Yahoo Tuning Groups Ultimate Backup tuning Ave Regina

Dear Kalle,

On 20 Mar 2009, at 10:57, Kalle Aho wrote:
> I uploaded my version of the Lassus example:
>
> /tuning/files/Kalle Aho/> AveRegina.mp3
>
> The example is played in three different tunings in this order: 12-> equal, TOP meantone and my adaptive JI.

What is your adaptive meantone? How does that work?

Thanks!

Best
Torsten

--
Torsten Anders
Interdisciplinary Centre for Computer Music Research
University of Plymouth
Office: +44-1752-586219
Private: +44-1752-558917
http://strasheela.sourceforge.net
http://www.torsten-anders.de

🔗Torsten Anders <torsten.anders@...>

On 20 Mar 2009, at 12:39, Torsten Anders wrote:

> Dear Kalle,
>
> On 20 Mar 2009, at 10:57, Kalle Aho wrote:
>> I uploaded my version of the Lassus example:
>>
>> /tuning/files/Kalle Aho/>> AveRegina.mp3
>>
>> The example is played in three different tunings in this order: 12->> equal, TOP meantone and my adaptive JI.
>
> What is your adaptive meantone? How does that work?

I mean of course your adaptive JI, sorry.

Best
Torsten

🔗Torsten Anders <torsten.anders@...>

On 20 Mar 2009, at 12:39, Torsten Anders wrote:

I mean of your "your adaptive JI", sorry.

> How does that work?
>
> Thanks!
>
> Best
> Torsten
>
> --
> Torsten Anders
> Interdisciplinary Centre for Computer Music Research
> University of Plymouth
> Office: +44-1752-586219
> Private: +44-1752-558917
> http://strasheela.sourceforge.net
> http://www.torsten-anders.de
>
>
>

🔗Marcel de Velde <m.develde@...>

Hi Kalle,

I uploaded my version of the Lassus example:
>
> /tuning/files/Kalle Aho/AveRegina.mp3
>
> The example is played in three different tunings in this order: 12-equal,
> TOP meantone and my adaptive JI.
>

Sorry but I find the timbre you used to be very unclear.
Even the 12tet sounds good to me with it.
I don't know how you produce these examples but if it's easy for you to make
a midi file of the 3 tunings so I can use my own timbres that would be
great.

Marcel

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, Torsten Anders <torsten.anders@...> wrote:

> > Dear Kalle,
> > What is your adaptive meantone?
> I mean of your "your adaptive JI", sorry.
> > How does that work?

Hi Torsten,

it works like this:

I determine what I want the tunings of all the different simultaneities to be in the piece I'm tuning. In this example that was quite an easy job to do (I can give a list of the sonorities if you want). Then I determine which pitches are "the same note" in the different simultaneities (note that this doesn't assume octave equivalence). In this example I followed the basic assumption of western music i.e. pitches that are separated by a syntonic comma in JI are treated as the same note. Then I find the (melodic) tuning that minimizes the maximum pitch range for a single note occurring in the piece. If there is more than one solution then I may for example choose the one that has the least number of retunings of sustained notes.

That's it.

I made a program in FreeBasic to find the tuning for just this piece. It randomly varies the bass pitches of the simultaneities and uses a sort of genetic algorithm to find the tuning.

My CSound instrument retunes sustained notes portamento during the attack of the new notes. I added a minuscule amount of pitch micromodulation to the sounds because I noticed that notes have a tendency to "disappear" in totally phase-locked JI sonorities.

Kalle Aho

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, Marcel de Velde <m.develde@...> wrote:

> Sorry but I find the timbre you used to be very unclear.
> Even the 12tet sounds good to me with it.
> I don't know how you produce these examples but if it's easy for you to make
> a midi file of the 3 tunings so I can use my own timbres that would be
> great.

Hi Marcel,

it is quite ironic that I decided to make an audio file because the midi files sounded so unclear to me! :D
I made this in Csound so there's no quick way to produce a corresponding midi file but I can make a mono file with steady sawtooth timbres and no reverb if you want. That won't sound too pretty but at least it shouldn't sound unclear to you.

