On the fly perfect chord tuning

[crapwonk]

Here is a link to an article on perhaps the ultimate in split key organs to let you choose exactly which interval you want:
36 Keys per Octave: The Arciorgano, https://www.voxhumanajournal.com/keller2018

Note that the two manuals shown in the picture are required to implement the 36 keys per octave. Interesting concept, but nothing I hanker after. I am quite sensitive to pitch changes, but am quite insensitive to the interval issues in the usual tuning.

[samibe]

I'm wondering if it hasn't been pursued because it is unnecessary to make a digital organ sound like a pipe organ. Almost no pipe organs are ever perfectly tuned just like pianos (even right after tuning) regardless of temperament. Also, I feel like mis-tuning in a digital is mainly to get an ensemble effect despite a lack of channels or speakers.

That being said, I think a digital organ with pure cord tuning in a church setting could make it a lot easier for the choir or congregation to sing to.

To me, beats on purpose (Celeste or trem) are fine. But I think beats just because that's how it's been done are unnecessary and probably worth eliminating especially if technology will allow it.

[jbird604]

This might well be a thing that will become common in digital organs, as one of the temperament choices on the menu. It won't work in all situations though. In a church where organ and piano are routinely played together, the piano is going to be tuned in equal temperament and thus might clash with the organ under certain conditions.

I think Hermode may well be a good thing when an organ is being used to play certain types of literature, particularly early Renaissance or pre-Renaissance pieces that may have been composed for pipes or other instruments with Just intonation. And it might be slightly beneficial when an organ accompanies singing, as singers tend to automatically adjust their intervals on the fly.

[samibe]

I think that especially for singers it would make it much easier to tell if you are singing in tune. I think it would eliminate the "Am I out of tune with the organ or the other members of the choir?" question because the organ would be matching what the choir could do.

I think my next thing to figure out is how to implement something like this with a VPO.

[voet]

I once professional soprano in my choir from China who was working on her MFA at Julliard. She had great difficulty singing Western music a capella. However, I went to her master's recital and she sang a number of Chinese folk songs based on scales with micortonal intervals. She had no problem with these pieces, but she found Western music very challenging. It is interesting how culturally conditioned this all is.

Another professional singer in a different choir always had a tendency to sing sharp. When we sang a capella she would throw the entire choir off pitch. I wondered if she was so fearful of singing flat that she went in the other direction. When she was gone, the pitch problems went away.

[rjsilva]

...Thus we have settled on the Equal Tempered Scale as championed by Bach (or so I've heard) because it allows one to play any piece of music in any key without being too offensive.

[samibe]

Well I tune a perfectly equal temperament every time.

I usually set the stretch with an octave, then set thirds between the octave (and one third below), then fill in the notes in between with fourths and fifths. I do check everything with thirds, fourths, fifths, and sixths but I definitely allow a bit of variability in consecutive beat rates. Occasionally, I have to adjust my temperament when I start working up the treble because some of the deviation becomes too wild for my taste.

To a certain extent, the piano and organ already clash because an organ's harmonic series lacks the piano's inharmonicity (and resulting stretch). That's assuming they were tuned to each and that the temperature and humidity hasn't changed significantly since tuning.

This thread has got me wondering if the pitch fixing that a lot of pop artists do in editing a studio album uses Hermode tuning or some other pure chord system (which is why they seem so fake sometimes). I've been noticing (while searching for pure chord tuning information) that there are several audio processing programs that seem to have Hermode pitch adjusting capabilities built in.

[rjsilva]

I usually set the stretch with an octave, then set thirds between the octave (and one third below), then fill in the notes in between with fourths and fifths. I do check everything with thirds, fourths, fifths, and sixths but I definitely allow a bit of variability in consecutive beat rates. Occasionally, I have to adjust my temperament when I start working up the treble because some of the deviation becomes too wild for my taste.

[samibe]

In researching this a bit more, I found a book and a couple articles by William A. Sethares fascinating. They are a bit mathy (which I appreciate) but they also seem very on the ball (or at least as good as you can get when dealing with human perception).
Parts of the book can be viewed here.
He has a website with links to his books and articles (1993 Article, 2002 Article, etc.).

As for how to implement this, I'm starting to figure out how it could be done on a high level, but where the rubber meets the road is a different animal altogether.
It looks like most synthesisers have an option to adjust the pitch of a midi channel via midi messages (usually to accommodate a pitch bend wheel).
The tricky part seems to be with dealing with midi channelization. It is easier to implement on a synthesizer because a synthesizer has one voice and only a couple channels worth of midi information. So, taking the unique notes played, assigning them to separate additional midi channels (in a C chord grouping all notes into C, E, and G note midi channels) with an appropriate pitch adjust midi signal, and running that through a series of synthesizer processors is fairly straightforward.
It gets much more convoluted with organs because the midi from an organ is already divided into multiple channels to capture all of the divisions, stops, expression information, etc. It's looking like I would have to take all of the note information on all of the separate division channels, run an analysis or pull the adjustment info from a table, then split each division midi channel into several more midi channels for each unique note on that division (up to 6 maybe), and apply an appropriate pitch adjust midi signal to each note/division channel. It's admittedly complicated but still doable, I think. I'm really just not sure how I would tackle it with the programs I have (jOrgan and fluidsynth).
It might be easier to implement a set of 12 (or so) just (or nearly just) temperaments (one for each key). Then a subroutine could check the played notes and pick the temperament that would fit it best (as well as an overall pitch adjustment so that the change isn't too crazy).

[voet]

This is an update regarding Hermode Tuning on Content organs. While that feature was not provisioned on my organ, I noticed that there was a reference to it in the manual. I contacted Content to see if it could be added to my organ and was informed that it cannot be added. It had to be requested when the organ was ordered. Also Content does not offer this feature any longer.

This appears to be a feature that was made available from a third party source. I suspect that there was not enough demand for it to justify offering it any longer.

[jbird604]

Too bad they dropped it. But perhaps the improvement is too subtle for most people to notice or care about.

But when Content was offering this feature, they had a quite convincing set of audio clips you could listen to. I know it sounded pretty good to me when I tried it, but it's surely a minor consideration these days, given the extensive tuning variations among samples in a typical modern digital organ.

[Havoc]

This can also be done (more or less) with microtonal tuning. Organs like this have been build (look for enharmonic organ) and they can/could be played where or you made de decision to play a c## or d with the microtonal keyboard or let the computer do it when you played a normal keyboard.

For early music, with a strong tonal center, it's doable...

Yes, but isn't this the music where you should absolutely not do it? Early music often used those out of tune sounding chords to set the mood, both by selecting the key and the chords.

About 8-10 years ago or so I had a new Johannus 397 organ that came factory equipped with 12 different tunings, most of which were quite nice to play around with.
The least liked I must confess was the Equal Temperament, the one most universally heard in North America all all music events, whether church or concert hall.
The Rembrandt also had a tuning (#13) that you could set up yourself. We toyed around with the so-called concept known as the "circle of fifth" and came up with a setting that had very nice sounding chords in almost all of the sharps and flat keys. We heard chord combinations never heard before, difficult to describe, but it seemed as if the music came alive.
Changing the half step transposer settings added another dimension to the overall projected sound and even more so when the pitch control was adjusted to get in between those half steps.
A major living accomodations downsizing made me sell it to a church and I bought a 2m Sweelinck with just Equal, Werkmeister and Meantone, also very useful. Our pitch is set at 425 and yes, every chord sounds different than what your ears are accustomed to than Equal at 440. Yes, I miss my Rembrandt custom tuning capability.

Just what I mean.