This is very interesting...
There is Bernhard, Only Pure, maybe Virgil, maybe Kent to some degree, if I am reading into his excellent articles without being totally lost, questioning, I think, what the utility of extracting 1st partial A 440 as an absolute benchmark is, when the perception created by the totality of the A4 tone may be registering, either aurally or with an appropriate algorithm, as a different value.
Then there is Fred, Don, Jason, and probably many or most tuners, saying that an extracted portion of the tone, disregarding the effect of the whole on the perception of the complete sound, is a valid benchmark. Saying that a model of the simplified tone, is more important than the perception of the whole tone, mainly because its easier to measure, and easier to fit into a simplified model of the tone.
One values the complete tone, attempting to allow a messier, more complicated reality to define what the perceived pitch is, the other, valuing a simplified model of the tone, because its easier and convenient to assume that the simplified model reflects the total sound.. .
Fred< orchestras do their own types of adaptations regardless of what piano technicians do.
If read him correctly, Fred does not present the customary orchestral tuner approach as having an absolute meaning. Rather, his comments suggest, that since the perception of the whole piano tone is so complex, no one without a device measuring select simplified signals of the complex tone, really can tell the difference, and adjusts aurally as necessary, no matter what the tuner thinks he did.
There is an level of unavoidable uncertainty here that I find quite comforting.
------------------------------
Jim Ialeggio
grandpianosolutions.com
Shirley, MA
978 425-9026
------------------------------
Original Message:
Sent: 12-01-2017 18:21
From: Fred Sturm
Subject: OnlyPure calibration
OnlyPure pitch information is not base pitch (first partial) of A4. It is some sort of autocorrelation function of the whole tone, that means an integral pitch impression including higher inharmonic frequencies and damping effects. So what is shown in OnlyPure for the pitch is a value that corresponds to a pitch impression compared with a pure harmonic tone of that pitch value.
The problem with this reasoning is that pitch standard is based on physical measurement, not on psychological "pitch impression." From a practical standpoint, for the past century and more piano tuners have used a reasonably accurate pitch source, and have matched the first partial of A4 (or C5 for those who used C forks) to the source. They have increased the accuracy of the pitch transfer from an aural source by comparing beat rates with F2, again ensuring that it is the first partial of A4 that is tuned to the standard. Furthermore, all electronic tuning aids (with the sole exception of Tunic software) do their best to place the first partial of A precisely at 440 as the default.
This is the tradition we live in, and orchestras have lived with it all along (to the extent that matching between pianos and orchestras is important - more for some of us than others, and orchestras do their own types of adaptations regardless of what piano technicians do). If matching a Steinway grand at first partial A4 440 means that the orchestra tunes to 441 (if they are that accurate so that you can say within 1 Hz what the average is), that is what they have been doing all along. There is no harm in it.
Most professional musicians, and many ordinary people with little musical background, have pitch reading devices that are quite accurate. If I were to tune to OnlyPure, I would have musicians and piano tuning colleagues commenting that I always tuned flat, wondering why. The oboist has his device sitting on the stand, as does the concertmaster, in the concert situation.
I think it is unwise to choose to be different in this way. I don't think it will reflect well on the software in the long run. I do understand the argument, but I don't see its relevance in the real world. And I don't see the relevance of the argument that if OnlyPure were to offset to produce 1st partial A4 at 440, that would make the software wrong when tuning a tone with little to no inharmonicity. It's piano tuning software, isn't it? it is meant for inharmonicity.
------------------------------
Fred Sturm
University of New Mexico
fssturm@unm.edu
http://fredsturm.net
http://www.artoftuning.com
"We either make ourselves happy or miserable. The amount of work is the same." - Carlos Casteneda
Original Message:
Sent: 12-01-2017 14:52
From: Bernhard Stopper
Subject: OnlyPure calibration
Ok i think i need to try to explain further:
Quote Fred Sturm: "I don't understand why it is essential for the base pitch (first partial) of A4 to be variable in order for the software to work. I especially don't understand how a variance of over 3< can possibly be acceptable."
There seem to be a misunderstanding: OnlyPure pitch information is not base pitch (first partial) of A4. It is some sort of autocorrelation function of the whole tone, that means an integral pitch impression including higher inharmonic frequencies and damping effects. So what is shown in OnlyPure for the pitch is a value that corresponds to a pitch impression compared with a pure harmonic tone of that pitch value.
