Melodeon.net Forums

Please login or register.

Login with username, password and session length
Advanced search  

News:

Welcome to the new melodeon.net forum

Pages: [1]   Go Down

Author Topic: Best unequal temperament for a two row melodeon?  (Read 2822 times)

0 Members and 1 Guest are viewing this topic.

Owen Woods

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3894
  • melodeonmusic.com
    • The website and blog of Owen Woods
Best unequal temperament for a two row melodeon?
« on: March 27, 2016, 06:09:28 PM »

I am in the midst of getting a new box - an A/D. I think I've decided to tune it other than 12TET but I can't quite decide what. It's basically quarter comma, sixth comma, vallotti or me try to work out my own temperament. Anybody have any thoughts?
Logged
Bergflodt D/G 4 voice, Saltarelle Bouebe D/G, Super Preciosa D/Em, Hohner Impiliput B/C+C#

Latest blog post: In Any Weather

http://melodeonmusic.com/blog

pgroff

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1827
Re: Best unequal temperament for a two row melodeon?
« Reply #1 on: March 27, 2016, 10:08:07 PM »

It depends on why you are going "off-road" (in the sense that ET is a well-frequented road - a general-purpose temperament used by so many musicians and instruments these days that solves so many potential problems of compatibility).  In my case, I like instruments tuned in other ways than ET because they can sound better, particularly when playing certain intervals and chords. But you're getting into a wild terrain with many compromises and challenges.

Will you ever play with other musicians and what instruments will the others play? Are you trying to optimize a solo sound? Do you require just major thirds? Can you tolerate very narrow, buzzy fifths and a degree of "sharpness or flatness" of some keys that will deviate a lot from 12ET instruments? Are you willing to include more than 12 different pitches in each octave, and to develop a complex technique to choose among those options when building harmonies? How many keys and modes will you use, how many accidentals (beyond the A and D diatonic scales), exactly which pairs, triads (or more) or notes will you sound together and in which bellows directions? Which pairs or groups of notes will you never play concurrently in the same bellows direction?

My point - if your goal is to optimize a solo sound for this instrument, with some compromise of compatibility with a range of other instruments,  you may be able to create a special tuning for an A/D box that provides many beautiful harmonies but that has better fifths than 1/4 comma meantone. Especially likely if you will not be playing in the keys of C, F, Bb, Eb, Ab on that instrument ( and perhaps not in C# or F# either); then you can force all the wolf intervals into certain intervals that would be mainly needed in those keys.


PG
« Last Edit: March 27, 2016, 10:27:37 PM by pgroff »
Logged

squeezy

  • Quick starter (now lagging behind)
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1317
  • Hohner Cornelia (mixte D/G/o&s)
    • www.johnspiers.co.uk
Re: Best unequal temperament for a two row melodeon?
« Reply #2 on: March 28, 2016, 01:16:50 AM »

Although intrigued by more perfect tuning scales and really appreciating what great sounds and harmonies they can bring to a piece of music - I'm increasingly coming to the conclusion that I don't think they have any place in the tuning for a 2+ row diatonic ... unless you are going for a fairly dry tuned solo instrument that will be predominantly played in one key against a drone or simple bass only left hand - it seems to create more problems than it solves to my ear - especially in the case of full chords on the left hand which either sound out of tune - or clash with the right hand.

You have to be sure of why you want it and how you're going to use it ... or you have to be prepared for it to be an expensive experiment that may yield no good results ... whichever way - all experimentation and exploration of what can be achieved on a box is valid and useful and I'd definitely like a go on it if you go ahead!
Logged
Squeezy

Sometimes wrong, sometimes right ... but always certain!

Steve_freereeder

  • Content Manager
  • Hero Member
  • ***
  • Offline Offline
  • Posts: 7511
  • MAD is inevitable. Keep Calm and Carry On
    • Lizzie Dripping
Re: Best unequal temperament for a two row melodeon?
« Reply #3 on: March 28, 2016, 07:17:20 AM »

I would echo what Paul and Squeezy have said.

You have to be sure of why you want it and how you're going to use it ... or you have to be prepared for it to be an expensive experiment that may yield no good results ... whichever way - all experimentation and exploration of what can be achieved on a box is valid and useful and I'd definitely like a go on it if you go ahead!

One way forward might be to get a Streb in which the individual notes can be tuned to whatever temperament you choose and changed as your experiments progress. Not sure this is possible with the current Streb design but it might be worth contacting the maker Steve Rouse. 