Kalle Aho

🔗Marcel de Velde <m.develde@...>

>
> it is quite ironic that I decided to make an audio file because the midi
> files sounded so unclear to me! :D
> I made this in Csound so there's no quick way to produce a corresponding
> midi file but I can make a mono file with steady sawtooth timbres and no
> reverb if you want. That won't sound too pretty but at least it shouldn't
> sound unclear to you
>

Ha, ironic indeed :)
But yes i think the steady sawtooth timbre without filter and reverb will
help.
Or if you post the csound score I can convert it to midi.
Midi is still the best I think as one can select your own timbres when you
load it in you midi sequencer and compare it directly with thesame sound to
other examples.

Marcel

🔗Torsten Anders <torsten.anders@...>

Dear Kalle,

On 20 Mar 2009, at 14:20, Kalle Aho wrote:
> > How does your adaptive JI work?
>
> I determine what I want the tunings of all the different > simultaneities to be in the piece I'm tuning.
[...]
> I added a minuscule amount of pitch micromodulation to the sounds > because I noticed that notes have a tendency to "disappear" in > totally phase-locked JI sonorities.

OK, I understand that for each chord you defined some tuning which is very close to JI, but with minor "errors" for some phasing.

> In this example that was quite an easy job to do (I can give a list > of the sonorities if you want). Then I determine which pitches are > "the same note" in the different simultaneities (note that this > doesn't assume octave equivalence). In this example I followed the > basic assumption of western music i.e. pitches that are separated > by a syntonic comma in JI are treated as the same note. Then I find > the (melodic) tuning that minimizes the maximum pitch range for a > single note occurring in the piece. If there is more than one > solution then I may for example choose the one that has the least > number of retunings of sustained notes.

I don't follow. You try to find a scale with minimal total number of notes to play the piece such that you only use your almost JI chords? Do you address the chord roots in any specific way?

Thank you!

Best
Torsten

>
> That's it.
>
> I made a program in FreeBasic to find the tuning for just this > piece. It randomly varies the bass pitches of the simultaneities > and uses a sort of genetic algorithm to find the tuning.
>
> My CSound instrument retunes sustained notes portamento during the > attack of the new notes. I added a minuscule amount of pitch > micromodulation to the sounds because I noticed that notes have a > tendency to "disappear" in totally phase-locked JI sonorities.

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, Torsten Anders <torsten.anders@...> wrote:

> OK, I understand that for each chord you defined some tuning which is
> very close to JI, but with minor "errors" for some phasing.

No, the "errors" are added by the Csound instrument with the jitter-opcode. The tunings defined for the simultaneities are exact JI.

> I don't follow. You try to find a scale with minimal total number of
> notes to play the piece such that you only use your almost JI chords?

No, each note (C1, E2 and so on) may be tuned to slightly different pitch in different simultaneities. That means that there is a pitch range for each note i.e. the log-frequency difference of the highest and lowest pitch of the note occurring in the piece. One of the notes will have the greatest pitch range. My adaptive tuning is the one
which minimizes this range. Note that no melodic center tuning is assumed beforehand.

> Do you address the chord roots in any specific way?

I don't.

🔗Carl Lumma <carl@...>

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
>
> I made a program in FreeBasic to find the tuning for just this
> piece. It randomly varies the bass pitches of the simultaneities
> and uses a sort of genetic algorithm to find the tuning.

Excellent result, Kalle, and I'm very glad to see you've coded
up your idea. (I think the timbre is perfect, too.)

I would like to see the list of simultaneities, if you could
post it.

Was there as case when two notes, e.g. D3 and D4, were not
tuned an octave apart (can you inspect the resulting tuning
in this way)?

-Carl

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:
>
> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
> >
> > I made a program in FreeBasic to find the tuning for just this
> > piece. It randomly varies the bass pitches of the simultaneities
> > and uses a sort of genetic algorithm to find the tuning.
>
> Excellent result, Kalle, and I'm very glad to see you've coded
> up your idea. (I think the timbre is perfect, too.)

