Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards Presto: From a Kawai MP10 to a Yamaha AvantGrand N2...

sullivang, I will check the MIDI codes and report back. The instrument doesn't need regulation. Its touch is even and consistent from bottom to top. In general, if it needed regulation, there'd be unevenness in the touch and SOME KEY would play off the jack, just not all keys. It is, of course, possible that the whole thing is out of wack, but I doubt that is the case. (But who knows, maybe another AG owner can get theirs to play off the jack.)

sullivang, Using Pianoteq's MIDI info, the AG sends an "Aftertouch Value 127" for the note being played when you try to play off the jack, but it does not send any actual note velocity. When you release the key, the AG sends an "Aftertouch Value 0." In a software piano, you could assign that info a low velocity and achieve the "playing off the jack" effect. However, it's a little jumpy, if you approach the letoff point with any speed, the AG sends the Aftertouch Value 127 without needing the key to be pressed all the way to the bottom. But if you approach the letoff point carefully (and don't trigger the Aftertouch info prematurely), you'll get the playing off the jack effect.

Thanks!

So, if you press a key very carefully, and stop just before the "notch", you can hold it there for as long as you like, right? If you then press down forcefully, a very soft note will sound. Do you get an Aftertouch AND a Note-On in this case, or just one or the other? Anyway, from your description, it appears that the re-trigger sensor (the middle sensor in a tri-sensor action) is positioned slightly above the "notch", which could be considered a small deficiency in the design, I suppose.

I thought the Avangtgrand would use the hammer for the velocity sensing, though. It looks like it might be using the key. (although it's hard to be sure yet)

Greg.

(i) Yes, you can hold it there for as long as you like; it is as though you've done nothing, pressed nothing. (ii) Not correct. You cannot "play off the jack."


If you slowly and gently depress to the notch, then, from that point, smash the key hard, from the notch-point to the bottom of the range, you get Aftertouch only. You do not get a note on. Again, you cannot "play off the jack."

What do you mean by this, exactly? Are you saying that if you stop just ABOVE the let-off point, you can then play a note forcefully, and a soft note will sound?

Greg.

I was talking hypothetically about a situation where, using the AG purely as a MIDI controller, you might reassign the Aftertouch 127 value to a very soft Note-on value (velocity 1, maybe?). In that event, if the AG sent an Aftertouch 127 value, Pianoteq (for example) would play a very soft note (Note-on 1).

The problem with reassigning the Aftertouch 127 value would be, I presume, that you would no longer be able to depress a note silently in order to lift the dampers -- in order, say, to utilize string resonance or to achieve what Dewster would call "silent replay."

Furthermore, the Aftertouch 127 value does not work in the same way as playing off the jack would work. It seems to be a signal that is sent at the point where the key has travelled to the notch, but not necessarily bottomed out.

In any event, now that I've thought about this a little more, I don't think reassigning Aftertouch 127 to, say, Note-on 1 would work very well at replicating playing off the jack. I hereby retract the portion of my previous post that you've quoted. I don't think that will work well.

I have to say, this is all getting fairly abstruse, even for my taste. The only reason I made the observation about "playing off the jack" was that I was amazed that the Roland RD-700NX achieved the effect. Is that because Roland intended to achieve the effect or is it just a happy coincidence? It's such a weird technique. As we all know, sometimes performers invent techniques that instrument designers may never have anticipated (playing a violin with the wood of the bow; playing an electric guitar near the speakers to intentionally achieve feedback; a glissando on a clarinet; a horse whinny on a trumpet; rolling a brush on a snare drum; banging on the cymbal stand with a drum stick; castrati). I put playing off the jack in that category. A nice trick to know that I'll probably never employ and would never have expected to find in a DP.

Note: There are no intentional double entendres in this post. I didn't invent these terms.

RE: the Roland, perhaps they simply positioned the middle sensor slightly after the let-off feel rubber flap? (why not design it as authentically as possible?) It doesn't seem to be such a hard thing to get right, even though it is admittedly an extremely subtle detail. Perhaps, though, if one tried hard enough, it would be possible to make the Roland fail as well. For example, there might be a tiny window of travel between the middle sensor and the let-off flap, where a note can be re-triggered without the let-off feel. This would not be authentic, because it would represent a piano that is repeating without the jack returning under the hammer knuckle.

Thanks for your clarification and patience, anyhow.

Greg.

Very curious?

Is this discussion to pluck holes in the AG as being "not worth the money, not as good as soft pianos, not as good as EP hammer actions?" How often do people play off the jack "whatever that is?" Don't get me wrong, I'm staying up with this thread cause I like my AG so much it's nice to hear people discussing an axe I own for a change.

For me practicing is more meaningful and jamming with friend with Aebersold is also closer to the real experience of gigging.

I don't think I could sit at the best grand piano (if I had room and money) and still get as useful of practice sessions. For me, the grand piano will be for performance, where I will hopefully be very comfortable after priming my skills on my AG.

Mate, just enjoy your N2. This particular argument is going around in ever decreasing circles - perhaps it will end when the participants disappear up their own bottoms...

I couldn't have said that better.

The next discussion will cover the imperceptible though extremely important air currents that arise from the movement of the hammers and how those currents effect the propagation of sound waves. The hammers in the AvantGrand are shaped differently and this subtle effect is lost.

On the other hand, these kinds of threads should be a useful mine of information for the R&D departments of Messrs Yamaha, Roland, Kawai and Casio etc.....in fact, I think they could probably save a few bob from their initial stages (the brainstorming, throwing a softball around a table stages) and put a few of these ideas into prototype production straight away.

