vs roland will be an interesting competition. It's good since the products appeal to different sections of the market (hardware and software users), but it's probably a matter of time before roland ports theirs to vsti. That is, if it can even be ported in the first place. I wouldn't be surprised if it required specific hw dsp only available from the physical unit.
Anyway, it's all speculations. Nothing can be said before I get to mess with it.
I wondered how long it would take; I first tried Pianoteq two years ago, and passed on it. However the concept of generating the sound - rather than copying the sound - always seemed logical to me than trying to record one
I'm listening to a Chopin piece as I type this, and it's being rendered with Pianoteq 2.3. I can hear all the notes clearly - no mud - on this particular piece, the polyphony sometimes shoots into the nineties.
Actually the RD-700GX is PH-II and the description of the V-Piano is PH-III, so it looks like it might be a next generation action. (The RD-700sx was PH-I, and there is a big difference, so at least with Roland, it doesn't seem to be just marketing . . .
It seems that you don't buy the sound you want (as in samples), you adjust the sound the way you want it.
It seems that the user controlled variables are much like those of a well known software solution. Both the strength and weakness of this approach is that some considerable knowledge is required in order to understand how the physical aspects of an acoustic piano affect the sound. Concepts such as hammer shaping, hammer hardness, non-linear responses to velocity, stretching, unison frequency variation, and others must be understood in order to achieve a "reasonable" sound.
One has great control if one understands these variables, but without patience, knowledge and experience, the sound may be worse than useless. I've seen numerous examples of the latter in attempts to achieve the sound of some well known piano. In knowledgeable hands, it's powerful; in the hands of others it can be not so wonderful.
Strange design to have almost all of the parameters controlled by one knob. Will mean a lot of button pushing to experiment or build an instrument.
Just for fun (if you haven't already seen it), go to a BMW dealer and look at their 700 series autos. They have one knob that takes you to multiple screens that control everything that isn't directly related to driving. I believe they call it I-drive. The concept hasn't been widely accepted by all owners.
I agree that it's strange. But then, Roland has never been a company that built products that were particularly well aimed at the general public. The manuals for their digital pianos are a wonderful case in point; at the opposite end of the spectrum is Yamaha when it comes to manuals. I'm on my second Roland DP, and I pretty well gave up on the manuals long ago.
Roland couldn't even provide me with the patch numbers for all the onboard instrument sounds - for each instrument, I had to record a few notes, load the recorded midi file into a sequencer, read the patch numbers in the sequencer, and write them down for future use.
Also, it's got 1 cpu (fantom has 1 cpu) dedicated to processing incoming signal from the keyboard. Which is very impressive, since it means the hammer heads aren't just modeled after key down, but the heads are actually moving in sync with the keys, as it's going down. That sort of key detection is only seen with v-piano, and also the new VAX77, which does vector detection also.
I just realized... it's a little ironic that yamaha is still stuck with sampling, even on their flag ship digital piano "Avantgrand"... and long time ago, they introduced physical modeling to the market with their VL1 synth.