Once you are happy with a temperament on a Streb you could then copy that tuning to a conventional A/D box. It's perhaps a fiddly way to go about it but it could save you making expensive mistakes in the long run.
Logged
Steve
Sheffield, UK.
www.lizziedripping.org.uk

forrest

  • Other Places::
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1009
  • digits still functioning
Re: Best unequal temperament for a two row melodeon?
« Reply #4 on: March 28, 2016, 08:03:57 AM »

In my obscure corner of the universe, I have settled on a simple temperament of sorts that includes  flatted thirds and sharpened fourths, as most of my playing is either solo or with violin. Violinistas can invent their own temperament on the fly, and fiddlers like their intervals sweet.

I'm increasingly coming to the conclusion that I don't think they have any place in the tuning for a 2+ row diatonic ... unless you are going for a fairly dry tuned solo instrument that will be predominantly played in one key against a drone or simple bass only left hand - it seems to create more problems than it solves to my ear - especially in the case of full chords on the left hand which either sound out of tune - or clash with the right hand.


  It follows that whatever tuning or temperament is situated on the treble side, must be mirrored on the bass/chord side. It is possible sometimes to vary a treble or chord note to be slightly different on push than on pull, as needs be, as long as they match on left and right. In other words, there should be no 'clash'.
 
Logged

deltasalmon

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1002
Re: Best unequal temperament for a two row melodeon?
« Reply #5 on: March 28, 2016, 12:24:26 PM »

In other words, there should be no 'clash'.
 

That sounds like the second part from above would be true then. I would consider wetness to be a "clash" in itself. I'd think changing the temperament would have little effect unless the box was a dry 2-voice or single voice box.
Logged
Sean McGinnis
Bordentown City, NJ, USA

Theo

  • Administrator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 14475
  • Hohner Club Too
    • The Box Place
Re: Best unequal temperament for a two row melodeon?
« Reply #6 on: March 28, 2016, 02:27:24 PM »

And the "clash" of a major third in ET is pretty appalling, but most of us have grown accustomed to it and don't notice how awful it is.
Logged
Theo Gibb - Gateshead UK

Proprietor of The Box Place for melodeon and concertina sales and service.
See my YouTube channel  for videos on repair and tuning

IanD

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1703
  • Too many melodeons...
Re: Best unequal temperament for a two row melodeon?
« Reply #7 on: March 28, 2016, 02:52:01 PM »

And the "clash" of a major third in ET is pretty appalling, but most of us have grown accustomed to it and don't notice how awful it is.
Go and have a listen here to see what you've grown used to...

http://leware.net/temper/js/acc.htm
Logged
Oakwood Model 4 D/G, Castagnari Dony D/G/#, Castagnari Tommy G/C, Baffetti Binci D/G, Hohner Preciosa D/G, Melos Bb/Eb, Lightwave SL5 and Kala California fretless basses

Owen Woods

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3894
  • melodeonmusic.com
    • The website and blog of Owen Woods
Re: Best unequal temperament for a two row melodeon?
« Reply #8 on: March 28, 2016, 05:54:40 PM »

It depends on why you are going "off-road" (in the sense that ET is a well-frequented road - a general-purpose temperament used by so many musicians and instruments these days that solves so many potential problems of compatibility).  In my case, I like instruments tuned in other ways than ET because they can sound better, particularly when playing certain intervals and chords. But you're getting into a wild terrain with many compromises and challenges.

Will you ever play with other musicians and what instruments will the others play? Are you trying to optimize a solo sound? Do you require just major thirds? Can you tolerate very narrow, buzzy fifths and a degree of "sharpness or flatness" of some keys that will deviate a lot from 12ET instruments? Are you willing to include more than 12 different pitches in each octave, and to develop a complex technique to choose among those options when building harmonies? How many keys and modes will you use, how many accidentals (beyond the A and D diatonic scales), exactly which pairs, triads (or more) or notes will you sound together and in which bellows directions? Which pairs or groups of notes will you never play concurrently in the same bellows direction?

My point - if your goal is to optimize a solo sound for this instrument, with some compromise of compatibility with a range of other instruments,  you may be able to create a special tuning for an A/D box that provides many beautiful harmonies but that has better fifths than 1/4 comma meantone. Especially likely if you will not be playing in the keys of C, F, Bb, Eb, Ab on that instrument ( and perhaps not in C# or F# either); then you can force all the wolf intervals into certain intervals that would be mainly needed in those keys.


PG

This box is for me to play on my own, at home as my "go to box" or as a solo instrument. It won't be for playing with others, in any circumstances really. It will be an A/D, which means that I will be playing in A, D and E majors, with E dorian, B dorian, aeolian and harmonic minor) and F# aeolian. Accidentals not in those keys probably limited to A# (for playing in B harmonic minor) and F natural (although it wouldn't be used often if at all) and If I want to stray out of those keys then I have other boxes available to do that. I don't want it to be so wonky that people who aren't used to listening to unequal temperaments will be put off. I am not willing to have more than 12 pitches per octave, nor to have more than one value of a particular note (my brain only stretches so far and this will only be a two row after all). I will play lots of different intervals within those tonalities, so any wolf intervals need to be a long way away. The box will be a Hohner and therefore tuned to a medium swing. This means that the effect of the temperament is likely to be subtle but hopefully still noticeable (comments welcome on that point).

