Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Novel Piano Tuning App

Could also be a feature for the pitch raise mode as it would allow an easier way to do muteless pitch raises.

Paul.

Just to add some encouragement, providing assistance like this would be really helpful for owners not wanting to do a full tuning but rather to touch up unisons between tunings. Automatically matching the peaks to the strings would be great, but if that proves unreliable you might ask us to pluck or mute the strings on a note so that you can definitively establish the peak/string relationships . That might be too time consuming for a full tuning but fine for some touching up.

Not affordable UI and UX design in this app. Think about creating my own.

In what way is it not affordable? You think you could do better for the price Frank is charging? You must not value your time very much ...

Sadly. I can't promise anything here. I would have to do some research first. I think it is easy to do for pitch raises, because the peaks are moving quite strongly when you start tuning the first string and so it is easy to identify them. When they are already close together and close to the target, it is harder, because there is less movement involved.

I have a question to those of you who have been using pianoscope in the last weeks. A senior piano tuner has contacted me and told me that he has tried pianoscope but he was not satisfied. He said, he ended up with a tuning in which the major thirds over the bass/tenor break did not have a progressive beat rate. He described them as sounding "lumpy". When I however look at the calculated beat rate curves for thirds for his instrument in pianoscope, they look as smooth as butter. So something must have gone wrong when he was applying the tuning to the instrument. Either pianoscope has somehow guided him to slightly wrong pitches, or he must have done something wrong, or both.
So to narrow things down, I am interested in your experience: How did the thirds progress in your tunings? Were you satisfied with them?

I will be tuning a very small Samick baby grand tomorrow. I will pay special attention to the thirds and will have verituner running at the same. I will let you know my observations.

That was one of the reasons why I opted for measuring inharmonicity value of each note rather than sampling. I assume Verituner has a similar approach too.

Hakki, thanks. This totally makes sense. I thought the gentleman must have measured all notes, but now that I think about it he might have fallen into this trap. I think I will need to make the IH setup clearer in my app. That for a rough tuning it is sufficient to only sample a few notes, and that you should measure more for fine tuning.

Just to be clear:

PiaTune uses an aural like scheme where inharmonicity is measured for every note.

As the tuner proceeds, the complete tuning is recalculated after tuning each note.

PiaTune's approach (like Verituner) might seem slow at first impression.

But I opted to employ this scheme in my app PiaTune, to ensure a fine tuning regardless of the scale design of the piano.

Simplicity can be good. If tracking the strings when open unison tuning proves hard a simple aid for muted tuning could be helpful, something like just leaving a coloured dot on the screen as a reference for the first string's main peak when tuning the second, then two reference dots for tuning the third string?

Yes, I am aware of that. It is a fundamental design decision for the app. And using an aural-like sequence has its benefits. My approach needs a full IH model upfront, because I want to perform a global optimization for the whole tuning. And for it I also want to take the partial intensity distribution into account. If the IH model is too coarse, because you measured too little notes, the resulting tuning might have problems. But I made pianoscope fast so that measuring the full bass & tenor should not take much longer than a minute.

I will improve the workflow of the IH measurements in pianoscope to make it clearer to the user, how many notes need to be measured to get a good result.

Is it possible that reducing the measurement of each partial down to a single inharmonicity constant loses the specific partail information needed to really accurately predict 3rds speeds for a specific instrument?

Hakki, does yours boil the data down on each note to an inharmonicity constant?

Ron Koval

Ron, I have tried both retaining the measured partials and using a single constant while developing PiaTune.

After extensive testing I came to the conclusion that, if modelled correctly, a single constant could reproduce the measured partials accurate enough not to cause any discrepancy in beat rates.

Currently PiaTune is using an inharmonicity constant.

Since PiaTune measures every note as you tune and bases its calculations on measured notes only and does not use any guesswork, the end result is similar to an aural concert tuner’s tuning.

Verituner keeps the partials information rather than using a single constant, as far as I know.

I don‘t think so. From my experience, the combination of fundamental + IH is pretty good at describing the first ten partials. The bigger source of error is not knowing a strings IH in the first place because you did not measure it. Especially for bass strings which are uniquely fabricated. As you know, even two bass strings in a unison can have a quite different inharmonicity. For strings you did not measure, pianoscope builds an interpolation, which is quite good for well balanced pianos. But if you want to be sure and do a really fine tuning, you need to know the IH of all bass and tenor notes. There are little surprises with IH in the treble, therefore you do not need to measure a lot there. I am working on an update which makes this clearer to the user.
And the IH slightly changes along the duration of a tone. pianoscope focuses on the first ~second of a tone, which is the part which you hear the most in actual music.

I have yet to meet an electronic device that can navigate the bass/tenor break ideally without having to make aural (analog) adjustments. But then again, I do not own any apple equipment. And then again, why spend my time looking at an app when I'm just going to have to correct it anyway with my ears and brain?

Peter Grey Piano Doctor

Hakki and Frank, thanks for the clarification. It was something that I've wondered about for some time!

Ron Koval

Well I went to tune this short Samick today. I don't know if something went wrong since the last build, but I measured all notes from A0 to A6 and the bass turned out so sharp it was unplayable. I did not have a lot of time so I just quit pianometer. I am not well versed in the technical terms so I am afraid I can't help you very much. All I know is that I never got to test some thirds as the octaves were just really bad I quit after an octave and a half.

I am sorry for this, but I guess this is what a beta test is for. Thanks for testing it. Can you send me the pianoscope document with the measurements?

Hi Frank,

At what email? I can't attach a document if I follow links on your website.