Thanks, Carl!

> I would like to see the list of simultaneities, if you could
> post it.

As chord types:

4:5:6
4:5:6:8
2:4:5:6
2:3:4:5
2:3:4:5
2:5:6:8
6:15:20:24
3:6:9:10
3:6:8:10
3:6:8:9
2:4:5:6
4:8:9:12
6:12:15:20
3:5:6:9
3:5:6:8
6:9:12:16
2:3:4:5
4:6:8:9
2:3:4:5
4:5:6:8

> Was there as case when two notes, e.g. D3 and D4, were not
> tuned an octave apart (can you inspect the resulting tuning
> in this way)?

Well, for example G3 has only one pitch while G4 varies between two different versions which differ by exactly 1/3-comma (I'm sure of this, the value is accurate to 15 digits). The center between max and min of G4 differs from G3 by octave + 1/6-comma. All the other notes seem to have two versions separated by 1/3-comma too! I don't know how to explain this.

Kalle Aho

🔗Marcel de Velde <m.develde@...>

🔗Carl Lumma <carl@...>

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:

> > I would like to see the list of simultaneities, if you could
> > post it.
>
> As chord types:
>
> 4:5:6
> 4:5:6:8
> 2:4:5:6
> 2:3:4:5
> 2:3:4:5
> 2:5:6:8
> 6:15:20:24
> 3:6:9:10
> 3:6:8:10
> 3:6:8:9
> 2:4:5:6
> 4:8:9:12
> 6:12:15:20
> 3:5:6:9
> 3:5:6:8
> 6:9:12:16
> 2:3:4:5
> 4:6:8:9
> 2:3:4:5
> 4:5:6:8

Thx!

> > Was there as case when two notes, e.g. D3 and D4, were not
> > tuned an octave apart (can you inspect the resulting tuning
> > in this way)?
>
> Well, for example G3 has only one pitch while G4 varies between
> two different versions which differ by exactly 1/3-comma (I'm
> sure of this, the value is accurate to 15 digits).

Interesting. Don't know if you saw my recent post where I
speculated whether the 1/4-comma 'Vicentino' solution was
really optimal. It gives you 1/4-comma motions on 5ths only.
But 1/5-comma meantone would give 1/5-comma motions on 5ths
and 3rds... probably there is a solution where 1/3-comma makes
sense (involving minor 3rds no doubt).

-Carl

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:
>
> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
>
> > > I would like to see the list of simultaneities, if you could
> > > post it.
> >
> > As chord types:
> >
> > 4:5:6
> > 4:5:6:8
> > 2:4:5:6
> > 2:3:4:5
> > 2:3:4:5
> > 2:5:6:8
> > 6:15:20:24
> > 3:6:9:10
> > 3:6:8:10
> > 3:6:8:9
> > 2:4:5:6
> > 4:8:9:12
> > 6:12:15:20
> > 3:5:6:9
> > 3:5:6:8
> > 6:9:12:16
> > 2:3:4:5
> > 4:6:8:9
> > 2:3:4:5
> > 4:5:6:8
>
> Thx!
>
> > > Was there as case when two notes, e.g. D3 and D4, were not
> > > tuned an octave apart (can you inspect the resulting tuning
> > > in this way)?
> >
> > Well, for example G3 has only one pitch while G4 varies between
> > two different versions which differ by exactly 1/3-comma (I'm
> > sure of this, the value is accurate to 15 digits).
>
> Interesting. Don't know if you saw my recent post where I
> speculated whether the 1/4-comma 'Vicentino' solution was
> really optimal. It gives you 1/4-comma motions on 5ths only.
> But 1/5-comma meantone would give 1/5-comma motions on 5ths
> and 3rds... probably there is a solution where 1/3-comma makes
> sense (involving minor 3rds no doubt).

As you can verify from the list, all simultaneities are 9-limit just. I don't think this is the case with Monz's Vicentino-solution or is it? My solution should be the one that minimizes the maximal pitch range of notes with these sonorities. In this example the minimized max pitch range happens to be exactly 1/3-comma. Maybe the 1/3-comma somehow comes from the 9-limit chord changes?

