Pianotech

Expand all | Collapse all

The Piano Sound Is A Moving Target. YouTube Video Presentation

  • 1.  The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-20-2024 22:01

    Click here -> Video Link on YouTube: The Piano Sound Is A Moving Target

    The Piano Sound Is A Moving Target

    This video presentation comes as the result of original research, with results that have never been prior shown in the technical literature. I believe it to be a major breakthrough to the understanding of the nature of the piano sound. The work presented was peer reviewed and vetted carefully. It will evolve into a future journal publication. The video is more than just theory, it presents sound demonstrations to validate the points. I will be at the upcoming convention in Reno for further discussion opportunities.

      

    Statement of the Problem :

    •       The piano sound is never in a steady state of settled frequency, hence, a 'moving target.'

    •       The cleanliness of sound is most perceived just after the attack in real piano music.

    •       There is no such thing as an instantaneous measure of frequency; it occurs over a 'window of time.'

    •       Ultimately, the question: "When, in time, do we set the tuning, and on what basis?"

    •       Each string and its partials generally has a unique frequency decay characteristic.

    •       When unisons coupled, the Weinrich Drift effect can be profound, at least 1 ½ cents, with a significant delay after the attack. Each coupled unison will have its own 'signature' and spreading effect.

    •       'Unintended consequences' as 'damaged inharmonicity' from slight spreading and not tuning in the first initial attack period.

    •       There is a lack of objective data in the literature showing the actual frequency changes over time.

    •       Therefore, the details of this phenomenon are not universally understood, nor agreed upon.

    •       We are left with many subjective methods dealing with it, many diametrically opposed.

    •       Each preferred method will produce a different outcome.

    Solution:

    •       Make 'new technology' to show in real time how the partials of a given string are changing frequency over time, giving the piano technician a 'target in time' that is consistent and repeatable.

    •       Make this a more objective practice, raising the bar to a higher level in the piano tuning industry.

    •       Sound demonstrations are presented to validate the points.



    ------------------------------
    Steven Norsworthy
    PianoSens.com
    steven@rf2bits.com
    Cardiff By The Sea, CA
    (619) 964-0101
    ------------------------------



  • 2.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-20-2024 22:21
    Thank you for making this video.  It is a very clear explanation of the most accurate piano tuning method available today.  Some might disagree, to each his own.  The presentation is easy to follow and understand.  
    Sent from my iPhone





  • 3.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-21-2024 08:01

    Steve,

    This issue has been debated ad infinitum over the decades. AFAIK you are the first to actually produce actual real analytical data to show graphically what many of us have simply theorized about (but could not prove). 

    What I see in this demonstration now is precisely WHY slight variations in unison tuning contributes to the often unique "signature" sound that any one of us seems to be able to impart to a piano due to our unison tuning habits. 

    It is VERY interesting to actually SEE some of what happens (at least in the case of your 308) with the interaction (or reaction) of partials upon combining with one another and being coupled with the bridge/soundboard vs. vibrating singularly. And as you say, each note has its own sonic signature based on how it gets tuned because of its coupled interaction. 

    Regardless of how one feels about the sound samples (obviously the first unison is far more pure than the second) this simply shows that there are various (actually numerous) ways of  slightly "coloring" the sound in the instrument, and why each of Faziolis' tuners make the same instrument sound different even though they all tune ET. 

    Bravo!  I think Harold Conklin and Gabriel Weinrich (and others) would have gone gaga over this had it it been available to them!

    Peter Grey Piano Doctor 



    ------------------------------
    Peter Grey
    Stratham NH
    (603) 686-2395
    pianodoctor57@gmail.com
    ------------------------------



  • 4.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-21-2024 10:52

    You have to watch this several times. It's the best explanation of why we need to tune unisons as perfectly as possible- how it affects the tone. The second example in the video has a sound that would make me think the string wasn't seated properly or the hammer needs to be filed. But that's not it - it's the response of the partials behaving from the result of a less than perfect unison. The kind of unison you would get when tuning by ear.

    Randy Woltz, RPT



    ------------------------------
    Randall Woltz RPT
    San Juan Capistrano CA
    (714) 731-7479
    ------------------------------



  • 5.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-21-2024 11:53

    In the collective minds of a given discipline there are two levels of thought – knowing and understanding. Steve, your observations are commendable. Now let's take it to the next level. What causes these gyrations in harmonics of such. I believe the unmentioned elephant in the room is the nature of string termination and internal string tensions. I will bring to the upcoming convention a demonstration that internal string tension may be the source of some of these gyrations. See me at the Moondog exhibit booth.



    ------------------------------
    Roger Gable RPT
    Gable Piano
    Everett WA
    (425) 252-5000
    ------------------------------



  • 6.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-21-2024 12:24
    The Article by Gabriel Weinreich in Scientific American Vol 240 No 1, Jan 1979, pp. 118-127 "The Coupled Motion of Piano Strings" is relevant covering "antisymmetric" motion of vibrations and coupling of harmonics even to result in an accent in the middle of a note.