Disappearing up your own bottom might never have been so valuable, despite the beneficiaries being elsewhere.

toddy, this technology is not exactly in its infancy. The first of this hybrid technology was introduced around 15 years ago or so, the Yamaha GranTouch.

Yeah, I'm surprised at the complaints about this thread. Pretty much everyone recognizes that AG is the best DP money can buy. Someone points out that it can't do a particular thing (playing off the jack). Interesting. No one cares about that particular technique too much (hardly anyone knew what it was when it first came up) but we try and figure out if there are other implications, like that Yamaha senses a repeated note only when the note is lifted higher than an acoustic would do (albeit by not far). That seems like a pretty legitimate topic to me, and very interesting. The fact that the best available piano is nonetheless imperfect shouldn't shatter anyone's world view. And any implications of that imperfection are interesting.

Remember, this is a gear forum. If we were uninterested in discussing the details of DP's and cared only about playing, we'd be playing instead of coming to a forum to discuss. I see no need to truncate this thread by offending its participants or suggesting that it's not a legitimate question or discussion.

In fact, since it turned rapidly to into a "complain about people who don't take as given that AG is perfect" thread, I kind of got lost with the conclusion that was being reached. In the end did we figure out what the issue is? Is it that Yamaha's sensation that a key has been lifted is triggered just a little higher than it should be relative to the jack friction point, or is the conclusion that the inability to play off the jack is unrelated to the use of the double escapement for rapid repetitions? I got lost.

I also think 128 final velocity levels are probably enough for piano (and take away level 0 for note-off, and perhaps level 1 for silent play). So in a sense the term "grayscale" is appropriate, since the eye seems to have a fairly limited resolution here too - black and white pictures look fine when I scan them using only 256 gray scales, but for post scan editing more levels are necessary so as not to easily reveal these levels, another parallel to the velocity shifting, scaling, and curving that most keyboards can do (and should use more resolution pre-filtering to do it right).

Using that MIDI data with, say, a PC based piano sample, and then performing further MIDI filtering on the MIDI data stream itself could be problematic, as this could eliminate some of the remaining 128 velocity codes from being useful, reducing resolution. So attempts at matching the response of the keyboard to the response of the sampler should start (and hopefully end) with the velocity filtering and scaling functions in the keyboard itself.

I think it's kind of a shame MIDI didn't use at least 256 levels - reserving a single bit in every byte strikes me as bad design. A 16 bit word with 4 bits of opcode and 12 bits of data might have been even better, but I imagine they were trying to keep costs down by employing the 8 bit UART chips that they had at the time (though why they didn't also pick a common RS-232 BAUD rate is beyond me).

Yes, I agree. While 128 *may* be enough...it does seem like they could have made it much better without stretching the limits of technology, even at the time. It would at least give me confidence that the 127 limitation isn't crippling my performances, and probably there are other instruments where a larger number of levels would be more important than it is on Piano.

I'm always rather bewildered when DP owners (myself included) take any hard analysis of the DP they happen to own to heart.

By almost all accounts the AGs are among the best widely manufactured DPs in existence today for replicating the full "true piano experience" but, being DPs and all, they are prone to certain limitations, and anyone contemplating the purchase of one should be fully aware of those limitations.

The DPBSD thread revealed audible looping on the AG sample set, admittedly probably not too obvious with normal playing (the looping is some of the best I've heard, though the loops themselves could use improvement) but IMO this somewhat or entirely precludes serious intimate solo recording of the instrument. I would argue that looping also hurts the tone, though that is perhaps perceived in a more indirect or subjective way than the repeated looping sound itself. It's my strong feeling that a DP in this class really shouldn't be hobbled by crude sample set compression techniques.

On this thread an honest misinterpretation of Yamaha's own AG press literature seems to have lead to the revelation that the AG keys are likely significantly shorter than those in the CFIIIS. Something which, if true, would have consequences for the evenness of touch when playing closer or farther from the key pivot point.

Kippsec brought up how his AG won't "play off the jack" even though many real pianos and some other DPs can do this.

All of this may seem like petty harping, but Yamaha implicitly issued something of a challenge with this line of DPs by marketing them as no-compromise AP replacements. And to those of us who are more technically inclined, they've more or less thrown down a gauntlet. Can AGs replace APs in many instances and do so realistically? Sure. Can they do so in all instances? No. Are they completely free of the typical DP issues? No. Could Yamaha significantly improve them? Yes. Is Yamaha's marketing "overselling" them? I would say yes and I find it annoying. Would a lot of people be really happy with one? Absolutely.

These are the key lengths from front of the key to the very back, right? The discussion in this thread has indicated that the measurement of interest is the distance between the fulcrum and keyslip, I believe. According to another website I was checking out, this is standardized at 10-3/8" (not too different from the RM3 key length). Do you guys know if that's correct or not? If it is, does that mean there would be no advantage to using a concert grand action instead of the smaller action in an AvantGrand?

In other words, is there an advantage to having a longer key from the fulcrum to the back of the key, the part the pushes up the hammers, as well?

I want to know this too. I sent an email to rick@roselandpiano.com and will share the info if/when he replies.

I asked the question on the piano technician's forum and got back the following answers:

1. The fulcrum is near the center of the key, regardless of key length, so there is a larger distance between keyslip and fulcrum in larger pianos.

2. Longer keys do have superior playability, so the 9-foot action is indeed better than that of the smaller grand.

I wonder if the sweet spot...the optimal key length...has been studied and found.

One reason I suspect that a longer keyslip-fulcrum distance is better is that this results in a key movement that is closer to purely vertical from front to back. I.e, I suspect the ideal would be a key that moved purely vertically. (just surmisal on my own part)

Greg.