Kalle Aho

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, Marcel de Velde <m.develde@...> wrote:
>
> I had a more carefull listen to the adaptive ji version and I can't say I
> like it.The way the tones shift all the time sounds very unnatural to me.
> How do you like the adaptive ji version yourself Kallo?

Hi Marcel,

my name is Kalle, not "Kallo" which actually means skull in Finnish. :D

The only thing that I don't like about my adaptive version is the very short overall pitch shift down and back up in in the fourth measure when the alto moves E4-D4-E4. Maybe shorter notes could be tuned against longer ones without bothering about their pitch range contribution at all. There is plenty of room for piece-specific decisions.

Kalle Aho

🔗Aaron Andrew Hunt <aaronhunt@...>

Hi Kalle.

I liked your adaptive Lassus JI a lot, and it gave me the idea to
show how my keyboard can be used to do something similar
in real time. I play the passage twice - once with pure intervals
(nearly pure 5-Limit of 205ET) and once adapting as I go. The
changes in pitch are more noticeable than yours, because the
keys have to be restruck, and also my choices are in some cases
different than yours. The only place I find problematic is the last
suspension. If I keep the soprano, I end up having to shift the last
root 2 JNDs (1/2 comma), which is too noticeable, so the
suspension is restruck, which takes away some of the beauty.
I wrote some more in the commentary for the video on YouTube:

<http://www.youtube.com/watch?v=63YrrA3X4tg>

Thanks for the kick!
Aaron
=====

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
>
> --- In tuning@yahoogroups.com, Torsten Anders <torsten.anders@> wrote:
>
> > > Dear Kalle,
> > > What is your adaptive meantone?
> > I mean of your "your adaptive JI", sorry.
> > > How does that work?
>
> Hi Torsten,
>
> it works like this:
>
> I determine what I want the tunings of all the different simultaneities to be in the piece I'm tuning. In this example that was quite an easy job to do (I can give a list of the sonorities if you want). Then I determine which pitches are "the same note" in the different simultaneities (note that this doesn't assume octave equivalence). In this example I followed the basic assumption of western music i.e. pitches that are separated by a syntonic comma in JI are treated as the same note. Then I find the (melodic) tuning that minimizes the maximum pitch range for a single note occurring in the piece. If there is more than one solution then I may for example choose the one that has the least number of retunings of sustained notes.
>
> That's it.
>
> I made a program in FreeBasic to find the tuning for just this piece. It randomly varies the bass pitches of the simultaneities and uses a sort of genetic algorithm to find the tuning.
>
> My CSound instrument retunes sustained notes portamento during the attack of the new notes. I added a minuscule amount of pitch micromodulation to the sounds because I noticed that notes have a tendency to "disappear" in totally phase-locked JI sonorities.
>
> Kalle Aho
>

🔗Carl Lumma <carl@...>

Kalle wrote:

> > Interesting. Don't know if you saw my recent post where I
> > speculated whether the 1/4-comma 'Vicentino' solution was
> > really optimal. It gives you 1/4-comma motions on 5ths only.
> > But 1/5-comma meantone would give 1/5-comma motions on 5ths
> > and 3rds... probably there is a solution where 1/3-comma makes
> > sense (involving minor 3rds no doubt).
>
> As you can verify from the list, all simultaneities are 9-limit
> just. I don't think this is the case with Monz's Vicentino-
> solution or is it?

There are only two 9-limit chords here, and the 9ths are
eighth-note passing tones. But right, the 9ths are in meantone
in Monz's version.

> My solution should be the one that minimizes the maximal pitch
> range of notes with these sonorities.

1/3 1/4 1/5 1/6
min3rds 0 1/4 2/5 3/6
maj3rds 1/3 0 1/5 2/6
5ths 1/3 1/4 1/5 1/6
MAX .33 .25 .4 .5

So for triads, you should be seeing 1/4-comma shifts, and
I was wrong about 1/5-comma being better. Let's add 9ths...