    Best wishes

    David P 

    --
    - - - - - - - - - - - - - - - - - - - - - - - -
    David Pinnegar, B.Sc., A.R.C.S.
    - - - - - - - - - - - - - - - - - - - - - - - -
    +44 1342 850594





  • 7.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-21-2024 13:55

    Roger,

    I appreciate your affirmation of my analysis. It required a significant effort to develop the signal processing software needed to extract this information. I look forward to meeting you in Reno. I will be at the PianoSens booth and will have my computer where we can conduct live experiments and extract data off any piano there.

    The data was taken on my Fazioli F308. Fazioli's are noted for their clean sound and few false beats. People who have heard my piano have remarked how clear it is, some who have opined already on this thread like Randall Woltz and Paul McCloud who live locally to me and played the instrument. Do you have any experimental data on Fazioli's method of string terminations and pin angles on the bridge that you can share? The first sound file in the recording certainly sounds super clean, and Randall Woltz commented that it ruled out a string termination issue! I made sure with a little tool that the strings were seated well. Have you developed special new IP on string terminations and improvements that you have shared and that you can point us to? 

    Best,

    Steve



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 8.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-22-2024 05:08

    All,

    Thanks for your replies. Just a heads up. There are others with vested interests in this subject. I am getting some private replies as to why this has not been shown in the past, casting doubts as to imply I may not have done a good enough prior survey. Here is my answer.

    A good researcher does two things: (1) Surveys the prior publications as best as possible, and (2) has highly credentialed peers privately provide feedback before publishing. I did both. Certainly we have 'spectral analysis,' a very big field in engineering and nearly all fields of science, to examine. Looking at very fine grain spectral movements in time, a 'spectrogram', on piano partials is a wide open opportunity that I seized upon. Most researchers are working on grants that have financial backing. I don't! It took hundreds of hours just to develop the signal processing software to even analyze this phenemona, much less then collecting the files off the piano countless times and weeks of experimentation. I have to acknowledge a 'silent partner', ex-MIT professor now retired, whom I have known since my Bell Labs days in the 80's, as he has been doing spectral analysis at the world's highest level for his entire career. He is also, like me, a pianist! We had a passion to want to do this. Neither of us could find prior research that shows what is in this video. I do believe that I will get some naysayers, it always happens in research, but I DO really appreciate all your great comments. 

    Kind and thankful regards,

    Steve N.



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 9.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-26-2024 12:25

    Hello Roger:

    i have read your short comments with interest.  I have long paid close attention to bridge terminations in my rebuilding and know how important those terminations are in getting a good pure tone.  I also have an ongoing interest in the stresses and strains that exist in pinned bridges and their relationship to the soundboard and their affect on tone.  I have done some research into alternate pinned designs with tonal characteristics that are somewhat different and promising.

    You speak of the effect of string tensions that exist within the string, and their effect on the harmonic (partials) structure.  You call them gyrations.  Why the choice of that particular word?  Can you elaborate on what you mean?  Also elaborate on internal string tensions.  That could mean a lot of different things. 

    Steven, forgive me if I am being a bit obtuse.  What do you mean by IP?



    ------------------------------
    William Truitt RPT
    Bridgewater NH
    (603) 744-2277
    ------------------------------



  • 10.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-26-2024 12:45
    Hi William!

    Whilst not being a fan of micro exact tuning nor of on-the-strike measurement as demonstrated here https://www.youtube.com/watch?v=Z_e94lRIrfo gyrations will occur as the strike vibration is horizontal and very efficient at transferring energy to the bridge whilst sustain vibrations will arise from horizontal motion given by all sorts of interactions with and through the bridge.

    In his paper "The coupled motion of piano strings" https://www.jstor.org/stable/24965072 Weinrich writes:
    "When two strings are in tune, their
    motion can always be expressed as
    the superposition of symmetric and an
    tisymmetric modes. When two strings
    are slightly mistuned, their motion can
    still be expressed as the superposition of
    two modes: an almost antisymmetric
    mode whose damping is small, although
    not quite zero, and an almost symmetric
    mode whose damping is large, although
    not twice as large as the single-string
    rate. In both modes the amplitudes of
    the two strings are equal. If a hammer
    strikes the strings at the same time and
    with the same strength, the exactly sym
    metric motion the hammer excites is
    not a normal mode; rather, it must
    be viewed as a superposition consisting
    mostly of the rapidly decaying (almost
    symmetric) mode but containing an ad
    mixture of the long-lived (almost anti
    symmetric) mode as well. The amount
    of this admixture depends on how differ
    ent these normal modes are from per
    fect symmetry and perfect antisymme
    try, which depends in turn on the extent
    of the mistuning.
    The major difference between the
    contribution to the aftersound due to the
    mistuning and the contributions due to
    the horizontal polarization and the anti
    symmetric motion is that a skilled piano
    tuner can adjust the former but not the
    latter. I think this explains seemingly
    random variations in unison tuning that
    were observed by Roger E. Kirk of the
    D. H. Baldwin Company. A skilled pi
    ano tuner varies the mistuning in such a
    way as to make the aftersound uniform
    and smooth from note to note by com
    pensating for the irregular effects of the 
    horizontal polarization and the antisym
    metric motion. In this way the piano at
    tains its characteristic beauty of tone
    that less skilled piano tuners are unable
    to induce. To test my hypothesis I would
    have one tuner tune the same piano a
    number of times, with someone else de
    tuning it in between. If the same "ran
    dom" mistunings manifested them
    selves each time, it would prove that the
    mistunings were not random at all."