Although I probably have the knowledge necessary to create my own temperament, I am not sure that I have the willpower! The current favourite is Werkmeister III which narrows some fifths by a quarter of a comma, resulting in some sweeter thirds whilst retaining some more or less pure fifths. I might end up rejigging it somewhat to give purer fifths though.
Logged
Bergflodt D/G 4 voice, Saltarelle Bouebe D/G, Super Preciosa D/Em, Hohner Impiliput B/C+C#

Latest blog post: In Any Weather

http://melodeonmusic.com/blog

forrest

  • Other Places::
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1009
  • digits still functioning
Re: Best unequal temperament for a two row melodeon?
« Reply #9 on: March 28, 2016, 06:15:27 PM »

In other words, there should be no 'clash'.
 

That sounds like the second part from above would be true then. I would consider wetness to be a "clash" in itself. I'd think changing the temperament would have little effect unless the box was a dry 2-voice or single voice box.

If the chords and bass are tuned such that they are centered between the + and minus of the treble, then it it does not "clash". You are correct though, that wet tuning is really nothing more than a controlled dissonance, and "clash" is a subjective judgement. Some of the Italian tunings are more than I can handle.

And the "clash" of a major third in ET is pretty appalling, but most of us have grown accustomed to it and don't notice how awful it is.
Go and have a listen here to see what you've grown used to...

http://leware.net/temper/js/acc.htm


Thanks for posting that link, Ian. It is a great tool for experimenting with the different established temperaments, and hearing their results as they are moved around the circle of fifths, and also building one's own temperaments. When testing, be sure to index the scale to  the pitch of the instrument, and then you can make sure that the 'wolf tones' wind up in territory that is obscure to those keys.
Logged

IanD

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1703
  • Too many melodeons...
Re: Best unequal temperament for a two row melodeon?
« Reply #10 on: March 28, 2016, 11:17:45 PM »

Incidentally I went round the same train of thought as ukebert more than 20 years ago, and came up with a temperament much sweeter than ET for all the likely keys playable on a D/G/acc 3-row, but never tried it out because each key away from the home key got a few cents further away from the "correct" pitch, which would make playing with other instruments pretty much impossible.

When I found the javatuner page about ten years ago I found I'd been beaten to it by about 400 years, I'd just reinvented Aaron's meantone 1/4 comma...:-(
Logged
Oakwood Model 4 D/G, Castagnari Dony D/G/#, Castagnari Tommy G/C, Baffetti Binci D/G, Hohner Preciosa D/G, Melos Bb/Eb, Lightwave SL5 and Kala California fretless basses

pgroff

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1827
Re: Best unequal temperament for a two row melodeon?
« Reply #11 on: March 29, 2016, 01:26:03 AM »

Incidentally I went round the same train of thought as ukebert more than 20 years ago, and came up with a temperament much sweeter than ET for all the likely keys playable on a D/G/acc 3-row, but never tried it out because each key away from the home key got a few cents further away from the "correct" pitch, which would make playing with other instruments pretty much impossible.

When I found the javatuner page about ten years ago I found I'd been beaten to it by about 400 years, I'd just reinvented Aaron's meantone 1/4 comma...:-(

Hi Ian,

I have used 1/4 comma meantone as a temperament in general purpose playing, and (for my purposes) I made it work reasonably well; for my repertoire and with the D note set at 0 deviation from the D in concert pitch ET, my own experience was that it would work pretty well in about 5 keys (C, G, D, A, E) before "the sharp keys got too flat in pitch" and "the flat keys got too sharp in pitch" to match well enough with other fixed pitch or fretted instruments. True, as you go around the "spiral of meantone fifths / fourths" away from the center, the deviation increases.

More of an issue in ukebert's case, since he doesn't need to match with other instruments, is that to gain pure thirds for every interval (for which the correct notes are available on your keyboard), 1/4 comma meantone sets the fifths very narrow (around 3.4 cents narrower than the already slightly narrow fifths of 12ET). That's great for keyboard music where there is a lot of modulation and every interval can be played on the instrument, but IMO this is may be an unneccessary compromise for a 2 row button accordion if there are some major thirds that you will never play concurrently. 

That's why I posed those questions to him. There might have been a "best compromise" tuning for him that gets him pure thirds (for all that are playable concurrently on his keyboard layout) while compromising the fifths less than does 1/4 comma.