1/3 1/4 1/5 1/6
min3rds 0 1/4 2/5 3/6
maj3rds 1/3 0 1/5 2/6
5ths 1/3 1/4 1/5 1/6
9ths 2/3 1/2 2/5 2/6
MAX .67 .5 .4 .5

Now 5th-comma is better.

So... not sure how to explain your results.

-Carl

🔗Carl Lumma <carl@...>

Hi Aaron,

This gives me a "this video has been removed by the user" message.
Will it come back up?

-Carl

> Hi Kalle.
>
> I liked your adaptive Lassus JI a lot, and it gave me the idea to
> show how my keyboard can be used to do something similar
> in real time. I play the passage twice - once with pure intervals
> (nearly pure 5-Limit of 205ET) and once adapting as I go. The
> changes in pitch are more noticeable than yours, because the
> keys have to be restruck, and also my choices are in some cases
> different than yours. The only place I find problematic is the last
> suspension. If I keep the soprano, I end up having to shift the last
> root 2 JNDs (1/2 comma), which is too noticeable, so the
> suspension is restruck, which takes away some of the beauty.
> I wrote some more in the commentary for the video on YouTube:
>
> <http://www.youtube.com/watch?v=63YrrA3X4tg>
>
> Thanks for the kick!
> Aaron

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:
>
> Kalle wrote:
>
> > > Interesting. Don't know if you saw my recent post where I
> > > speculated whether the 1/4-comma 'Vicentino' solution was
> > > really optimal. It gives you 1/4-comma motions on 5ths only.
> > > But 1/5-comma meantone would give 1/5-comma motions on 5ths
> > > and 3rds... probably there is a solution where 1/3-comma makes
> > > sense (involving minor 3rds no doubt).
> >
> > As you can verify from the list, all simultaneities are 9-limit
> > just. I don't think this is the case with Monz's Vicentino-
> > solution or is it?
>
> There are only two 9-limit chords here, and the 9ths are
> eighth-note passing tones. But right, the 9ths are in meantone
> in Monz's version.
>
> > My solution should be the one that minimizes the maximal pitch
> > range of notes with these sonorities.
>
> 1/3 1/4 1/5 1/6
> min3rds 0 1/4 2/5 3/6
> maj3rds 1/3 0 1/5 2/6
> 5ths 1/3 1/4 1/5 1/6
> MAX .33 .25 .4 .5
>
> So for triads, you should be seeing 1/4-comma shifts, and
> I was wrong about 1/5-comma being better. Let's add 9ths...
>
> 1/3 1/4 1/5 1/6
> min3rds 0 1/4 2/5 3/6
> maj3rds 1/3 0 1/5 2/6
> 5ths 1/3 1/4 1/5 1/6
> 9ths 2/3 1/2 2/5 2/6
> MAX .67 .5 .4 .5
>
> Now 5th-comma is better.
>
> So... not sure how to explain your results.

Hi Carl,

sevenths of 9:5 (or seconds of 10:9) should be considered too as they occur in some of the simultaneities.

You are measuring just the max error of the center tuning. Instead of that I think one should find the maximum pitch movement in successive chords and this I believe is calculated by observing also the direction (positive or negative) of the errors and summing the absolute values of the largest negative and positive errors.

This gives

1/3 1/4 1/5 1/6
.67 .75 1 1.17

Kalle Aho

🔗Aaron Andrew Hunt <aaronhunt@...>

OK, YouTube finally did not ruin the audio of this upload,
although the video now looks terrible:

<http://www.youtube.com/watch?v=Jo1HenpSjFM>

AAH
=====

--- In tuning@yahoogroups.com, "Aaron Andrew Hunt" <aaronhunt@...> wrote:
>
> Hi Kalle.
>
> I liked your adaptive Lassus JI a lot, and it gave me the idea to
> show how my keyboard can be used to do something similar
> in real time. I play the passage twice - once with pure intervals
> (nearly pure 5-Limit of 205ET) and once adapting as I go. The
> changes in pitch are more noticeable than yours, because the
> keys have to be restruck, and also my choices are in some cases
> different than yours. The only place I find problematic is the last
> suspension. If I keep the soprano, I end up having to shift the last
> root 2 JNDs (1/2 comma), which is too noticeable, so the
> suspension is restruck, which takes away some of the beauty.
> I wrote some more in the commentary for the video on YouTube:
>
> <http://www.youtube.com/watch?v=63YrrA3X4tg>
>
> Thanks for the kick!
> Aaron
> =====
>
>
> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
> >
> > --- In tuning@yahoogroups.com, Torsten Anders <torsten.anders@> wrote:
> >
> > > > Dear Kalle,
> > > > What is your adaptive meantone?
> > > I mean of your "your adaptive JI", sorry.
> > > > How does that work?
> >
> > Hi Torsten,
> >
> > it works like this:
> >
> > I determine what I want the tunings of all the different simultaneities to be in the piece I'm tuning. In this example that was quite an easy job to do (I can give a list of the sonorities if you want). Then I determine which pitches are "the same note" in the different simultaneities (note that this doesn't assume octave equivalence). In this example I followed the basic assumption of western music i.e. pitches that are separated by a syntonic comma in JI are treated as the same note. Then I find the (melodic) tuning that minimizes the maximum pitch range for a single note occurring in the piece. If there is more than one solution then I may for example choose the one that has the least number of retunings of sustained notes.
> >
> > That's it.
> >
> > I made a program in FreeBasic to find the tuning for just this piece. It randomly varies the bass pitches of the simultaneities and uses a sort of genetic algorithm to find the tuning.
> >
> > My CSound instrument retunes sustained notes portamento during the attack of the new notes. I added a minuscule amount of pitch micromodulation to the sounds because I noticed that notes have a tendency to "disappear" in totally phase-locked JI sonorities.
> >
> > Kalle Aho
> >
>

🔗Aaron Andrew Hunt <aaronhunt@...>

YouTube was distorting the audio so I removed it. Then uploaded
in a different format and the same thing happened - tried several
other formats always with the same result, until I tried .AVI and the
audio now sounds fine, but the video quality is bad. My verdict is
that Flash is not the greatest format on the planet.

Anyway, I hope you like the video!

Here's the link again:
<http://www.youtube.com/watch?v=Jo1HenpSjFM>

It says you can watch in HD, but I tried and it does not improve
the video quality at all. I don't know why YouTube would not
correctly reproduce the audio on an .mp4 or .mov or .wmv file;
I've not had such a problem before.

Yours,
AAH
=====

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:
>
> Hi Aaron,
>
> This gives me a "this video has been removed by the user" message.
> Will it come back up?
>
> -Carl
>
> > Hi Kalle.
> >
> > I liked your adaptive Lassus JI a lot, and it gave me the idea to
> > show how my keyboard can be used to do something similar
> > in real time. I play the passage twice - once with pure intervals
> > (nearly pure 5-Limit of 205ET) and once adapting as I go. The
> > changes in pitch are more noticeable than yours, because the
> > keys have to be restruck, and also my choices are in some cases
> > different than yours. The only place I find problematic is the last
> > suspension. If I keep the soprano, I end up having to shift the last
> > root 2 JNDs (1/2 comma), which is too noticeable, so the
> > suspension is restruck, which takes away some of the beauty.
> > I wrote some more in the commentary for the video on YouTube:
> >
> > <http://www.youtube.com/watch?v=63YrrA3X4tg>
> >
> > Thanks for the kick!
> > Aaron
>

🔗Kalle Aho <kalleaho@...>

Hi Aaron,

this is excellent, thanks!