    In my posting https://my.ptg.org/discussion/yamaha-c31859-broadwood-concert-grand-1885-bechstein-model-iii-1960s-steinway-b-and-tolerances-of-unisons I raise issues of the exactitude to which we should be tuning unisons, not as a matter of theoretical piano perfection, but for the purpose of creating an alive sound for the music and the musician.

    Weinrich in the passage quoted above gives priority to the aftersound rather than the strike sound "A skilled piano tuner varies the mistuning in such a way as to make the aftersound uniform and smooth from note to note by compensating for the irregular effects of the horizontal polarization and the antisymmetric motion. In this way the piano attains its characteristic beauty of tone
    that less skilled piano tuners are unable to induce." and it's in that beauty of tone that I'm aiming for in tuning for the sustain and audible in the example of the Yamaha CG1 demonstrated in the video above.

    If anyone listens to the process I've documented choosing the sustain rather than the strike and feels that they'd do it differently I'm very happy to learn from criticism.

    Best wishes

    David P

    --
    - - - - - - - - - - - - - - - - - - - - - - - -
    David Pinnegar, B.Sc., A.R.C.S.
    - - - - - - - - - - - - - - - - - - - - - - - -
    +44 1342 850594





  • 11.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-26-2024 15:20

    I am really interested in meeting Roger at the convention. I published a video (link below) months ago and it spurred quite a bit of controversy and discussion. I stand behind it. It shows there are two sources of false beats: One being Acoustic and the other being Mechanical (probably in the bridge termination). I provide a very useful 'tool' for distinguishing between the two sources. Many have confirmed the usefulness and are getting better more accurate tuning resolution from this. I'll demonstrate this live for you at the convention, but the video shows the essence of the issue and the solution to distinguishing the difference.

    Link to YouTube Video Below.

    https://youtu.be/e_UvfoUevy8

    Steve



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 12.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-26-2024 17:55

    Steve,

    Sorry for the delayed response, I'm dealing with the annual summer STEM student workshop which requires most of my non-piano work.

    No, I don't have access to a Fazioli for experimentation and no bridge "data" I can share at this time.  My comments were directed to the question of string termination as to the possible reason for the varying harmonic movements during the few microseconds after attack as illustrated in your graph.

     I believe we think differently, as I approach a problem with the goal of developing a way to correct these anomalies. So far what I see with your latest research is data only. What do you plan to do with the information?

    William,

    My research strongly suggests – I might even say demonstrates – that the nature of the termination plays a significant role in the tonal envelope and false beat character of a vibrating string.  For example, a string that is terminated on both ends via a 360-degree clamping (such as drilling a hole the same diameter in a piece of steel as the string) will exhibit a different tonal envelope than a string that is terminated in the traditional piano bridge design. Also, keep in mind that a piano string is terminated differently from its opposite ends – capo verses flat bridge.

    Lastly, I will demonstrate that the internal stresses in a wire greatly affect the tonal and false beat characteristics. This is evident by using the counter reactive wire bending tool sold by Supply 88 --the tool bends the wire around the bridge pin.

     See me in Reno.

    Roger



    ------------------------------
    Roger Gable RPT
    Gable Piano
    Everett WA
    (425) 252-5000
    ------------------------------



  • 13.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-26-2024 18:46

    Roger, my 'solution' is to own a Fazioli! Few if any false beats! I am an analyst and not in the business of doing mechanical after-market solutions. I posted the research tool that helps identify the source of the false beats, Mechanical vs Acoustical, and it is the best diagnostic tool I know of. You should try it. Thanks. Steve

    Link to YouTube Video Below.

    https://youtu.be/e_UvfoUevy8



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 14.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-26-2024 19:55

    I watched the video twice, thank you for putting it together. While I'm not going to object to the data shared in the video, I'm struggling with the two sound samples provided (C5).  While the second has a noticeable wobble compared with the first one as noted in the video, neither of them are satisfying my ear. The first example has a clean sounding fundamental (partial 1) but the 3rd partial seems to be beatng 4-5 times per second. When I tune unisons, I am primarily listening to the third partial of the note to judge when my unison is locked in. Does anyone else hear this?



