Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum Interesting comparison of 2 ETD's.............

Grandpianoman,

When you say that the "damper pedal is up" or "the damper pedal is raised", do you mean that the sustain pedal is not engaged or depressed and the dampers are therefore muting the strings?

Jun

Hi Jun,

Sorry about that.....When I said "damper pedal is up/raised" I meant that the dampers are up...all of them raised. I will go back and correct that.

Peter, thanks for that tip, and for your help! Will give it a try when I have some time. Using the RCT, I always tune in "Fine" mode, and I can usually stop the spinner, except for a few notes that never seem to want to stop the spinner. Another factor here is that my tuning pins are a bit high in the pin block, so they flagpole somewhat when I tune. Not being a pro, it's more difficult for me to set them. I notice that after I tune, in the 5th-6th octaves and a little of the 7th, they tend to go a bit flat fairly quickly. You can hear it more in the RCT tuning than in the Stopper. Even with this issue, I still hear more of that "sustaining" quality in the RCT tuning.

Hi GP,

A fascinating comparison! I hope you don't mind a non-tuner and rather amateurish player commenting in the hallowed technicians' halls?

Although listening only through smallish speakers on my work PC, it seems to me as though the Stopper tuning produces a clearer sound, especially in the treble, e.g. C6 to F4 at 00:11 to 00:13. The tenor (octave 3), e.g. in the opening notes, sounds warmer [edit: not really warmer, difficult to say - perhaps more extrovert] and clearer to me than in the RCT tuning. The latter strikes me as bland, almost muddy in places, while the former makes the piano sing and sparkle.

But, on the other hand, I perceive a difference in the way that bass notes blend in with their harmonics. To me, this becomes particularly clear in the often-played tonic of the piece, i.e. the way that F2 blends with F3 and its other harmonics - even more so where F2 is played sufficiently loudly to bring out its partials. Some examples: 00:34 and 00:43 (also to a lesser extent in 00:29-30). I am no professional, but my ear hears much more beating in the Stopper tuning than in the RCT tuning - in fact, my first reaction to the Stopper F2 was, "is that unison in tune?"

I'll also listen to the recordings using earphones, but my preliminary conclusion is that it appears as though a clear, singing tone comes at the expense of more beating between (partials of) bass and treble notes?

It's just an opinion (and some conjecture to boot), but there you have it.

A lovely piece, either way!

Regards,
Mark

Grandpianoman, Peter's point about changing the spinner speed is very helpful for improving your tuning skills. The default setting of 2 cents I found is also related to the amount and duration that the spinner blushes, indicating a stable, on target setting. Setting the speed to the highest discrepancy introduces a new hurdle to overcome. At that speed you will often encounter strings, especially on lesser quality pianos, that waiver or more often roll sharp on attack, stabilize for a second, and then slowly roll away flat. When RCT is set to record mode (in the initial sampling or when you click on the ear in CyberEar), I think it picks a small segment of the sustain/decay to come up with its analysis. For this reason it is important to develop a tuning technique that produces the same amplitude (mf) and experiment with different mike locations, and then gauge your results with the variance numbers. With practice you can get the variance numbers down to less than .1 which is well under the "good" description. If you rely heavily on the ETD and have not developed good aural skills it is worthwhile to check the 5-6 Sample A strings with CyberEar first to eliminate any unstable ones from your original sampling. Often the tuning graph will bear this out with a more even curve for intervals than if you happen to sample a rogue string. When cleaning up unisons by ear on these not so stable strings I am baffled that I have found no consistent visual pattern in respect to the phase to compare to on the spinner. Sometimes they clean up closer to the attack and other times closer to the end of the decay. I guess that this is because of the somewhat inconsistent nature in which the strings couple together. The ear still rules in this respect.

Hello again,

I've listened to the recordings through earphones and asked my wife to play them at random, so that I can test objectively whether I can hear any difference.

I strongly advise those who say that they recognise each tuning to do the same, and post the results here!

After this exercise, I'm no longer convinced that I can really hear a difference (other than mechanical noises that differ from one recording to the other). We ran a total of six tests:

1) RCT played, heard (wrongly) as Stopper.
2) Same as 1.
3) Stopper played, first identified as RCT, but unsure, then decided on Stopper.
4) Same as 3.
5) RCT played and identified correctly.
6) Stopper played and identified correctly.

Not a very convincing tally! More tests may follow. But I'm not sure whether I'm beginning to rely on things like single notes or mechanical noises that differ between the two recordings.

My earlier posting on blending harmonies does not seem very solid any more.

Regards,
Mark

Interesting results, Mark. A colleague of mine sent me recordings about a year ago where he played various pieces in ET and others in some kind of very mild well-temperament. Surely, I thought, I would be able to identify each correctly in each case. WRONG! The music was all very complex, mind you but the person who sent me the CDs said that I was wrong more than half of the time! Yet, I know very clearly when I hear my own style of tuning.