Kalle

--- In tuning@yahoogroups.com, "Aaron Andrew Hunt" <aaronhunt@...> wrote:
>
> YouTube was distorting the audio so I removed it. Then uploaded
> in a different format and the same thing happened - tried several
> other formats always with the same result, until I tried .AVI and the
> audio now sounds fine, but the video quality is bad. My verdict is
> that Flash is not the greatest format on the planet.
>
> Anyway, I hope you like the video!
>
> Here's the link again:
> <http://www.youtube.com/watch?v=Jo1HenpSjFM>
>
> It says you can watch in HD, but I tried and it does not improve
> the video quality at all. I don't know why YouTube would not
> correctly reproduce the audio on an .mp4 or .mov or .wmv file;
> I've not had such a problem before.
>
> Yours,
> AAH
> =====
>
> --- In tuning@yahoogroups.com, "Carl Lumma" <carl@> wrote:
> >
> > Hi Aaron,
> >
> > This gives me a "this video has been removed by the user" message.
> > Will it come back up?
> >
> > -Carl
> >
> > > Hi Kalle.
> > >
> > > I liked your adaptive Lassus JI a lot, and it gave me the idea to
> > > show how my keyboard can be used to do something similar
> > > in real time. I play the passage twice - once with pure intervals
> > > (nearly pure 5-Limit of 205ET) and once adapting as I go. The
> > > changes in pitch are more noticeable than yours, because the
> > > keys have to be restruck, and also my choices are in some cases
> > > different than yours. The only place I find problematic is the last
> > > suspension. If I keep the soprano, I end up having to shift the last
> > > root 2 JNDs (1/2 comma), which is too noticeable, so the
> > > suspension is restruck, which takes away some of the beauty.
> > > I wrote some more in the commentary for the video on YouTube:
> > >
> > > <http://www.youtube.com/watch?v=63YrrA3X4tg>
> > >
> > > Thanks for the kick!
> > > Aaron
> >
>

🔗Carl Lumma <carl@...>

--- In tuning@yahoogroups.com, "Aaron Andrew Hunt" <aaronhunt@...> wrote:
>
> OK, YouTube finally did not ruin the audio of this upload,
> although the video now looks terrible:
>
> <http://www.youtube.com/watch?v=Jo1HenpSjFM>
>
> AAH

I've heard the HQ mode does something bad to the audio.
Or did when they first debuted it. Or something.

Great work! It really demonstrates the power of an
extended keyboard.

-Carl

🔗Carl Lumma <carl@...>

I agree that Flash is obnoxious (proprietary binary format that
doesn't integrate well with the www or the browsing experience),
but it's just a wrapper for a smorgasbord of video codecs it
supports. The blame probably lies with youtube. They have been
applying audio compression to uploads lately. They also
transcode the video. Vimeo is supposed to be better on these
counts but the traffic is obviously not as high.

-Carl

🔗Carl Lumma <carl@...>

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
>
> Hi Carl,
>
> sevenths of 9:5 (or seconds of 10:9) should be considered too
> as they occur in some of the simultaneities.

OK.

> You are measuring just the max error of the center tuning.
> Instead of that I think one should find the maximum pitch
> movement in successive chords and this I believe is calculated
> by observing also the direction (positive or negative) of the
> errors and summing the absolute values of the largest negative
> and positive errors.
>
> This gives
>
> 1/3 1/4 1/5 1/6
> .67 .75 1 1.17

You're right. But I don't get these numbers.
I get:

1/3 1/4 1/5 1/6
roots 0 0 0 0
min3rds 0 -1/4 -2/5 -3/6
maj3rds -1/3 0 +1/5 +2/6
5ths -1/3 -1/4 -1/5 -1/6
MAX .33 .25 .6 .83
9th -2/3 -2/4 -2/5 -2/6
MAX .67 .5 .6 .83

-Carl

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, "Carl Lumma" <carl@...> wrote:
>
> --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
> >
> > Hi Carl,
> >
> > sevenths of 9:5 (or seconds of 10:9) should be considered too
> > as they occur in some of the simultaneities.
>
> OK.
>
> > You are measuring just the max error of the center tuning.
> > Instead of that I think one should find the maximum pitch
> > movement in successive chords and this I believe is calculated
> > by observing also the direction (positive or negative) of the
> > errors and summing the absolute values of the largest negative
> > and positive errors.
> >
> > This gives
> >
> > 1/3 1/4 1/5 1/6
> > .67 .75 1 1.17
>
> You're right. But I don't get these numbers.
> I get:
>
> 1/3 1/4 1/5 1/6
> roots 0 0 0 0
> min3rds 0 -1/4 -2/5 -3/6
> maj3rds -1/3 0 +1/5 +2/6
> 5ths -1/3 -1/4 -1/5 -1/6
> MAX .33 .25 .6 .83
> 9th -2/3 -2/4 -2/5 -2/6
> MAX .67 .5 .6 .83