Quote Fred Sturm: "Just to add another specific observation to Jason's that opened this thread, I tuned a couple Steinway Ds this morning. A4 spot on at 440. Measured with OnlyPure. It told me the note was about 5< sharp. I calibrated OnlyPure (changed the 440 window) and it needed to be set to 441.1 - 441.3 to read an A440 as being in tune. This is not a minor quirk."
As mentioned before, OnlyPure pitch information is not pitch of partial 1 of A4, but a pitch impression compared with a harmonic tone: In the example above of the Steinway D this would mean the following: An A4 tone of a Steinway D with an A4 partial 1 tuned to 440 Hz, a harmonic tone with no inharmonicity and no damping (with harmonic partials) of a pitch of about 441.2 Hz is required to produce about the same aural pitch impression as the A4 of the Steinway D tone with partial 1 tuned to 440 Hz, because of the higher inharmonic partials involved.
So the difference of OnlyPure effective pitch (OEP) and pitch of partial 1 on A4 is not a quirk, but by definition and intention.
As i mentioned before and as Fred proposed, this could be handled by calibrating things out relative to partial 1 (as possible with former versions of OnlyPure). The drawback would the be that after calibration the autocorrelation (whole tone) reading of OnlyPure on a harmonic tone would be off the amount of calibration in the other direction, what would be not meaningful at all.
To avoid too strong beating between instruments with relative high inharmonicity on A4 (Steinway for example) and a harmonic instrument (Oboe of orchestra), a value half between OEP and partial 1/harmonic tone difference may be preferrable. In the former example when OnlyPure pitch is set to 440 Hz, Oboe can have between 440,6 Hz for a lowest overall beating on A4, or 441,2 Hz for the same pitch impression the piano does with OEP (not partial1) set to 440 Hz on A4. Tuning partial 1 of A4 of the piano about a half beat lower than the nominal pitch of the orchestra (Oboe, Violin) is pretty common in the concert circus.
------------------------------
Bernhard Stopper
Klavierbaumeister
Tuebingen
Original Message:
Sent: 12-01-2017 13:39
From: Fred Sturm
Subject: OnlyPure calibration
Just to add another specific observation to Jason's that opened this thread, I tuned a couple Steinway Ds this morning. A4 spot on at 440. Measured with OnlyPure. It told me the note was about 5< sharp. I calibrated OnlyPure (changed the 440 window) and it needed to be set to 441.1 - 441.3 to read an A440 as being in tune. This is not a minor quirk.
------------------------------
Fred Sturm
University of New Mexico
fssturm@unm.edu
http://fredsturm.net
http://www.artoftuning.com
"We either make ourselves happy or miserable. The amount of work is the same." - Carlos Casteneda
Original Message:
Sent: 12-01-2017 10:56
From: Fred Sturm
Subject: OnlyPure calibration
I don't understand why it is essential for the base pitch (first partial) of A4 to be variable in order for the software to work. I especially don't understand how a variance of over 3< can possibly be acceptable. A4 is very close the the middle of the piano, so why can't it serve as the nexus around which the rest of the tuning is crafted? Alternately, why not include within the software the ability to play A4 first, and have the software automatically read the profile of that note and adjust the inner pitch of the template?
The user should not be responsible for finding some other way (outside of using the app) to assure that the pitch of the tuning will be accurate. Accuracy of 0.1< isn't really important, though it is desirable that that degree of accuracy is available, but + or - 3.0<+? Making the piano I am tuning compatible with an additional piano played at the same time comes up very rarely, and I would say that that ability should be an additional optional feature rather than a major driving force in designing the software.
------------------------------
Fred Sturm
University of New Mexico
fssturm@unm.edu
http://fredsturm.net
http://www.artoftuning.com
"We either make ourselves happy or miserable. The amount of work is the same." - Carlos Casteneda
Original Message:
Sent: 11-28-2017 15:08
From: Bernhard Stopper
Subject: OnlyPure calibration
A deviation of 3.4 cents you encounter on A4 is of a correct magnitude for OEP (OnlyPure effective pitch) for a typical piano scale, so there is no error with your devices or your software. OEP functionality is essential for the ability that OnlyPure can deal with piano nonlinearity without the need of pre-measuring notes before the tuning process or OnlyPure's ability to tune pianos with different nonlinearity profiles nicely together by using the same OEP for both instruments.