I wonder if Grandpianoman would like to use the figures for the EBVT III that are on record to tune his player piano and offer yet again some comparisons? I already know that it would not be exactly the way I would tune the piano but I also know that many people would do it the way he would: enter the "correction figures" for the EBVT III and tune the piano with the RCT default settings. I, for one, would be interested in the results. Grandpianoman could post the same pieces played by the player piano in three different versions: RCT default ET, Stopper tuning and the EBVT III with RCT default settings.

I am sure that I would get both positive and negative reactions but it is all good. I expect that. I would just like to hear the results.

Grandpianoman, you would need to consult your RCT manual on how to do this. If you need for me to provide the exact correction figures, please ask either on here or by private post. They are available on my website at www.billbremmer.com but I would be glad to send them to you.

Hi Bill,

Thanks for the offer...you know I thought about this awhile back, but did not follow through. I think it's a great idea and I would be happy to do it. I was just thinking it might be fun to throw in another ETD in the mix, my original ETD that I bought 4+ years ago, that started me on this journey. It's not as sophisticated as the RCT or the Stopper, as it has only pre-set stretches, but it served me well. It would be interesting to see now, how it fares against the other 2, plus your EBVT III. I won't be able to do this project probably until next month, as I am knee deep in work for the next month. So let's shoot for mid December! Bill, please pm me the correct figures for your EBVT III. Also it might be fun, once I get the 4 tunings recorded, to first do a "guess which ETD this is" test. What do you think?

It seems interesting.

But, instead of your amazing recordings of music, which sound great! I would prefer to hear some intervals. With an entire striped piano, to hear only one string of each unison.

Runs of 8ves, 5ths, 4ths, M3s, M6ths, M10ths, 12ths, double and triple octaves, will give us a much better idea of what is achieved by each ETD in terms of stretch and quality of the intervals calculated by RCT and Onlypure software for the same piano.

Don't you think?

BTW, how do you get such a professional quality in your recordings?

I could certainly include that test Rafael.

My setup is really very simple. There is no processing or enhancement of the sound, except in Audacity, where I amplify the original recording to bring the levels up a bit. I use the Samson Zoom H4 and 2 Rode NT5 mics connected to the H4's built-in pre-amps, with 2 boom mic stands, lid raised. You can see a picture of the mic set-up on my YouTube site, plus a few more recordings etc. http://www.youtube.com/user/AmpicoGPM

I think the digital recorders of today are leaps ahead of what we had before. Of course, the M&Hamlin BB is what it is, and with the 2 modifications, the sound is quite something now. I really did not expect this good of a sound from this setup, so I was very surprised they have turned out so well.

Another recording that might be good to include in this test: "Il Postino" which is not on my YouTube page. It has a magical quality to it, lot's of pedal. I have it here only in the RCT tuning: http://www.box.net/shared/fav9kacmpv

I saw your pictures.

Il Postino: Great music!

GPM:

Let give you another option for comparisons. I can construct and publish beat rate curves for a tuning given the iH and either cents deviation or frequencies. This would show objectively what octave types are tuned where.

I already have an iH curve for a BB, but if you have the iH numbers for your piano, those could be used instead. Up to you, it’s just an offer.

I listened to the two recordings, but didn’t want to give what are really just impressions. Knowing which tuning is which could easily influence my opinion. I don’t have time to set up a blind test. But my probably biased opinion is that the RCT bass is better (being more harmonious) and the Stopper treble is better (being more melodious.) And in between, I don’t know. They are just different.

Hi Jeff,

Sounds great. As I get closer to doing this in December I will PM you for instructions etc.

I can see where knowing which ETD is which could influence opinions....that's another good reason to perhaps first present it as a question to try and guess which ETD is which.

Your comment on the bass of the 2 tunings is interesting. I remember when I was re-tuning to the Stopper in the bass from the RCT, as I reached the last 6-7 bass notes, I noticed there was more of difference in those notes compared to the RCT. The rest of the Stopper tuning for the whole piano did not deviate as much from the RCT tuning.

Here is another selection that might be good to include for hearing the bass integration between the different ETD's. This is only in the RCT version: "Theme from ET" http://www.box.net/shared/gb6hc6pt4u I recorded this selection with a different mic position. They were right above where the music desk would be, (it was removed) and the mics were in the X-Y crossed position. You can hear more action noise etc, but it gives a different perspective.

Thanks for your comment Gardzar.

Hi, you mention using a very simple ETD with pre-determined stretches and no ability to sample notes. I've seen several piano tuning software with those pre-determined stretches (ex. Grand Piano 6 to 7 ft, Vertical Piano 39 to 51 inches, etc). I've never tuned using pre-defined tunings. How would you describe these pre-defined tunings compared to those by ETD's that allow you to sample notes? Just wondering...

