[whafro]

Not to be a buzzkill – this is a cool story! But it’s worth noting that perfect/absolute pitch can be a negative for musicians in some contexts, especially vocal music.

Since even the vast majority of musicians employ relative pitch, entire choruses can move together off of the original key, for good reasons and bad, but those with perfect pitch will (sometimes stubbornly) maintain the original key, even when doing so is counterproductive to the performance.

Lead singer in the ensemble is a little under the weather and can’t hit the high notes? Normally, you’d consider starting the piece down a step or two, and get on with the show without much trouble. But if you have members with perfect pitch, that may not be an option without some significant rehearsal to familiarize them.

This also translates to musical appreciation – I know people who can’t stand when a singer covers a song in a modified key, saying it sounds “wrong” and “terrible” compared to the original. For the vast majority of the audience, the key doesn’t matter terribly much, but for those with perfect pitch, the key is a significant attribute of the original piece, and it’s just as major as changing the words might be.

In other contexts, perfect pitch can be very handy, but it’s not always quite as “perfect” as it’s sometimes portrayed.

[pclmulqdq]

I used to be a harpsichord tuner, and I can tell you that perfect pitch would have driven me nuts dealing with A being 415, 430, or 440 (+/- 2) Hz on any given day, as well as dealing with unequal temperaments. I have very good relative pitch in comparison to the average musician, and that is a lot more useful (and an entirely learnable skill as an adult). I know a lot of musicians with perfect pitch, but only one piano tuner.

My sister has perfect pitch, and she definitely had a leg up learning music, but she can't stand baroque music played in authentic pitch/tuning. Some modern music also uses effects to raise and lower the pitch of the song, and those annoy her too: think about the Janet Jackson song that breaks hard drives - it is in E, but the tuning is almost A=450 thanks to the use of varispeed. That one is pretty far, but many other songs have A=435-445 thanks to post-production.

Her orchestra plays at A=441, and I think she has basically learned that tuning or doesn't care - it's only about 5 cents sharp (1/20th of a half step).

[lloeki]

I was about to chime in about A=432Hz and stuff!

Not to mention training for absolute pitch on the chromatic scale has a heavy bias towards typical western music.

[smeyer]

>I used to be a harpsichord tuner

I have to ask out of curiosity, does this mean you worked full-time tuning harpsichords or rather that you did a lot of e.g. piano tuning and also occasionally tuned harpsichords? I'm hoping the former but expecting the latter.

[pclmulqdq]

I was a student at the time, and harpsichord tuning was a side job (~10 hours/week). My big "competitive advantage" over piano tuners was that I was very much into playing baroque music and knew a lot about unequal temperaments and harpsichord technology. I could also do pianos (tuning only, no maintenance), and did a few when needed, but harpsichords need tuning once a week plus an extra tune before every concert. For comparison, most pianos tend to get tuned on a several-monthly cadence, so you need a lot more clients to fill up a schedule.

Essentially, instead of a fee for service (like piano tuning) product, harpsichord tuning is a subscription product. However, I think there were <10 other people who tuned harpsichords in the same major metro area, so the market is pretty tiny.

[smeyer]

This is all fascinating! Thank you for sharing! If I can keep bothering you with questions (or if you happen to be able to point me to a place to learn more about harpsichord tuning), I have more questions. But I also understand if you don't want to keep answering them! Despite being someone who doesn't know much about nor listen to much music, I've always had a soft spot for harpsichords.

Why do harpsichords need to be tuned so much more frequently? How long did they take to tune (and how does that compare to tuning a piano)? How was the pricing structured with such a regular need? Were most of the harpsichords you were tuning in academic institutions, arts institutions, private use, et cetera? And not a question but another thought, I'm surprised there were even a handful of people tuning them in your metro area!