------------------------------
Bernhard Stopper
Klavierbaumeister
Tuebingen
Original Message:
Sent: 11-28-2017 14:29
From: Jason Kanter
Subject: OnlyPure calibration
Bernhard, thank you for chiming in. Your answer makes sense and yet it does not account for the size of the deviation that I find. The inharmonicity of the partials cannot account for 3.4 cents difference in the fundamental. Kent says that even without tapping the offset to match a known 440, one can just start tuning and it will be very close. Not so in my case.
To review: the first time I used OnlyPure, I did tap the offset to match the given piano because I did not wish to do a pitch correction. The second time I used it, I first raised the pitch using Tunelab's overpull function, so the piano was now close to A4=440. Then I switched to OnlyPure, whose offset display showed 440.0, but the actual A4 string's green triangles were 3+ cents off to the right. If I just trusted OnlyPure, I would have retuned this piano at 439.14.
When I tapped the offset to match the correctly tuned A4 string, the displayed number in the upper left of the OnlyPure screen was 440.9 (which is 3.4 cents sharp).
I understand that this displayed number is not germane to the actual tuning: that I use the offset buttons to match A4 to a trusted source, and proceed to tune from there. That's fine. But something happened with my instance of the software, which resulted in the displayed number being wrong. The size of the discrepancy is not adequately explained by your description of OEP. This is an error that should be investigated and fixed.
Perhaps there is a clue as to what happened, in the odd fact that the same unwitting offset that I describe above (on an iPad) appears also in my iPhone, which I never "calibrated."
It seems that the first tuning sets a stamp on the software.
Please comment again, after which I expect further discussion will lead to diminishing returns.
Cheers
Jason Kanter
| || ||| || ||| || ||| || ||| || ||| || ||| || |||
jason's cell 425 830 1561
On Tue, Nov 28, 2017 at 2:25 AM, Bernhard Stopper via Piano Technicians Guild
Original Message------
Jason,
OnlyPure does not measure partial 1 on A4 by design. It rather interprets some sort of "OnlyPure Effective Pitch" (OEP) by a proprietary autocorrelation function, where OEP deviates from partial 1 pitch with increasing nonlinearity (inharmonicity, damping). On a signal with harmonic partials, OEP and partial 1 pitch is exactly the same. With former versions (still there with the Android version) it was possible, to "calibrate" this deviation out. Generally calibration was also necessary, because soundcards were not consistent enough, which is no longer true on the newer devices. The drawback by calibrating the difference of OEP and partial 1 caused by nonlinearity is that with a harmonic signal, OnlyPure would read a difference between OEP and partial 1 then, which makes no sense and is the reason why the calibration has been pulled off.
Kent Swafford correctly outlined how to deal with this.
------------------------------
Bernhard Stopper
Klavierbaumeister
Tuebingen
Original Message:
Sent: 11-27-2017 12:57
From: Jason Kanter
Subject: OnlyPure calibration
Users of Tunelab and OnlyPure
I've encountered a sticky calibration problem. I posted this on the ptg googlegroup, but I want input from Kent Swafford and Bernhard Stopper, please.
I just tuned my piano with OnlyPure on my ipad mini 4. Then I play it for a while and the treble is gorgeous but I want to investigate the bass, so I open TuneLab to check the frequencies of selected partials. I look at A4 and Tunelab says it is 3.4-ish cents flat. Verituner agrees: 3.4 cents flat.
Hmm, I think to myself - maybe it's something about this iPad, though I know I have it properly calibrated per Tunelab setup instructions. I switch to my iPhone and open OnlyPure for the first time on that device. It shows the A4 as wandering slightly sharp of 440, by 0.07 to .8 cents. So OnlyPure on the iPhone agrees with OnlyPure on the iPad. Again, on the iPhone, both TuneLab and Verituner agree that this note is 3.4 cents south.
I've uninstalled and reinstalled OP, but it remains in a state where it thinks 439.1 is 440.0. And this is so on BOTH devices, though I have only used it on one device. How can I get OnlyPure calibrated correctly, on both devices?
Thoughts?
| || ||| || ||| || ||| || ||| || ||| || ||| || |||
jason's cell 425 830 1561