I have come to the same conclusion as your mentor and do essentially the same thing. It took me about a year and a half of deliberate effort to be able to consistently accomplish this feat (i.e. make it "correct"). The one thing I add to it is that after tuning the third string I will often go back to the middle string and tweak it to whatever "sweet spot" I can find (if I can find one).
This method has the additional benefit of generally overcoming the tendency of the pitch to drop a little upon completion of the unison, so I do not have to worry so much about tuning the first string a little sharp in anticipation (as Virgil spoke of) as I accomplish what I want in the process of tuning the unison.
It is extremely difficult to do this accurately and stably, therefore putting the lie to those who have attempted to accuse me of "detuning" or being "sloppy" in tuning unisons (which is a very common reaction I've found within our circles in the tuning world). It quite the OPPOSITE of detuning. It is more like "retuning" with a purpose. And although it doesn't work on all pianos it's a great tool to have in the toolbox.
Original Message:
Sent: 02-18-2023 19:30
From: Fred Sturm
Subject: New Pianoscope Feature Freezes Pitch During Attack Phase
You might be interested in reading the articles about unisons on the Education Hub in the unison segment of the beginning
Tuning Curriculum. Scroll down and you will find links to nine articles on the topic. Those were the ones I selected from 63 hits searching "unisons" on the
Journal index.
Virgil Smith, of course, was one who advocated most strongly for "voicing unisons" (tuning them to make them 'sound the best' in terms of bloom, sustain, etc. In describing this in his autobiography, he noted that tunings done like that didn't last very long.
If you can tune every unison within 0.1¢ and then "detune" one of the strings to be 0.2¢ off from the other two (and be stable), my hat's off to you. My take is that the real struggle is over stability of all the strings of the piano. No matter how carefully and skillfully you tune, a retuning after a rehearsal will result in a few notes having a bit of a "wowr" in them. Everything may be quite acceptable in that condition, no complaints, but some movement has taken place.
My take is that there may be a few isolated situations where it may make sense to "detune" a unison enough to give a particular note more sustain, if that note sticks out for being dead for some reason (maybe the note is over a rib or something), but for my purposes as clean and stable as I can achieve is the gold standard. I have yet to have complaints about the results. I do get occasional complaints about fairly small deviations from clean.
About tuning to the attack, use of various free and reasonably precise tuning apps shows clearly that the initial pitch read is significantly different (sharp) from the decay pitch. When I was beta-testing OnlyPure (which lacked a buffering for the display and had a noticeable time lag between sound and display), the initial visual was way off what happened subsequently. High speed videography shows that it takes quite a few microseconds before the string settles into a steady pattern.
Attack includes considerable noise (percussive non-pitch sound), the chaos of the string resolving its initial deformation into a predictable pattern of partials, and a delay in response of the bridge/soundboard. In my experience, it makes no sense to aim for the initial 0.5 seconds or less (though it is nice if your ETD will give you information real time about that initial onset of pitch). Pitch within this narrow a parameter only becomes significantly audible when it is sounding with the pitch of another note, and that relationship doesn't become apparent until at least nearly one second into the sound envelope.
(BTW, looking at the visual read out of intervals in an audio application like Ocenaudio, as I described in my articles "Measuring How We Tune" recently, is an interesting way to confirm this: the regular patterns of beats don't begin until after a few tenths of a second).
Regards,
Fred Sturm
http://fredsturm.net
www.artoftuning.com
Original Message:
Sent: 2/18/2023 6:12:00 PM
From: James Busby
Subject: RE: New Pianoscope Feature Freezes Pitch During Attack Phase
All
I've had this conversation (about absolutely perfect unisons not sounding as good as unisons that are out a tiny amount) with many top notch technicians. Here are some things I've gleaned and some bits of info from years of experience.
1. At least one Journal article from maybe 60+ years ago said to put the first two strings as perfect as possible, then put the third string to where the "timbre thickens" but does not beat. I haven't been able to find that article in again. Maybe you techies can find it.
2. My first mentor (born in 1898 died 1996) told me the same thing, and demonstrated that to me. He tuned several notes, and I measured them with my Sanderson 1, as well as listened carefully and counted beats. To the best of my ability at the time, I discovered nearly every string he tuned "out" (now that is my word. He said they "correct") were somewhere between .1 and .3 cents off from the others. Besides using the Sanderson's lights, I found that the two strings would beat in about 4 to 5 seconds to do a full cycle. By that time the volume had diminished enough that you could barely hear it complete the beat. He claimed it was "pure, but better than a dead sounding unison. It has life." His words remember.
3. I had a concert artist tell me he wanted more "bloom" in the notes. Remembering my mentors admonition to put one string out (again, my word) I tuned the first two dead on (SAT) then made the third string about .2 out. Now that is very hard to do! It was as good as I could anyway. The artist nearly jumped up and down with excitement. "That's what I want!"
4. Upon discussing this topic with Richard Davenport he laughed and said "There's a saying that goes 'Tunings always sound better on the second day'".
I hope this adds to the discussion. I tend to tune everything as pure as I possibly can, hoping Richard was right.