    ------------------------------
    Tim Foster RPT
    New Oxford PA
    (470) 231-6074
    ------------------------------



  • 15.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-26-2024 20:02

    Tim, why don't you make a wave file recording of your own tuning examples, and let us hear them. Some of what you hear could be MP3 compression going on with your device and YouTube. That is a common problem with reconstruction of audio. When I hear the raw uncompressed wave files, they don't have what you describe. No social media server allows raw uncompressed wave files. They take up 10x more space. Your playback decompresser makes it worse yet. Audio compression does funny things to the spectrum. If someone wants me to send them my raw uncompressed wave files, I can do that. Just send an email at steven@rf2bits.com. Tim, I'll put your recordings through the same partial spectral analysis and let's see what is going on. Certainly every piano is going to be different. 

    The whole point of the video is to finally show the graphs that clearly indicate that the partials are changing over time. Weinrich Drift is going on and it is different for each partial. Each partial has its own rate of frequency changes. That is part of what makes a piano sound like a piano, not a steady state synthesizer. That was the whole point. And mind you, this comes off a $300,000 'world's best' concert grand Fazioli F308 that was hand selected for the Franz Liszt International Piano Competition.

    As I argue in the video, do you want the attack phase in the first second to have unwaivering clarity and power, or do you want to wait 3 seconds later and tune when the power is down 100x (40 decibels) when no piano literature is listening to a single note in isolation with no pedal and no accompaniment notes? That is purely academic. Go play the Chopin Op. 25 No 7 Étude in C# Minor and the opening line in the left hand by itself uses pedal to blend that single melodic line, then all bets are off trying to figure out what any of those notes sound like counting 3rd partial beats 3 seconds later in isolation with no pedal. But guaranteed you will hear the cleanliness of the attack in the first second. This is a major problem with the mindset of the tuning community. They don't understand this, and the video points it out clearly. Go to my YouTube channel where this video is posted and listen to my other files where I play intervals after I tuned the Fazioli this way. They are very clean and clear. Then make some recordings like that and post them for us to all hear. Let's compare.

    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 16.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-27-2024 07:43

    Thank you, Steven, David, and Roger for each of your responses.  I don't have time to answer all of your excellent remarks, but I will make a stab here. 

    Steven, I am very much aware of your Pianosens, and believe that you are making a significant contribution to the state of the art.  I must confess that I like it when someone comes in and upsets the apple cart of conventional wisdom.  For persons who practice intellectual honesty, you have made them put on their thinking caps, even when they disagree with you as to its value.  All for the good.

    Your Fazioli is a remarkable instrument, I do not think you will find any argument here.  Its clarity is of real value both for your research and demonstrations, and we learn from it.

    To me, it is both an asset and a distraction in that it represents the zenith of piano making to a degree that very few instruments can achieve.  The reality of our existence is that we work in a world that is much less tidy, and we must muddle our way through it with in an ever-changing effort to do our best work with what we encounter.  The argument lies in whether tuning the unison to an absolute fidelity to pitch that you advocate yields the best tonal result.     

    Your argument for absolute pitch fidelity is very narrowly targeted, and I must confess that I champ at the restraint of the bit.  Elsewhere, Roger has used the term "tonal envelope".  That speaks to the totality of a sounding note from onset to end, and everything that comprises it.  That is a whole lot of things, and it is fantastically interesting to me, if one is interested in making the best possible sounding piano. 

    Roger has said that "the nature of the termination plays a significant role in the tonal envelope."  I could not agree more with that statement.  My own experiments tell me that is so, for better and worse.  😊  What is important?  Everything. 

    Roger talks about the 360 degree termination clamping at both ends in steel, I have envisioned something like that in a proper tonewood.  (I have not built a model yet).  Roger has hinted that, in ideal terms, the boundary conditions should be identical for each end of the speaking length of the string.  To the degree that we can achieve that through redesign, that would maintain the purity of the originating signal.  The hammer strikes the string and sets it into motion and it organizes its harmonic structure.  Everything that string encounters adds its own artifacts to the signal – the wood surface on the contacting surface at the pin and on top of the bridge – the species of wood, its density, hardness, resistance to crushing, speed of sound.  All these things have their effects.  Where there are pins, the material used, spacing front to back, side to side spacing, diameter, and the notching protocol.  Add to this downbearing and side bearing, it is really complex for something so simple.  With the aid of my friends Tom Thiel and Doug Kirkwood, we designed and built a working model of a pinned design that has 3 pins per string.  We subjected about a dozen people to listening tests which confirmed what our ears were hearing.  The attack was shorter and not as loud, and very quickly resolved into the harmonic structure and remained there thru the duration.  The pitch drifted than would be typical of a two-pin design.  It had a stronger sense of enduring order.  It lacked the droop in pitch that follows the attack, which is so characteristic of the two-pin design.

    The two pins per string configuration is a dinosaur.  For all its simplicity, it brings its own set of problems, Fazioli notwithstanding.  Done right, we can hear great beauty from it. 

