Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Tuning philosophy Q:

Well, perhaps.

I was trying some pianos today in a good shop and noticed that there was a real dissonance between F and a lower D-flat-- the first pitches of the Chopin Prelude #15. Real bone-rattling, fingers-on-chalkboard kinds of beats being thrown out of the thing.

I'd thought that maybe the piano was out of tune, but there were no other obvious discrepancies, and it was a Yamaha AvantGrand, in any case. So presumably, that is what the manufacturer wanted us to hear.

Has anyone else noticed this? And is it a correct fall-out from equal temperament? I never picked that up when I was a kid, and after I focused on violin there was no reason not to play that-- or any other interval-- in tune.

Is there a way of minimizing the clash? Something, perhaps, about Steinway-ish lack of clarity or some other gimmick that fudges the problem?

If you want to test whether it is out of tune, compare this Db-F with the adjacent major thirds, C-E and D-F#. They should beat approximately the same, C-E a little slower, D-F# a little faster.

Yes it may have been out of tune, as it was perhaps an equally tempered third, so it is 14 cents away from what a violinist or string player would play if they were listening to their intonation. It is an unavoidable by product of equal spacing of the semi-tones. This amount of dissonance is, by modern convention, accepted as the maximum dissonance that is acceptable. What many of us find unfortunate is that in equal temperament, every single third is this far out of tune, hence, every single third is at the limit of tolerance so that none are over it.

In practice, I find the vast majority of pianists I tune for prefer a graduated size of third that follows the historical pattern of widening tonic thirds in the keys as one progresses around the circle of fifths.
Regards,

I should have compared that with other thirds, then (or 10ths, or 17ths). Or listened through headphones to see any differences between the Yamaha and Bosendorfer samples.
Odd that it hit me so hard, though, when it never did before. Perhaps the Yamaha series of overtones coming through the speakers is pure enough that it sets up some really nasty clashes, that would normally be disguised by the more familiar Steinway lower-midrange soup?
Anyone else with a well-engineered hybrid, or perfectly-tuned acoustic piano is welcome to share results.

Just so you understand what Ed Foote is saying, he says that particular interval should beat more than just about all other major thirds on the piano, because it is the third for the scale which is most remote from C major. But if Yamaha is using such a radical departure from equal temperament, they should tell you.

Check whether the octaves, fourths, and fifths of those particular notes beat excessively. That will tell you whether it is just one of those particular notes, or if it is the tuning.

There may be some self-diagnostic or reset if there is a problem.

um, no. What I am saying is that the Db-F, just like every third in equal temperament, its nearly 14 cents wider, ("sharper" in string player's vernacular) than pure. I am assuming that the OP was listening to an equally tempered scale. Had the temperament been one of the conventional, irregular, circulating, tunings, that third would have been on the more expressive side of the options, but without specifying such, the normal assumption is that ET was in place .
Regards,

I just sold my AvantGrand NU1, so I cannot record this test for you. However if you go to this site digital forum (or one of plenty other digital piano forum and clack out there on the internet) you will find no shortage of people with that instrument and I am sure a few will record that interval for you. That is very easy to do on an AvantGrand, even with all the temperaments available to the instrument. Then we can talk about some specifics recordings, rather than just speculations about what you might or might not have heard.

Please don't judge equal temperament by any electronic keyboard. There isn't one on the market that sounds a decently tuned ET. They all have tunings which have been "adjusted" to sound more "interesting".
No real tuner would dare leave a piano tuned as these electronic ones are.

LOL - that's errant nonsense! There was an old thread a few years ago whereby people were trying to tune a perfectly progressive ET ... someone sampled an electronic instrument and it was pretty close. There are instruments which have multiple samples of each note - so the tuning is as good as the original instrument was tuned. On my ancient Yamaha P120 I can clearly demonstrate the piano voice's tuning is stretched (as expected) when playing along with a voice that has no inharmonicity (such as the organ or harpsichord). Then, of course, there's purely modeled instruments (like PianoTeq) where you can tune each note individually if you want.

I suspect the instrument the OP was playing had had a non-ET temperement set.

Here's my challenge to you
find an electronic keyboard with either or both of these...
A scale with thirds that progressively speed up nicely. Or
An octave within which the individual notes sound pure like a well tuned unison should. No meowing going on.
If you really think you have one please tell me the make and model, or video it with good sound and I will withdraw my claim!

Nick

For PianoTeq you could try this one as it's a relatively slow piece:

For the Yamaha AvantGrand, you could always look at the Yamaha marketing videos as the sound is pretty well recorded. Eg.

And now I see on the N1X thread that there is a choice of temperaments available. I wonder if someone had messed with the samples on the showroom?

That would be my guess Maestro!

I think that there's confusion here. These videos don't even attempt to address what I'm talking about.
My challenge remains.

You said "An octave within which the individual notes sound pure like a well tuned unison should. No meowing going on."

I didn't hear any miaowing in either of the two videos - if you did, can you point out the times where you could hear it?

I do own a 24 year old Yamaha P120 and I can definitely hear the loops repeating as the samples are only a couple of seconds. But modern sampled pianos do not use loops at all - so any miaowing would come from any mis-tuning or imperfections in the piano being sampled. As for the physically modelled pianos like PianoTeq, you can set the "unison width" to whatever suits your taste - in the video above you can see the control and hear the result. If I had access to this I'd create a recording of consecutive 3rds ... but I don't. Given they you can set the tuning on any note, you can really create any kind of temperement you want. Note also that PianoTeq has an octave stretch parameter too ... so you can also create impure octave as well.

Ok let's have another go. Go to https://www.modartt.com/grotrian then listen to this track - it should have the "no meowing well tuned unison" sound you're asking for.

https://www.modartt.com/data/audio/...ge%20a%20Rameau%20-%20Young-Ah%20Tak.mp3

Here is the Steinway, Yamaha and Bosendorfer on Logic Pro.

Thanks Jean - that's PianoTeq, right?

No. That is the stock sampler in Logic Pro X on Mac.

Sorry to put you to the trouble and I appreciate your post, but no music will demonstrate what I'm referring to.
Accordeur has posted a suitable video which I think demonstrates my point exactly...listen to the beat rates as he goes up the thirds...

Also I was actually referring to stand alone electronic pianos, not computer programs, but even so, hopefully he will post a video of an octave's worth of single notes so that you can hear the unison quality.