[pclmulqdq]

A sibling comment mentioned that harpsichords tend to have wood frames, and so they are incredibly reactive to temperature and humidity. The pin blocks in harpsichords also are often a single piece of hardwood, while in a piano they are a laminate of specially selected quartersawn hardwoods, so they hold the tuning pins a lot more strongly. If you hire a piano tuner for your harpsichord, they also tend to torque the pins too hard, which weakens the pin block even more.

Depending on the instrument and temperament, a tuning could be ~45 min or up to 90 min. Small instruments in the family (spinets and virginals) could have <4 octaves and one stop, meaning <50 strings to tune, and the biggest instrument had 5 stops and a 5-octave keyboard, meaning more strings than a piano (300 vs about 230). The "standard" instrument is ~4.5 octaves with 3 stops, meaning ~150 strings.

The customer picks the temperament generally, and that has some effect on how long it takes. Quarter-comma tunings (4 fifths flat by a quarter of a comma, the rest remaining pure) like Werckmeister are the quickest, and took under an hour on the standard instrument, but tunings like Kellner (1/5th comma, but harder from an A reference) and Valloti (1/6th comma) took me at least 2 passes to touch up, so over an hour. I also did equal temperament tunings with a tuner, which are quick. For reference, it takes me about 90 minutes to 2 hours to do a piano, so I am a little slow by professional piano tuner standards, but harpsichords are definitely quicker.

I also frequently adjusted the tuning based on what repertoire was being played, rotating it so that the near-pure thirds would be in the keys of the repertoire and possibly raising leading tones a bit. It was also not uncommon to have a modern woodwind in an ensemble, which often meant only doing a slightly spicy version of equal temperament rather than using a full-on baroque tuning.

The harpsichords were pretty much all at schools - I started by tuning my school's instruments and expanded from there.

I generally charged my hourly rate for 4 services/instrument/month + some padding, with extras (concerts) going at an hourly rate. This was a "work study" arrangement at my school (although I had 15 hours/week of work study for <3 hours of work, and still gave them a discount at the standard student rate) and contract work outside.

[analog31]

There's a harpsichord in my family. The traditional harpsichord had a wooden frame, thus the materials just weren't stable.

An American maker, John Challis, developed a harpsichord using modern materials, that stays in tune for much longer.

They didn't take as long to tune, as mentioned above, because the historic temperaments were easier if you knew what you were doing, and there were no unisons (multiple strings per note) to get into agreement. Before the age of the modern piano, keyboard players had to tune their own instruments, so it was just part of learning to play.

[pclmulqdq]

The Challis harpsichords are fascinating, but I think also a bit of an acquired taste. The aluminum parts sound odd to me, but I assume they basically never lose their tuning.

[plaguuuuuu]

I can finally ask someone this! I was thinking of getting my piano tuned to a spicier temperament, happy for certain keys to be noticeably out as long as other keys sound a bit sweeter.

Anything you'd recommend, or is it a shit idea on piano?

[pclmulqdq]

I would suggest Kellner or Valloti if you tend to play romantic period music, although those are a hair harder to tune than the easy ones like Werckmeister. Pianos can definitely take temperaments as long as you keep the A at 440 (to equalize tension on the harp).

Personally, I think Beethoven and Chopin sound great in Kellner, and I had my piano in Kellner for a very long time.

Definitely do it at least once.

[rerdavies]

Which leads to exactly the question I was curious about. Is absolute pitch good enough to tune a piano by ear?

[pclmulqdq]

I don't think so. Equal tempered intervals have a very distinct sound, and nailing that just from absolute pitch seems hard.

[tkgally]

When I was a teenager, I took piano lessons from a teacher with perfect pitch. When I told her that our piano at home—an old American-made upright—couldn’t be tuned any higher than about A=435, she visibly shivered. She said she couldn’t stand playing pianos that were not tuned to A=440.

Ever since, I have never wished that I had perfect pitch.