    We haven't even started talking about bridge agraffes….



    ------------------------------
    William Truitt RPT
    Bridgewater NH
    (603) 744-2277
    ------------------------------



  • 17.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-27-2024 10:33

    Steven,

    I don't know why I'm hearing the G6 (coincident partial) beating with the C5 unison locked in as it is. If not an audio problem, perhaps an aliquot or higher open string with no damper sympathetically resonating and slightly out of tune, at least at the coincident partial? This is not necessarily a critique of your content or tuning, I'm just trying to figure out what I'm hearing. 

    I'll try to get a Wav recording later of C5 on my Yamaha C5. 

    One question, and sorry if I've missed this in other content of yours: it seems that most of your testing has taken place on your 308. Have you tested many other instruments and if so, how have the results compared?

    Thank you!



    ------------------------------
    Tim Foster RPT
    New Oxford PA
    (470) 231-6074
    ------------------------------



  • 18.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-27-2024 13:41

    Here is C5 on my Yamaha C5:

    https://drive.google.com/file/d/1m76bgbme-IPaxqyeWI8RD6CYjtgTrDgq/view?usp=drivesdk



    ------------------------------
    Tim Foster RPT
    New Oxford PA
    (470) 231-6074
    ------------------------------



  • 19.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-27-2024 13:57

    Anthony Willey posted a video on his Pianometer YouTube site refuting many of my findings. Here is the trouble with his refutes:

    1) An apples-to-apples comparison should use the same register area, C5 in my video, and not A2 in his video. The lower register is a longer wavelength and more stable. Anthony used the note A2 in the mic variation. The lower the note, the longer the wavelength and the less senstivive to mic placement. Every user of ETD's knows this, how jittery the readings are in the mid- to high register with a mic as compared to the sensor. He should have used C5 or higher.

    2) The tuning point in time I explained was more based on the point of reference from POWER and MUSIC. Where is the power the highest? In the first 1 sec after the attack. What does the piano literature do with duration of a single note in isolation 3 seconds later after the attack with no pedal and no other notes? That does not exist in the piano literature. I gave all my reasons both technically and musically. Go back and listen to my explanation. He missed my points.

    3) My graphs are all taken from the sensor, which has no hammer noise on the strike, of course. He is capturing the mechanical hammer noise. I am not.

    I fully stand behind everything in my video.

    Steve



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 20.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-27-2024 14:06

    Similarly, I have to question the objectivity of the presentation considering that the chosen note in no way is representing the scale location of the note in Steve's video. This is bothersome to me. It was the first thing that struck me as I was watching the video, considering that the stated intention was to "reproduce" the experiment. 

    Needless to say that using mic vs a sensor is another anomaly. If you want to reproduce something you do it the same way. AFTER doing it identically, THEN do it your way. Then we can all see the true differences. 

    Not being critical...just observational. 

    Peter Grey Piano Doctor 



    ------------------------------
    Peter Grey
    Stratham NH
    (603) 686-2395
    pianodoctor57@gmail.com
    ------------------------------



  • 21.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-28-2024 11:03
    Part 2 Video: The Piano Sound Is A Moving Targetk
    Link Here:
    Here is the first review on YouTube from Rick Clark:
    Rick Clark
    Great work! My whole life I have tuned on the sustain tail, whether by ear or ETD, because that was really the only way we could tune. I look forward to using a new method snap-shotting the pitch a fraction of a second in as suggested by your work and hearing the aesthetic result as music is played. The idea of tuning to a better standard than has ever been possible before in history is very exciting.



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 22.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-28-2024 12:40

    Another comment from the new Part 2 Video, from tuner Richard Adkins:

    Thank you. This is what we tuners need to see. The late Ron Nossaman was one of the tuners who recommended "tuning in the attack". You may find some discussion in the PTG groups.
    I tried to follow his advice, but I still listen to sustain, too!
    Your research and videos confirm what Ron advocated and which I try to do. 
    I appreciate you sharing all this, which you could have just kept to yourself. 
    You are helping good tuners become better tuners.
    Richard



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 23.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-27-2024 15:35
    Another reality is that the tuning of C5 is governed by C4 and its relationship with the C4 harmonics. When we're tuning C5 its harmonics at C6 really don't matter greatly as from C6 to C7 we're getting into the range where the ear wants the notes to be higher than the harmonics and if we tune them to the harmonics we perceive the notes to be flat, and above that we're looking at an octave which is simply tuned percussion giving fizz and sparkle to what's below. If I tune the top two octaves harmonically then it sounds flat in performance. Indeed it's so frustrating in tuning the top octave that one takes the note up through the resonance of the 2nd partial of the octave below and has to knock it higher above resonance for it not to sound flat.

    The choice of C5 to exemplify why we need to buy a special device to tune it, and measure it to tune on the strike, indicates a technical mindset of great acuity but without the experience that with which the musician tuner will recognise as symptomatic of an invention of a problem that doesn't exist.