Grandpianoman, I am so glad you are willing to tune the piano in the EBVT III! I will send you the specs as soon as I can get around to it. The kind of comparison you want to do, a "blind" comparison is a very interesting idea! It has been proposed many times over the years but rarely anything close has ever happened. The last time I remember the EBVT being compared to other temperaments including ET tuned by Virgil Smith was in 1998 and the EBVT (original version) won! (On a 4:1 vote of raised hands). And that was with an audience of piano technicians at a PTG convention!

Another PTG colleague of mine for whom I have great respect and who is a university technician and who has another favorite mild Victorian style temperament, has often commented that when he presented ET and a Victorian temperament side by side on similar pianos, he would consistently get a favorable reaction to the Victorian temperament.

Now, let me say first, that I don't expect you to like or prefer the EBVT III to the way you customarily tune. Obviously, you are a fine pianist and have also learned to tune very well with an ETD. For many technicians, the slightest departure from ET simply sounds "out of tune". I know that and accept it. But if my customers who are not tuners consistently had the same reaction, I would have abandoned non-equal temperament long ago. But the fact is, I have not tuned in ET now for more than 20 years.

I like Jeff's idea too, that only intervals and such be played in various ETD programs but I also know that it would require you to tune the piano again in all of the various ways to produce files. This should come later and should be done but only when you can find the time for it.

There are other points to consider about such trials. Some people resent being presented with a tuning in a different temperament or style and not being told about it. They feel that there is only one kind of tuning and if "something else" is provided, it is an act of deception.

A comparison as you have suggested can serve to cut through all of that. If you label the files simply as #1, #2, #3, etc., and most people guess incorrectly, it will really reveal personal bias against what really is appealing.

If you set it up so that listeners are to try to guess which is which and which they prefer, I am sure that you will receive an interesting mix of results. People will have subjective descriptions and preferences but will more than likely guess incorrectly more than half of the time. That is my hypothesis, anyway.

It will really be interesting to see where the original ETD you used falls into the mix since it was so thoroughly condemned by those who read a description of its premise. I am among those who were highly doubtful that it could produce an acceptable sounding tuning. Wouldn't it be interesting if it, in fact, got a substantial number of comments about "warmth", "soul", "musicality", "wood vs. plastic", etc.? I am not suggesting that I expect that but the results of a totally "blind" comparison will surely be interesting from an audience of mostly piano technicians and others who are at least audiophiles and/or pianists.

Another point to consider is that the way a piano is tuned is expected to influence the way a pianist may play it. I have witnessed many times over how a high caliber artist has reacted to a piano tuned in a temperament far stronger in inequality than the EBVT III. The piano was otherwise very expertly regulated and voiced. When the artist sat down to play, they heard "other voices" from the piano they had never encountered before and instantly adapted to them. The piano "spoke" to these artists as no other piano had ever before and they went with the phenomenon and played a unique recital that was never heard before and would likely never be heard again.

In the case of these trials, you will have recorded the piece as you normally play it but the differences in tuning will manifest themselves regardless of your technique. It will not, therefore, be a totally valid comparison but a valuable one nevertheless.

I recall, for example, a German pianist who was asked to play a Beethoven sonata tuned in the EBVT III at the PTG convention in 2006. When asked about his reaction to it, he expressed in his imperfect version of English, "When I tried to make the expression, I suddenly realized the expression was already there for me". That statement revealed a lot to me but clearly, when one kind of pianistic technique is applied to several different styles of tuning, the whole concept is invalid (out the window).

Nevertheless, an experiment such has been proposed will be interesting for its results. After all of the guesses have been received and compiled, a list of the recordings and which tuning styles they were actually in with comments invited should be posted. Will it change anyone's mind? Will it open anyone's mind? We will all have to wait to see. If nothing more, I hope it will instill in people's minds to "never say never".

Its interesting how these various tunings from different ETD's are being commented on from the recordings of musical pieces. Would the comments be any different (from the techs at least) if a series of various interval progressions were recorded comparatively similar to the aural tests we commonly use to tune? It might shed some light on more than this ETD vs that, and answer the question why the end result of one is more pleasing or has different subtle characteristics than the other when used to make music. On a different lighter note, I had this strange dream last night where I was 30 years older and was tuning a piano note by note playing a little Bach, then some jazz, then some pop, no interval checks though....weird.

Emmery, I agree that an interval study would have value and I think it should be done but remember that Grandpianoman would have to tune that piano again with all of the devices. As technicians, we do a lot of interval and other technical study. That leads to the condition we sometimes get into where we no longer hear music, we only hear beats!

