Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Should There Be A Standard?

Absolutely! All persons, living or dead, are fictional, and any resemblance is purely coincidental.

0.9>0.2 so you can score 100% with nonprogressive M3's.

Kees

I wonder what temperament this is, and whether we can gather any information for a "Standard"? Listen particularly to the tuning.

Wonderful instrument from the silent movie era...a lot of movie houses would buy this instrument to accompany the silent films...it came in different sizes as well. He must be one of the best in world at playing it. (watch what happens at the very end)


https://www.youtube.com/watch?v=tT1LTdUnyVk


adding a nice story about one of the pieces.......

https://www.youtube.com/watch?v=wTGEvqEmR-4

Exactly! As I point out in my own version of "Let the piano tell you", when you compare a series of 4 CM3's (not just three), if the top M3 (F4-A4) is too slow, then the bottom M3 (F3-A3) is too slow. You then lower both F3 & F4, then possibly also slightly adjust C#4 so that everything fits.

Works the same if the top M3 is too fast.

When Jim Coleman Sr. was reviewing what I had written about that, I now recall that he said he had written virtually the same thin sometime in the 1980's. However, when you posted what he had written in another thread, that was the first time I had ever seen it.

I certainly have not seen or read everything that Jim Coleman wrote but it does show that when something exists to be discovered, it will be. It was actually Viviano, someone who was trying to learn how to tune a temperament octave and who was trying to help me edit my material so that it would be clearly understandable who pointed out to me that what I was doing was treating the F3-F4 octave as a pair. If one F is moved, the other must be moved by the same amount. If that does not yield the correct result, then try again.

Usually, it only takes one, maybe two adjustments of both F's to find the correct balance but for a beginner, I can see how one may have to go back and forth a few times before getting it right.

I would also like to say that there is no reason why a person cannot use a traditional 4ths & 5ths sequence that is practiced and familiar and then use CM3 tests to sort out ("nitpick") small errors.

In the example that Jeff posted, if he had listened to the chain of CM3's from F3-A4 after having arrived at the results in the usual and practiced way, the small error would probably have been evident.

In the end, all 4ths & 5ths need to sound very much a like, none too pure, none "beating", all M3's and M6's progressive and all CM3's also having the proper but small, slower/faster relationship. When you have all of that, you have what would be an indisputable ET. It can exist on any piano, regardless of scale.

It can exist, but this does not prove that it does exist. Every single recording posted here of M3/6's, including your own which I took from your PTG educational video, were not completely progressive.

Moreover the tolerances of the PTG tuning exam are such that it is possible to score 100% without having progressive M3/M6.

Kees

Something strange : as I am actually testing for methods to allow different levels of consonance in chords, so to favor more contrasts in modulations, I had to tune back to a "standard" temperament as I did before, using a F3- F4 sequence.

I immediately noticed how what some tuners call "the shimmering of ET" was back, it was mostly sensitive in the higher region of the temperament and beginning of the 5th octave. it get better for one octave and a half then, and then high treble can be acceptable, depending the way it is tuned.

In basses not all octaves are harmonious, notes are sounding as without clear link to mediums (again , depending of the compromising used)

The point there is not at all that "all chords sound alike" , but they have a strange behavior, that does not seem to have much tonal or harmonious "meaning"

I seem to hear that strange sensation linked to fast beat rates, which are in the end unavoidable, but seem to me that when the notes from the octaves , 12th, etc under the one played are consonant clearly, they lower the harsh sensation of fast beats (6ths M3 for instance)

I am not so surprised that musicians regularly seem to appreciate Well tuning, while I still believe this is basically due to a misconception of how to install ET on a piano.

My point is that tuners could work some high consonance tuning , so they begin to recognize the level of consonance they obtain when tuning in their more usual way.

"pure" 12, Chas, pure 5th, whatever, the quality of the "halo" provided and the crispness of the attack are the signs of consonance at a larger span than the octave.

Much of that being decided by the piano inharmonicity, we have not as much control as we think.

What I am after now is how to use the provided leeway to allow definitive changes in modulations "color" . (color = consonance level, to me, I do not wish to go farther than that)

I will record that if you wish, (did retune with "standard tuning" and stack of M3) there is always some small imbalance in the reconciliation of slow and fast beating intervals, but if the goal is to have progressive M3 or 10th, it is in now way impossible.

I am far from sure this is what gives the most consonant piano, but it is possible to realize, and the more enlarged the 1st octave is, the easier it is .