    Tuning Fazioli is one thing but Bechsteins play around with us with false beats, false notes and I believe deliberately so in order to give the sound life. When we tune one string by itself it's haywire but the three strings together pull their vibrations together to give a note. Again Faziolis might be well behaved but the Duplex scale on Steinways is rarely on pitch in my experience so there are false harmonics all over the place. From the infamous Wigmore Hall Bechstein with the logarithmic rather than stepped bridge, Bechsteins were deliberately experimenting with putting partials all over the place and their approach to partials was key to their tone which musicians regarded as special. 

    Today I was tuning for a pianist who actually _likes_ a meeeow to her unisons rather than sterile straight ones. As technicians we might like and relish the technical perfection of clinical purity to the sound, but not all musicians consider it to be music.

    The bottom line is whether we're tuning an instrument for the instrument to be art in itself or whether we're tuning the instrument to be able to express music as music, and music as a manifestation of emotion. 

    Best wishes

    David P

    --
    - - - - - - - - - - - - - - - - - - - - - - - -
    David Pinnegar, B.Sc., A.R.C.S.
    - - - - - - - - - - - - - - - - - - - - - - - -
    +44 1342 850594





  • 24.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-28-2024 16:51

    Steven N,

    Please don't understand my questions to be antagonistic, I assure you they are not. If your assertions about your device are correct, the accuracy in which this would allow pianos to be tuned would be unprecedented in the entirety of the piano's history. With anything new, there can be unintended consequences, both favorable and unfavorable. I'm also wanting to hear from a variety of musicians since this discussion has predominantly been among technicians who do not necessarily represent the desires of musicians. Some questions:

    1. Much of the discussion and tuning examples seem to focus on volume potential. Are the partials behaving similarly on the attack of a MF, MP, P or PP dynamic to your F or FF example? To my ears, quiet playing seems to have a more stable pitch at the onset, but of course this could be inaccurate.
    2. Is the sensor able to read partials above the fundamental, or do we rely on microphones for that?
    3. How are we defining clarity of sound? To my ears, European pianos sound clearer to me. I loved playing jazz on the Bösendorfer at convention last year, I much preferred Chopin and Medtner on the American pianos (a wonderful Mason and Hamlin in particular). In other words, onset clarity wasn't always what I was after on a musical level, but I don't know if there is a fundamental difference in what we are calling clarity.
    4. Lastly, the subject of the experiments I've seen were isolated notes. We know that the vibrations of one string will excite other strings to sympathetically resonate when their dampers are lifted, either when other notes being played, by a depressed pedal, or a combination of both. Have you looked at the "moving target" when combinations of strings are simultaneously resonating? 

    I think a lot more studying and experimenting needs to be done before adopting your propositions are superior. Perhaps they are superior, but this should be proven with further experimentation and most importantly, its adoption by musicians who believe it to be a superior sound allowing better self expression on the instrument.



    ------------------------------
    Tim Foster RPT
    New Oxford PA
    (470) 231-6074
    ------------------------------



  • 25.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-28-2024 17:17

    Answers for Tim F.

    1) Volume of attack initially sharpens a few tenths of a cent for the first couple hundred milliseconds, then it settles. The piano wire is a 'spring under tension' and has nonlinearity when stuck hard. I advocate tuning at mp-mf range and so do all my other collaborators who have adopted this methodology. Statistically it is the 'average force' for music anyway.

    2) The sensor has over 20 kHz of frequency range, way more than the C8 (4300 Hz), and so it hears everything by 'reading string movement' instead of the bias that comes from variation on the soundboard (a mic pickup). We 'get to the source' without bias with a string sensor. It is 'deaf' acoustically. The soundboard colors things in thousands of ways of variation and the variations are significant to creating tuning bias. The sensor bypasses that whole problem. It faithfully reproduces the initial source of all sound, i.e., the string resonance standing waves.

    3) By 'clarity' I mean no extra coloration imposed by the coupling of the string onto the bridge and soundboard. The string vibration is source or 'pure essence' of everything acoustically that happens and that we hear.

    4) The sensor is not biased by hearing other strings, it has a limited focus area that is plus/minus one unison set to the left or right, and that is 'by design' so that we don't get interference when tuning. Mics cannot do that, and hence, impose interferences that prevent the best accuracy of the strings being tuned. I have not heard anyone say that they want 'other string interference!'

    5) If you want a private list of top tuners using it and validating it, PM me and I will ask some of them if they will receive your calls.

    6) I am always trying different experiments and discovering new things from it. Examples even last night from updated posts.

    7) You can have a jack hammer next to you and tune perfectly because it is 'deaf', a 'good thing'. So far I have not had one person say anything less than aural confirmations of what it reads. There is no downside except for an initial learning curve.

    Feel free to call. Thank.

    Steve N.