Actually I was wrong: of course one shouldn't consider the sign of the errors as the octave inverse of an interval has the opposite error. But theoretically wouldn't the maximal corrective pitch movement in successive simultaneities be the double of the max error? That would amount to a complete comma in the case of 1/4-comma meantone if there are sonorities with 10:9 or 9:5.

Kalle Aho

🔗Carl Lumma <carl@...>

[Click "Show Message Option" > "Use Fixed Width Font" to view
this message correctly]

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:

> > > You are measuring just the max error of the center tuning.
> > > Instead of that I think one should find the maximum pitch
> > > movement in successive chords and this I believe is calculated
> > > by observing also the direction (positive or negative) of the
> > > errors and summing the absolute values of the largest negative
> > > and positive errors. This gives
> > >
> > > 1/3 1/4 1/5 1/6
> > > .67 .75 1 1.17
> >
> > You're right. But I don't get these numbers.
> > I get:
> >
> > 1/3 1/4 1/5 1/6
> > roots 0 0 0 0
> > min3rds 0 -1/4 -2/5 -3/6
> > maj3rds -1/3 0 +1/5 +2/6
> > 5ths -1/3 -1/4 -1/5 -1/6
> > MAX .33 .25 .6 .83
> > 9th -2/3 -2/4 -2/5 -2/6
> > MAX .67 .5 .6 .83
>
> Actually I was wrong: of course one shouldn't consider the sign
> of the errors as the octave inverse of an interval has the
> opposite error.

I'm thinking of chord identities (pitches), not intervals.

If you've got the progression CM -> Amin and 1/5-comma
meantone is the center tuning, the E is adjusted +1/5 comma
in CM and -1/5 comma in Amin, resulting in a 2/5 comma shift,
right? I was confused because the max error is the max shift
in 1/4-comma meantone, since its errors all have the same sign.

-Carl

🔗Kalle Aho <kalleaho@...>

--- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@...> wrote:
>
> --- In tuning@yahoogroups.com, "Carl Lumma" <carl@> wrote:
> >
> > --- In tuning@yahoogroups.com, "Kalle Aho" <kalleaho@> wrote:
> > >
> > > Hi Carl,
> > >
> > > sevenths of 9:5 (or seconds of 10:9) should be considered too
> > > as they occur in some of the simultaneities.
> >
> > OK.
> >
> > > You are measuring just the max error of the center tuning.
> > > Instead of that I think one should find the maximum pitch
> > > movement in successive chords and this I believe is calculated
> > > by observing also the direction (positive or negative) of the
> > > errors and summing the absolute values of the largest negative
> > > and positive errors.
> > >
> > > This gives
> > >
> > > 1/3 1/4 1/5 1/6
> > > .67 .75 1 1.17
> >
> > You're right. But I don't get these numbers.
> > I get:
> >
> > 1/3 1/4 1/5 1/6
> > roots 0 0 0 0
> > min3rds 0 -1/4 -2/5 -3/6
> > maj3rds -1/3 0 +1/5 +2/6
> > 5ths -1/3 -1/4 -1/5 -1/6
> > MAX .33 .25 .6 .83
> > 9th -2/3 -2/4 -2/5 -2/6
> > MAX .67 .5 .6 .83
>
> Actually I was wrong: of course one shouldn't consider the sign of the errors as the octave inverse of an interval has the opposite error. But theoretically wouldn't the maximal corrective pitch movement in successive simultaneities be the double of the max error? That would amount to a complete comma in the case of 1/4-comma meantone if there are sonorities with 10:9 or 9:5.

Sorry, I was wrong again: the maximal movement won't be double of the max error but equal to max error.

Kalle Aho