BTW the first octave is tuned "enlarged (focusing on 4:2 6:3) because the 5ths partials are then more easily flowing together)

Due to the way I listen to intervals activity, I believe I can leave some "mistakes" as my beat rates are influenced by more than one partial match.

I am also not happy with a straightforward beat rate that flow straight and relatively staticly from the intervals. That gives a somewhat harsh sensation. Ideally, the beats may flow like the tone of an unison, something that is firmly set but is flowing somewhat freely.

THe most audible beat rate is generally influenced by the second set of partials, this gives the beat rate some sort of "respiration" .

When tuners say they need to have a very quiet ear to tune, that mean they can allow anything to flow in a natural way, out of the piano.
Make sense to me.

Excellent point, Kees.

I think we return to an important point here: there are a lot of tuners who overestimate their ability, or who no longer hear the imperfection inherent in their tuning. Imperfect scales mean tuning variations to find compromises over them.

The amount 4 notes would have to cumulatively vary is about 0.2c to upset the balance of CM3rds. However, if four notes cumulatively varied this much in one direction, it wouldn't upset progression. Localized variation is required and for a single note, this is about 0.76c.

This is underneath the PTG test limit for errors. Built into the PTG test is an understanding that up to 1.0c variations are required to find scale compromises in the temperament region.

Yes, if you (or anyone else) would be so kind to record progressive M3/6 in the temperament range I would be grateful. I will believe it when I see it, and I think I'm not the only one.

Thanks,
Kees

This thread is getting really interesting!

I recorded this, with a strip mute in. I was happy with the progression at the time, until I listened back to the recording. I know I can do better, but I'm emboldened to post it now

https://soundcloud.com/phil-dickson-1/piano-tuning-1920s-chappell

ugh... those 5ths are really uneven!

Hi Tunewerk,

Beyond "...compromises...", I look forward to listening to your pure 12ths ET tuning.




Hi Kees, I think you (and those you are hosting in your mind) may well start a new Topic, kind of "I will believe it when I see it", so that here you will make less noise for those who have already seen "it".




Hi Bill,

You wrote: ..."Exactly! As I point out in my own version of "Let the piano tell you", when you compare a series of 4 CM3's (not just three), if the top M3 (F4-A4) is too slow, then the bottom M3 (F3-A3) is too slow. You then lower both F3 & F4, then possibly also slightly adjust C#4 so that everything fits."...

IMO, the problem with that sequence is that three or 4 CM3's are not telling you how the M3's in between will be, those initial 4 M3's are too far away and anything can happen with the remaining M3's. It gets much easier when you can compare M3's that are closer, like A3-C#4, A#3-D4 and B3-D#4, and it is much easier (I would say) to evaluate the A3-E4 slow breathing (a very slow beat), or the A3-D4 1bps beat, than having to guess how those 4 CM3's compare to each other. Of course, this is only my opinion, which is based only on my own routines.

..."Works the same if the top M3 is too fast."...

IMO, it might work, but it is far more open to approximations.

..."When Jim Coleman Sr. was reviewing what I had written about that, I now recall that he said he had written virtually the same thin sometime in the 1980's. However, when you posted what he had written in another thread, that was the first time I had ever seen it. I certainly have not seen or read everything that Jim Coleman wrote but it does show that when something exists to be discovered, it will be. It was actually Viviano, someone who was trying to learn how to tune a temperament octave and who was trying to help me edit my material so that it would be clearly understandable who pointed out to me that what I was doing was treating the F3-F4 octave as a pair. If one F is moved, the other must be moved by the same amount. If that does not yield the correct result, then try again."...

I agree, good practice is to check any other related interval.

..."Usually, it only takes one, maybe two adjustments of both F's to find the correct balance but for a beginner, I can see how one may have to go back and forth a few times before getting it right."...

I would say that we need to be willing "to go back" as many times it is needed, and once we expand the first octave we get more and more checks, like 10ths, 12ths, 15ths and 17ths, which may suggest we... go back.

..."I would also like to say that there is no reason why a person cannot use a traditional 4ths & 5ths sequence that is practiced and familiar and then use CM3 tests to sort out ("nitpick") small errors."...

Yes, I am glad you are open to other methods.

..."In the example that Jeff posted, if he had listened to the chain of CM3's from F3-A4 after having arrived at the results in the usual and practiced way, the small error would probably have been evident."...

Hmmm... IMO, there were three*2 other issues there, perhaps method*knowledge, accuracy*skill and will*head.