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 26.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-28-2024 22:07

    So, there's something I'm missing here.  The measurements made for this presentation are from a sensor picking up the vibrations of 3 strings in a unison simultaneously, if I heard correctly around 6:50.  The graphs show that the frequencies of different partials drift over time.  But tuning is a comparison of two or more frequencies.  It seems to me that the same partial on two or more strings could change together in the same pattern over time, which seems totally possible if all the physical characteristics of the strings and the terminations are the same.  So if a partial's frequency changes over time, but that partial on another string changes so that their frequencies are the same at any given moment, then they are in tune.  The fact that a partial can change over time, even if that change can be measured accurately, does not tell us if two strings are out of tune at any given time.  It seems to me that you would have to put a small sensor over each string and measure these changes over time for each string in order to compare them and make the claim that the drift in frequencies can cause the strings to be out of tune several seconds into the sustain, compared to the first fraction of a second.

    The graph displays the 4 different partials and how they deviate in different ways over time.  Steven says "the partials are fairly tight and close together."  But that is only true in the graph, which show cents deviation over time, not the actual frequency of the different strings.  The fact that the different partials of a single string drift over time would not cause a beat.  What determines whether a partial is in tune or not is whether the two or three strings have the same frequency or not.  This graph does not give us the information you would need to know to make this determination.

    Lastly, I'm just not getting the point of favoring the first 0.7 seconds for precise tuning.  It seems that something is missing from the analysis.  If the partials drift over time AND this can cause them to be in tune at one moment but out of tune at another, it seems you are allowing for the possibility that you could get the strings perfectly in tune shortly after attack, but then a long sustain will result in the unison being a bit out of tune.  If this doesn't happen, then what's the point?  If tuning using the first 0.7 seconds results in a note that is perfectly in tune for the duration of the sustain, then tuning at any time during the sustain will do the trick.  On the other hand, if it's perfect for the first 0.7 seconds, but that results in an imperfect unison later on, you have favored one part of the sound over the other, which is an artistic choice, not something that is necessarily determined using technical results.  It seems possible that you could test this:  get the first 0.7 sec perfect, then try to tell if there is any beat several seconds later, either by ear or with the recording system.






  • 27.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-29-2024 01:42

    As for the Y axis zero, it means that at the end of 500 msec, we have a reference of zero 'derivative'. What you should be understanding is that the Y axis is a derivative, or the change of the end point of the window with respect to the end point of the prior window. The end of the first window is 500 msec so there is the reference point for the next change, which is a window from 250-750 msec, so at the end of 750 msec we have a derivative or change from the end of the first window that is from 0-500 msec. 
    The sensor placement has virtually no variation of reading in frequency but in amplitude it of course varies because we are at different phase points of the partials.



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 28.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-29-2024 08:10

    Loren,

    The "coupling" or "entrainment" action of the bridge structure contains the unison once it is in tune (actually it has a progressive effect when tuning entirely aurally since once the strings get close enough [i.e. about .25 cent or so] the bridge eliminates the audible beat thereby making us believe we've reached zero when in fact we have not [but we're close]). So you're not going to get a "beat" in the unison several seconds after strike (unless of course it's a weird piano). 

    What I find very interesting in these tests and graphs is that it shows a significant difference in how the partial structure changes its behavior when coupled (entrained) to the bridge vs sounding singularly. Additionally it clearly points out WHY the sound of a note can be altered by either tuning dead on at zero or slightly spread. We have long known this to be a fact, but precisely why has been a bit of a mystery. Furthermore, both the degree and the pattern of spread changes the way the partials interact and behave within this coupling action of the bridge. 

    Tuning the unison as close to zero (as Steve N. AND MANY OTHERS) advocate (past and present) creates the best chance at keeping the unison clean for the longest period of time. This practice also (obviously) has a specific effect on the tone and (as shown) the volume of the initial strike. It also affects the decay envelope of the note in most cases, as does spreading the unison (every .1 cent added changes this decay envelope to some degree). Some of us refer to this as "voicing" the unison, which brings me to what I think is the greatest potential for this kind of research which is that it can help us SEE what our voicing techniques (with needles, etc) are doing (or can be done) to selectively change the partials behavior when the note is struck. If I had this kind of graphic information available to me on the fly I'd be very interested in seeing how I could voice the hammer in specific ways to achieve specific outcomes. At present, no ETD has the "guts" to display this information, but Steve's work shows that it can be done, and I personally find it very interesting and an eye opener. 

    The argument about tuning on the attack or somewhere past that point is not new (it goes back long before my nearly 50 years in the business) and remains a point of personal preference in musicality. The freeze feature/with sensor, coupled with an ETD that has the accuracy to use it clearly gives the tuner the best shot at hitting zero (if that is what is desired) on the unison. The sensor (without the freeze feature) produces an effect all it's own which I find quite useful as well. 