What I find interesting is that even among many piano technicians, people will say upon hearing a piano tuned in the EBVT III, "I can't tell the difference between that and ET." When I went down to the Fazioli dealer in Chicago a few years ago and tuned two fine pianos in the EBVT, all entirely by ear, that is exactly what he said. He was rather disappointed. He had been expecting something startling (at least, startling to him) and he didn't hear any difference at all.

From the recordings on my website (none of which are really that good but it is all I have that I can use on there), Jeff quickly proclaimed that he would never tune that way. I wonder if Jeff and many others would actually correctly identify which tuning is the EBVT III?

There should be 2 questions and grandpianoman can compile the results: Identify the tuning and then tell which you liked best and why? I'd bet that not one person identifies all of them correctly. Wouldn't it be something if I misidentified the EBVT? I wouldn't rule it out.

While I think I probably would identify the EBVT correctly, I know that when presented with another very mild ET (Moore, I think), I was wrong more than half the time. For me, there was not a lot of difference between the two. When the music gets complex, shades of temperament aren't heard so clearly as they are when you're playing intervals as a technician would. A local colleague has salon concerts at his dealership on a regular basis and has high caliber artists perform. He uses the 1/7 comma meantone (a far stronger temperament than the EBVT). Never once has an artist ever said anything about the temperament or tuning but they often did say something positive about the piano.

I am also interested in hearing the piano tuned by the original ETD that grandpianoman used. It was condemned by several people. It works on an entirely different concept than the other devices. But would I or many others really be able to pick it out of the crowd? Would people who previously said they preferred the Stopper or the RCT now misidentify them and say the same things they did before but about the wrong piano? Would people think the EBVT is ET and the ET is the EBVT? I surely think some will.

It will all be very interesting and fascinating.

We'd all like to attend a live event like this but that would be nearly impossible. On here, grandpianoman's great piano and great recording devices can give as a virtual alternative.

Bill, you might have missed the part about grandpianoman's set-up being a player-system? For just the reasons you describe, I've been thinking about the yamaha players at the school to use for tuning testing playback... It takes player perfomance out of the mix and just lets the tuning speak for itself.

The beauty of a recorded test is that the same piano is used, hopefully with the same equipment set up to record. By using a player system, the artist interacting with the tuning is taken out of the picture. (though that is a HUGE part of why some people prefer alternate temperament, it leads them to other interpretations than they normally would chose.)

I actually prefer to avoid "tuning" test intervals during the listening phase. I do prefer some very simple pieces, as well as those that span the range of the keyboard to get a feel for the musicality of a tuning - and how well it fits the particular instrument.

Ron Koval
chicagoland

Hi Bill,

I was just going to post that my recordings are not me playing. I wish I could play that well! The "LX" system is what is at work here, as well as the "Ampico" system. www.live-performance.com All the recordings I have posted here were played by professional pianists, Earl Wild, Brian Pezzone, etc.

Ranger, the ETD I first used and posted about here awhile back, was made by a retired engineer. He only made a small number, and is not making them any more due to lack of parts. He is however, experimenting with a new version of this tuner!......if I hear anyting about it, I will post it here. He has promised me a trial run with it if it works! A lot of info about my original ETD is in this post: http://www.pianoworld.com/forum/ubbthreads.php/topics/642321/1.html


The LX playback system does indeed take the variable human interpretation out of the mix, as it plays everything the same way, all the time. Whatever interpretation that pianist played the day of the recording, that is what you hear every time the file is played back. The same is true for the Ampico rolls.

This is turning into quite a project, and I am looking forward to doing it! Before I start it, I would like to get your opinions on how to set it up, procedures etc, and how to present it here on PW.

I admit that my opinion was not based on a side by side comparison. But if few can tell the difference then what use is it anyway? I will say that I found a sample of a Vallotti temperament on the web and felt that my sensibilities were being assaulted. With the EBVT III I was expecting to hear something that I would recognise as cyclic. (Fourth and Fifth tuning is cyclic by nature.) What I experienced was something different. So, in my own mind I was evaluating objectively. I did not hear what I expected.

But really, this started out as a comparison on ETDs. If GPM chooses to expand this into comparisons of temperaments, fine. But lets not pressure him.

All:

Even listening to beats, which is a more objective way to evaluate the stretch than listening to music, is still a subjective test. I mean, how can you decide just where the 4:1 octave turns from wide to narrow, or if one tuning had faster beating 6:3 octaves in the tenor than another?

That is why I am suggesting doing a mathematical analysis of the tunings to determine the beat rates. Then the beat rates can be used to calculate the stretch. The beat rate curve of each octave type and the twelfths can be compared between different tunings. All that is needed is the iH curve (which can be constructed from samples, as long as the same construction is used for comparison) and either the frequencies of the partials or their cents deviation.

But of course, that does not mean that anyone would be able to tell the difference between one tuning and another. But in cases of a proven preference, it may give an indication what it is that is being preferred.