[musicale]

Some hardware and software instruments can easily switch from modern A440 to A400 or any desired A frequency, and from equal temperament to various other temperaments - even dynamic tuning. They are really fun to play, particularly the dynamic tunings where you can get highly in-tune chords and scales in any key. Vocal ensembles can also dynamically adjust their tuning, producing amazing overtones. I imagine string and brass ensembles can as well.

Some (non equal-tempered) keyboard instruments have split sharp/flat keys so you can play more in tune in certain keys/scales.

Being able to distinguish these all of these tunings accurately and switch between them as needed seems like it would be a nice skill to have.

[tkgally]

That sounds great. I would like to try playing a keyboard with dynamic tuning sometime.

I should have mentioned that I was a teenager in the early 1970s, and the only pianos I had access to then had strings and could not be retuned on the fly. In retrospect, I came to feel sorry for my teacher, as being conditioned to the arbitrary standard of A=440 prevented her from enjoying playing music on many of the pianos she would have happened to come across.

[cascades42]

Wow this is so interesting! So you are saying that the person with absolute pitch often has lost the ability to intuitively follow relative pitch, such that they are having to transpose in their heads?

I had always assumed they could still intuitively match pitch and just had an extra information overlay.

Do these people you know who dislike transposed covers also dislike genres of music with dissonant elements, such as certain types of jazz or microtonal music?

[gnulinux]

It's not losing relative pitch at all[0], it's actually kind of the opposite. Relative pitch and absolute pitch are at odds with each other in some contexts. There are many reasons as to why, and if you search tuning theory [1] you can find some amount of technical information. In this post I'll only cover a tiny portion of the reason, there are many other reasons, but this is one fundamental reason why.

To give a basic gist, two of the most fundamental intervals in music are octave (2:1 frequency ratio) which is 1200 cents, and perfect fifth (3:2 frequency ratio) which is about 702 cents. You'll find that if you stack 12 of these perfect fifths you come back to the same note (seven octaves up) but 23.46 cents off. 23.46 cents off is very much audible by every human being who is not speech impaired, so it'll sound extremely jarring (dissonant). This makes musical composition within the tradition of Western art/church music challenging. So, to fix this, we use 700 cents as the interval of approximate perfect fifth and each semitones apart by 100 cents (so that perfect fifth is 7th note and octave the 12th). We call this system "12 tone equal temperement" which is standard in all genres of Western music (from classical to jazz to pop to rock... but other cultures have many other systems). Now your piano will be tuned to these notes (0, 100, 200, 300... cents) such that it's impossible to play other notes. When people learn absolute pitch, they learn these notes are C, C#, D, D# etc. But when an instrument with continuous pitch plays (such as violin, cello, human voice etc) you do not need to be bound by this tempering. So you can actually play a perfect fifth as 702 cents. As long as the piece is not so chromatic/atonal such that you need 12 perfect fifths to add up to seven octaves, it'll work out. But when someone with perfect pitch listens to this effect, it can feel jarring, particularly because music is "out of tune". This can make piano music feel "out of tune" for people who are used to just intonation (e.g. violinists) and violin music feel "out of tune" for people who are used to 12TET (e.g. pianists with perfect pitch).

[0] Note that relative pitch is required to understand spoken human language, so as long as you don't have a speech impediment, you can likely understand relative pitch just fine. Of course, ear training can help you label the intervals you hear and associate them with names, not something all laymen can do.

[1] https://en.wikipedia.org/wiki/Musical_tuning

[pclmulqdq]

Most professional violinists play in perfect equal temperament. I never got that deep into string playing, but I assume that a lot of study of "intonation" is actually about unlearning the natural frequency ratios (3/2 for 5ths, 5/4 for 3rds, etc) and learning to use the equal tempered counterparts (2^(7/12) and 2^(1/3) respectively).

However, there are a lot of times when you can make music more interesting and exciting by adding some pure thirds (equal temperament is off by the most on thirds, and thirds are very harmonically important) at strategic places. You just can't do this on a keyboard instrument.