Unfortunately on reading his answer, such a tuning is unlikely to work for him -- it's a little harder to solve the problem that neatly if you stick with only 12 pitches to the octave.

But for example, this might be an interesting idea to try:


1) The fifths from D-A-E are left as in ET (both of those intervals will be around 2 cents narrow of "just"). I know ukebert doesn't plan to play this box with others but if it ever should happen, the fiddle has a good A to match and the familiar intervals on either side will give comfort to the A440-ET-shackled.

2) The fifth E-B is narrowed a little more. The G-D fifth also. These are to taste, but the narrower these are (within reason), the better the G-B major third will sound. Since the Fifth E-B is more likely to be sounded on an A/D box (for E dorian or E major), you might want to narrow the G-D fifth more.

3) This might mean setting G to a pitch that is sharp 3 cents from its value in 12ET, and B note set flat 2 cents from its value in 12ET.  That would get you a G-B major third that is 5 cents narrower than ET. Of course, you could narrow those G-D and E-B fifths right down to their value in 1/4 comma (-3.4 cents each) and then you would sweeten the G-B major third by 6.8 cents (narrower than the third in ET). Yes, that slightly sweetened third will sound better.

4) The F# is flattened a bit more than the B. How much it's flattened will be a compromise between the purity of the D - F# major third and the B - F# fifth. But if you take the F# down to -6.8, then the D-F# major third will be sweeter than ET by that much also.

5) The C natural (if present as an accidental, or in the basses) is sharpened relative to the G.

And so on.

Just a thought - one concept to try out. 

You can get a lot more leverage on sweetening the intervals that you can actually play on an A/D accordion, while imposing fewer compromises on other intervals, if you are willing to let the press vs. draw duplicates of the same note deviate in pitch from one another.

PG
« Last Edit: March 29, 2016, 03:06:41 AM by pgroff »
Logged

pgroff

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1827
Re: Best unequal temperament for a two row melodeon?
« Reply #12 on: March 29, 2016, 01:32:40 AM »



This box is for me to play on my own, at home as my "go to box" or as a solo instrument. It won't be for playing with others, in any circumstances really. It will be an A/D, which means that I will be playing in A, D and E majors, with E dorian, B dorian, aeolian and harmonic minor) and F# aeolian. Accidentals not in those keys probably limited to A# (for playing in B harmonic minor) and F natural (although it wouldn't be used often if at all) and If I want to stray out of those keys then I have other boxes available to do that. I don't want it to be so wonky that people who aren't used to listening to unequal temperaments will be put off. I am not willing to have more than 12 pitches per octave, nor to have more than one value of a particular note (my brain only stretches so far and this will only be a two row after all). I will play lots of different intervals within those tonalities, so any wolf intervals need to be a long way away. The box will be a Hohner and therefore tuned to a medium swing. This means that the effect of the temperament is likely to be subtle but hopefully still noticeable (comments welcome on that point).

Although I probably have the knowledge necessary to create my own temperament, I am not sure that I have the willpower! The current favourite is Werkmeister III which narrows some fifths by a quarter of a comma, resulting in some sweeter thirds whilst retaining some more or less pure fifths. I might end up rejigging it somewhat to give purer fifths though.

Hi ukebert,

See my reply just above, to Ian.  IMO if you want to play in 3 or more keys that are adjacent in the circle of fifths (e.g. EAD), and you don't want more than 12 pitches to the octave or multiple pitches for some notes, then you'd probably best give up on any pure (just) fifths. However, you could have fifths as good as ET fifths in those keys and still start to sweeten up the major thirds a little.

The 1-row diatonic boxes are such a special case, because they only include 7 different pitches (typically) and only a few major thirds and fifths can be sounded as concurrent intervals. (C-G, D-A, G-D and C-E, F-A, G-B on a 1-row box in C). Thus you can go with just tuning or something close to it. An A/D box greatly multiplies the options of concurrent intervals and so it multiplies the need for compromising -- to improve some intervals comes at the expense of others. Not as many concurrent intervals are available as on a piano keyboard though! There are still a lot of concurrent intervals that you won't play on your A/D and that's where you should stow away the wolfs IMO.

The end result may not be too different from the various "modified (irregular) meantone" temperaments and the "well temperaments," but you may be more likely to avoid unnecessary compromises.  Those historical temperaments were made with the piano (organ/harpsichord/clavichord) etc keyboard in mind, not an A/D box.

PG

« Last Edit: March 30, 2016, 11:21:29 AM by pgroff »
Logged
Pages: [1]   Go Up
 


Melodeon.net - (c) Theo Gibb; Clive Williams 2010. The access and use of this website and forum featuring these terms and conditions constitutes your acceptance of these terms and conditions.
SimplePortal 2.3.5 © 2008-2012, SimplePortal