Jim Busby
Sent from my iPhone
Original Message:
Sent: 2/18/2023 5:08:00 PM
From: Peter Grey
Subject: RE: New Pianoscope Feature Freezes Pitch During Attack Phase
I had about an hour long conversation with steve yesterday and tried to help him (as a scientist) to understand that many of us tuners (who are definitely NOT scientists) have come to believe that tuning actual "perfecto" unison is not necessarily the best thing from a musical standpoint. True, we certainly don't want beats in our unisons, however slight deviations from absolute phase unity can lend a rather more pleasant sound (which the coupling action of the bridge structure then unifies in the aftersound), and can actually improve sustain and melodic color, etc.
After some discussion he seemed to understand this (though not actually agreeing, he fully understood what I was saying). I indicated that I felt this software tweak could have some value in doing this little unison tweak. I also indicated that there are some pianos that won't tolerate any deviation from "perfect" wherein Frank's feature could have value as well.
He was surprised to learn that we analog tuners actually have analog ways of reproducing this digital feature to a very close degree (maybe not one tenth of a cent but certainly in the ballpark), and have had this ability for a long time but we don't write scholarly peer-reviewed articles on it (like he's used to in his field).
We had a good chat.
Peter Grey Piano Doctor
------------------------------
Peter Grey
Stratham NH
(603) 686-2395
pianodoctor57@gmail.com
------------------------------
Original Message:
Sent: 02-18-2023 08:40
From: Cy Shuster
Subject: New Pianoscope Feature Freezes Pitch During Attack Phase
I'm reminded of the quote "Everything should be as simple as possible, but not simpler."
The app does a great job of snipping out a tiny piece of information. It would be useful to know if tuning using that snippet results in a better sound. But there's so much more going on when a note plays that is not shown.
The TuneLab spectrum display (or any RTA spectrum display) shows pitch and volume over time, which greatly helps me make my own best judgement, especially in tasks like measuring a pitch for a tuning exam. The SAT LED display similarly presents a small subset of the available information.
When aircraft altimeters were first digitized, the numbers would snap to their new values, like a thermometer or any other simple meter. Pilots soon discovered that the old rolling wheels of digits gave additional critical information, specifically a fine display of the rate of change of altitude. Digital altimeters were changed to mimic the old rolling wheels.
Secondly, I disagree with the value of tuning to a tenth of a cent. That's about as useful as tuning to a hundredth of a cent, or mowing a lawn to a tenth of a millimeter. It would be like painting the Golden Gate Bridge: when you get to one end, go back and start over, because it's all changed.
The one part of the piano where focusing on a time slice would be useful are the notes beginning an octave above the highest damper. I wish TuneLab would freeze the display before the sympathetic vibration of the lower notes kicks in (like when tuning F7 to C8). I usually lay a mute on those lower strings (thanks to Don Mannino's class).
--Cy--
------------------------------
Cy Shuster, RPT
Fairfax, VA
http://www.shusterpiano.com
Original Message:
Sent: 02-17-2023 19:41
From: Paul McCloud
Subject: New Pianoscope Feature Freezes Pitch During Attack Phase
Hi Ed:
I don't know if you have the app, or have tried it. The inharmonicity curves show very clearly where there are scaling irregularities. If you make experiments, you can take the measurement again and see any differences on the curves. You can examine various octaves, and all intervals calculated from the inharmonicity measurements. What you're able to do with that information I guess is up to you. Because of the fact that every note is measured for inharmonicity, you can go in and see what is happening for each individual note. On the lower notes in the tenor, the tension is much less, so you get a kind of 'boingy' effect. Think like the low tenor of a Baldwin Studio Upright. Unless you do some major changes there, I doubt much can be done. Right now, with this capability, and the red line indicator, you can see how much a hard blow vs a soft blow affects the change in frequency during the attack, and at what time frame. That might help to understand what these GB-1 strings are doing.
Refinements are being made and will be forthcoming soon to deal with some anomalies of this feature. Curiously, it's a side-gig for Frank, so he has to find the time to work on it while he's doing his day job. But yes, he's quite open to any suggestions and tries to implement any ideas very quickly. Steve and I had a Zoom meeting this morning to discuss the new feature and answer questions.
It appears that there is much inconsistency from note to note in how they respond to hard and soft blows. I'm sure it also has to do with setting the pins while tuning and other factors in the tuning process. I try to use the same technique on each pin so that each string has the same tension in the non speaking segments. You can easily see where the tendency of the pitch changes are sharp or flat in spite of the indicator being very close. The background "strobe" works very nicely for this. Any left or right movement gives you hints of too much or too little tension that might affect the pitch of the note.
------------------------------
Paul McCloud, RPT
Accutone Piano Service
www.AccutonePianoService.com
pavadasa@gmail.com
Original Message:
Sent: 02-17-2023 12:47
From: Ed Sutton
Subject: New Pianoscope Feature Freezes Pitch During Attack Phase
I've watched the development of PianoScope over the last two years. Frank has consistently asked for technician response and responded quickly to many suggestions. The app is extremely adaptable to user preferences.
This particular video was made with a Fazioli piano. I don't know to what extent the variables so nicely measured would be audible. I certainly don't hear the pitch drop of a few hundredths of a cent.
But what if the same tests were performed on the lowest plainwire string of a Yamaha GH-1?
Could this program in effect offer a way to quantify the poor behavior of a foreshortened scale or the irregular impedance of a poorly made soundboard?
Could it measure and demonstrate the improvements made by adding riblets or bridge weights?
------------------------------
Ed Sutton
ed440@me.com
(980) 254-7413