[plaguuuuuu]

I think I find it a bit dull when violinists stick purely to ET. It sounds a bit less lyrical.

Otherwise the hard part must be choosing just the right pitch vs the other instruments.

For cadenzas or solo, do what you want I guess

[gnulinux]

I think this very much depends on the context, and being a good virtuoso violinist (or cellist, or singer etc) picking the right temperament for the right effect. If you're playing in an orchestra with many other instruments, you likely have to stick with 12TET. If you're playing a violin concerto cadenza, if you're playing a piece for solo violin, you likely want to play in just intonation as much as possible. If you're playing a piece for accompanied solo violin (violin + accompanying piano or orchestra i.e. sonata or concerto) then it very much depends on the moment and what sounds good for the music. Especially for an instrument like violin, which is extremely sensitive to every tiny expression performer can add, it's hard to make blanket generalizations. Ultimately, it's all about the artistic style of the performer, and composer's vision.

[musicale]

> You just can't do this on a keyboard instrument

Excepting split sharp/flat keys (as seen in some non-equal-tempered harpsichords and organs), or some electronic instruments/plugins which can dynamically vary the pitch of each note.

[KnobbleMcKnees]

It gets worse than that. It can drift over time so that even if you're in the right key, you end up as much as a half step out of tune.

https://m.youtube.com/watch?v=QRaACa1Mrd4

[musicfan1]

The voice is the only instrument where you can "play" (sing) what note you think. So, the advantage of absolute pitch for improvisation is minimised.

However, AP would be advantage in vocal sight reading. With AP, you will never sing a wrong note, whereas a non-AP could make some mistakes, depending on how strong there musical ability/relative pitch is.

I am not sure shifting keys in singing would be hard for people with AP. Of course they would be aware of the exact new notes they would be singing whereas nonAP would simply thing "everything is X steps up/down" but relatively the same.

I don't think AP has an effect on tolerance to listening to a song in a different key, it is more your personal taste. I don't have AP, but I can tell when a song is in a different key from the original. I find it acceptable, so long as it is in tune. But I vastly prefer the original key simply due to familiarity. Also, some songs really do sound better in certain keys than others.

[nardi]

> The voice is the only instrument where you can "play" (sing) what note you think.

This is not even close to being true.

[kelnos]

How so? I can definitely sing random pitches that don't conform to, say, 12-tone equal temperament. If I sing into a guitar tuner, I can make the meter move continuously from its flattest to sharpest position.

[PainShake]

Their point is that there are dozens of instruments that also fit that criterion; i.e. the human voice is not the only instrument with that capacity.

Fretless sting instruments allow for continuous pitch modulation - that includes the violin family as well as others like pedal steel guitars, the Japanese shamisen, etc. Certain wind instruments like slide trombones and slide whistles do the same. There are also electronic instruments like the theremin or any synthesizer with a pitch bend wheel.

[cordellwren]

Audiation is a basic skill for any instrument. Accomplished musicians are trivially able to improvise complex melodies that they can play and sing simultaneously.

[rerdavies]

I'm a jazz guitarist who has developed the very bad habit of singing what I play while I improvise. Working hard to break it. But what I sing, and what I play is the same.

Being able to "think it, play it" is absolutely central to what I do when I improvise.

[someweirdperson]

> The voice is the only instrument where you can "play" (sing) what note you think.

More often than not it is:

The voice is the only instrument where you think you "play" (sing) what note you are thinking of.

With other instruments it is much easier to notice to be wrong.

[appletrotter]

It sounds like people who have absolute pitch are a little worse at relative pitch than normal?

[coliveira]

Relative and absolute pitch are orthogonal. A person with absolute pitch needs to practice relative pitch. So, this is not necessarily the case. However, people with absolute pitch have far easier time reading notation, which may result in less familiarity with relative pitch.