    Peter Grey Piano Doctor 



    ------------------------------
    Peter Grey
    Stratham NH
    (603) 686-2395
    pianodoctor57@gmail.com
    ------------------------------



  • 29.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 06-29-2024 14:36

    Steven N,

    Thank you for your explanations (post 25), that was very helpful. I think I understand the sensor a little better. A couple follow-up questions if I may.

    1. How are you certain the microphone discrepancies are due to soundboard distortion? 

    2. After the fundamental, all other partials have multiple places of origin (2nd partial front half and back half of the string for example). Does the sensor placement affect what it "hears"? For example, would we get the exact same reading with three sensors, one near the front, another near the back and another near the middle if they were all picking up the movement of the same string on the same strike?

    3. In relation to question #2 above, I would be especially interested if the placement of a sensor on the back end of the string has the same characteristics as the front since the front is the starting point of string disturbance (strike point). I would imagine that the string temporarily stretches around this point as it pulls the rest of the string into motion. (Of course, the fundamental itself would also be slightly higher because the string is being stretched the most at the beginning of the strike). If this is true, would it be possible that the front portion of the string stretches more than the back portion, making 2nd and higher partials  slightly sharper in the front of the string than the same partials on the back of the string near the time of the hammer strike?

     Thanks in advance for your input.



    ------------------------------
    Tim Foster RPT
    New Oxford PA
    (470) 231-6074
    ------------------------------



  • 30.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 06-29-2024 15:12

    Tim,

    There is significant research on the eigenmodes of the soundboard. The resonance points or eigenmodes 'tilt' the spectral peaks from location to location causing peak frequency detection errors of several cents worst case, and I have documented that experimentally. I have posted various videos of this analysis. That is why there are so many 'stories' I have heard of 'tuning examines' from PTG tests where one judge hears different 'frequency results' from another judge in another location spot. This is an issue the sensor 'fixes'. Remember, your ear is also an acoustic mic and your brain is the processor. 

    Each partial at any given timeframe has three variables: (1) frequency, (2) amplitude, and (3) phase. The sensor location on the wire does not affect frequency like the mic placement eigenmode resonance does (as per above). It affects the phase points of the partials, hence, the amplitude is net affected. That is not a problem for tuning purposes.

    Each 'string' is a 'spring under tension' and has nonlinearity when stretched beyond its linear region. That is why the initial attack from a hard blow will temporarily cause a small rise in pitch that lasts for a (at most) a few hundred milliseconds (as I have measured). After about 500 milliseconds, it settles down. I advocate tuning at mp-mf blow forces.

    The mic placement issue affecting 'frequency' is a big deal. Did you see my past posts on it?

    Best,

    Steve N.



    ------------------------------
    Steven Norsworthy
    PianoSens
    Cardiff By The Sea CA
    (619) 964-0101
    ------------------------------



  • 31.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Registered Piano Technician
    Posted 07-02-2024 22:38

    Thanks for the info.

    I was speaking with a friend today about tuning within the half second and this question came up. With all digital ETDs (that I know of at least), there is a small delay before the pitch goes on display. When milliseconds are of the essence and there is a 250 millisecond (or so) digital delay, how can we be sure that we are tuning within a tight enough window, unless using a stroboscopic tuner?



    ------------------------------
    Tim Foster RPT
    New Oxford PA
    (470) 231-6074
    ------------------------------



  • 32.  RE: The Piano Sound Is A Moving Target. YouTube Video Presentation

    Posted 07-02-2024 23:31

    Tim, 

    You are referring to the latency in the display of a tuning app. I understand that. Here is the solution. The Freeze Window concept that I was the first to disclose, and thereafter I first proposed that it be put into Pianoscope. The author of Pianoscope did his own implementation from my disclosure. It is the only app so far that has implemented the Freeze Window, as far as I know. Numerous videos on my YouTube site show the Freeze Widow of Pianoscope in action. It allows the user to exactly set the time window length from the hammer strike to the stop point and that is the window length and then the result of the calculations are displayed. 

    In my latest video, How Spatial Position Affects Piano Tuning, link below,

    https://www.youtube.com/watch?v=26zskT_GvkA&t=6s

    I show exactly how a spectrogram is made. You will not need to worry about any latency issues if the app uses and displays the Freeze Window, which is essentially the tuning result from first FFT block calculation. In this example it is from the point in time near the hammer strike to the end of the block time thereafter. The samples of sound that are captured during that block of time are used in the calculation. You then see the tuning result from that block. How far off it is from zero is then used to re-strike the hammer until iteratively the indicator shows nearly zero cents while of course the continuous indicator in the app keeps moving. The jittery spreading of the partials can be seen in the continuous indicator of an app, but most apps don't show us the details that I propose in the video. Jitter and spreading is 'real physics' going in in the piano sound, and of course the ear and brain hear it too, but can't measure it with the kind of precision that a machine can. 

    Steve N.



    ------------------------------
    Steven Norsworthy
    CEO/President
    RF2BITS, Inc.
    Cardiff CA
    619-964-0101
    steven@rf2bits.com
    ------------------------------