..."In the end, all 4ths & 5ths need to sound very much a like, none too pure, none "beating", all M3's and M6's progressive and all CM3's also having the proper but small, slower/faster relationship. When you have all of that, you have what would be an indisputable ET. It can exist on any piano, regardless of scale."...

I agree on the last part of that sentence, "..It (edit: an indisputable ET) can exist on any piano, regardless of scale".

I do not agree with the first part, where you say "all 4ths & 5ths need to sound very much a like, none too pure, none "beating"... Let me know if you want to know more about how 4ths and 5ths need to sound, and it depends precisely on 4ths and 5ths, if they do not sound how they should, there is no way you can achieve "truly" and "smoothly" progressive RBI's (all across the keyboard).

Regards, a.c.
.

I can't download that file, just play it, so I can't analyze it.

Kees

[...off topic stuff cut...]
Hi Alfredo.

Kees

Hi dear,

Yes, cut stuff and play digital if you like, but... how about starting a new Topic where you can analyze ad libitum and keep together all the material you get?
.

I'm probably wasting my time by responding, but since we have been considering progressive M3/6 as a possible standard for ET, don't you think determining if this standard is practical by examining if anyone can actually tune to this level of accuracy is on-topic in this thread?

Kees

Kees, two things. That is the ET via Marpurg and from viewing Jason Kanter's graph, you would not expect perfect progressions, especially in the M6's. The main reason that video does not have better progressions than it does is that I forgot to execute the very last step: reconcile D#4 with the G#3 and A#3 below it.

It remains true, however that if the ET via Marpurg is perfectly executed, it will "pass" the PTG tuning exam. Several people I know of have done that. I am not allowed to talk about anyone's actual scores by name unless I have permission to do so but I do know of at least one person who passed the temperament portion of the exam with a perfect 100.

If I were to guess at what Jeff's temperament would have scored, it might be a 98 (only one "error" scored). But that is hypothetical because a Walter Console does not meet exam standards and there was no "Master Tuning" to compare it to. Nevertheless, a slight error such as is heard would probably score an electronic point.

The criterion for whether a scored point can be aurally verified is: Can the note which has a scored point be improved by either sharpening it or flattening it. If any aural checks (including CM3 checks) say that it can be, then the scored point is confirmed.

Let's all remember this: The tuning exam was set up so that only about 50% of tuners can pass it. After some years in use, the exam standards were tightened by adding multipliers to the pitch, temperament, midrange, unisons and stability portions of the exam. Today, it still remains true that only about half of first time attempts at the exam actually pass it.

This means that you don't have to be able to be perfect. No one has ever taken the exam and scored perfect 100's in all 8 categories. If I were to take the exam today or any day, I would not expect to get perfect scores in every category either.

So, one may hold up the PTG exam as being a model for ET and a standard to which one may aspire but the exam itself has tolerances that mean even if someone ever does get a perfect score in all 8 sections of the exam, the resultant tuning would still not be absolutely "perfect".

It is all just as hypothetical as BDB's nearly 20,000 posts are.

There is already a standard. It seems retarded to try to reinvent the wheel. Experienced tuners ALREADY know how to properly tune a piano. Just do it

All very true. Question remains if it is humanly possible to tune (aurally or with an ETD) fully progressive M3/6's or not, irrespective of the PTG exam standards.

It is a somewhat theoretical issue, at least to me, as I prefer an unequal temperament for the music I like anyways. But for ET aficionados it would be nice to know. After all theory and practice should agree in theory, and preferably also in practice.

Kees

OK, I figured out a way to get the audio off that URL you provided. Here are the results:


M3

C#3F 5.3
DF# 6.0
D#G 6.5
EG# 6.6
FA 6.9
F#A# 6.7
GB 7.3
G#C 8.0
AC# 8.6
A#D 9.5
BD# 9.1
CE 10.7
C#F 10.8
DF# 11.6
D#G 11.6
EG# 9.2
F4A 10.9

M6

C#3A# 6.9
DB 7.0
D#C 6.4
EC# 7.0
FD 8.3
F#D# 8.3
GE 8.6
G#F 8.9
AF# 11.1
A#G 10.2
BG# 9.0
C4A 11.7

m3

C#3E 7.6
DF 7.6
D#F# 9.6
EG 9.5
FG# 10.7
F#A 12.3
GA# 12.7
G#B 11.5
AC 12.5
A#C# 14.1
BD 14.4
CD# 13.9
C#E 16.4
DF 18.2
D#F# 17.7
EG 20.8
FG# 23.5

Kees

A rapid perusal of this leads me to believe there must be some really bad octaves.