Hot seat

Hey Johnny...I don't feel like I'm on a hot seat...and, I have listened to non et Bach, and would be happy to have a list of just which artists recorded non et Bach music...I've asked several times on this list over the years, but had very few responses. Perhaps you could post some sort of list here, I'm sure others may be interested as well..best...HHH

microstick@msn.com wrote:

> Hey Johnny...I don't feel like I'm on a hot seat...and, I have > listened to non et Bach, and would be happy to have a list of just > which artists recorded non et Bach music...I've asked several times on > this list over the years, but had very few responses. Perhaps you > could post some sort of list here, I'm sure others may be interested > as well..best...HHH
> I wouldn't mind seeing a discography of non-12tet Bach.

--
* David Beardsley
* microtonal guitar
* http://biink.com/db

[ Attachment content not displayed ]

> I would assume that every harpsichord recording done in the
> last 30 years is non12et

And you would probably be right.

-Carl

Re: Hot seat

> I would assume that every harpsichord recording done in the
> last 30 years is non12et

And you would probably be right.

-Carl

However, re Bach recordings, there is nothing I know of besides what has
been previously stated. Bach has been basically ET or WIII. No one has
suggested otherwise.

-Johnny

[ Attachment content not displayed ]

Dante Rosati wrote:
>
> I would assume that every harpsichord recording done in the last 30
years is non12et

Hi Dante,
I wouldn't make that assumption. Probably the vast majority, yes.
And I have no ready counter-example. Still, it stands to reason that
some experimentalist would have taken a distinctively 12-EDO piano
composer's work and transcribed or transferred them to harpsichord.
I'd rather like to hear some of Debussy done that way.

> On 1/14/07, David Beardsley <db@...> wrote:
> >
> > microstick@... <microstick%40msn.com> wrote:
> >
> > > Hey Johnny...I don't feel like I'm on a hot seat...and, I have
listened to non et Bach, and would be happy to have a list of just
which artists recorded non et Bach music...I've asked several times
on this list over the years, but had very few responses. Perhaps you
could post some sort of list here, I'm sure others may be interested
as well..best...HHH
> > >
> >
> > I wouldn't mind seeing a discography of non-12tet Bach.
> >
> > * David Beardsley

I agree with these guys. It would be worthwhile exploring the
variety of approaches taken to tuning Bach, especially with a view to
finding out which ones really work in performance.

Regards,
Yahya

> > > I wouldn't mind seeing a discography of non-12tet Bach.
> > >
> > > * David Beardsley
>
> I agree with these guys. It would be worthwhile exploring the
> variety of approaches taken to tuning Bach, especially with a view
> to finding out which ones really work in performance.

Or perhaps we should revisit this post of Aaron Johnson's

/tuning/topicId_60103.html#60103

I'll spill the beans: nobody has correctly identified the
three Bach files yet, though Cameron's second guess was correct.

Granted, a single timbre helps to hear the differences.
Especially a bright timbre like the harpsichord. And on
a real instument resonances between strings might help
further. So the buddha isn't dead yet, but it seems to
me there's a lot of B.S. written about well temperaments
on this list.

-Carl

> I'll spill the beans: nobody has correctly identified the
> three Bach files yet, though Cameron's second guess was correct.

It might be worth saying that according to my spreadsheet
anyway, Werck 3 has about the same total error as ET in
A major, with Young a bit better. But the only difference
I can hear in these files, if I'm not fooling myself, is to
listen for the 12-cent difference in starting pitch between
Werck 3 (first file) and ET (second file).

-Carl

Interesting...so I take it that, at least in these files, Johnny
preferred Young?

-A.

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > > > I wouldn't mind seeing a discography of non-12tet Bach.
> > > >
> > > > * David Beardsley
> >
> > I agree with these guys. It would be worthwhile exploring the
> > variety of approaches taken to tuning Bach, especially with a view
> > to finding out which ones really work in performance.
>
> Or perhaps we should revisit this post of Aaron Johnson's
>
> /tuning/topicId_60103.html#60103
>
> I'll spill the beans: nobody has correctly identified the
> three Bach files yet, though Cameron's second guess was correct.
>
> Granted, a single timbre helps to hear the differences.
> Especially a bright timbre like the harpsichord. And on
> a real instument resonances between strings might help
> further. So the buddha isn't dead yet, but it seems to
> me there's a lot of B.S. written about well temperaments
> on this list.
>
> -Carl
>

Obviously, Johnny could not discern cent differences as he claimed to be
able to. The differences between Werckmeister (1681) and Young (1807) appear
on the 2nd, 4th, 7th, 9th, and 11th degrees of the chromatic twelve-tone
scale as given below:

W2 - Y2 = -3.910 cents
W4 - Y4 = -1.955 cents
W7 - Y7 = -1.955 cents
W9 - Y9 = -5.865 cents
W11 - Y11 = 1.955 cents

Total absolute difference : 15.6400 cents
Average absolute difference: 1.3033 cents
Root mean square difference: 2.2574 cents
Highest absolute difference: 5.8650 cents
Number of notes different: 5

The consecutive intervals of A Major respectively in Werckmeister and Young
are:

AM________W_______________Y____________D______
1-2:____203.910 cents____196.090 cents____7.820 cents
2-3:____198.045 cents____200.000 cents____-1.955 cents
3-4:____101.955 cents____105.865 cents____-3.910 cents
4-5:____198.045 cents____196.090 cents____1.955 cents
5-6:____198.045 cents____196.090 cents____1.955 cents
6-7:____203.910 cents____203.910 cents____0.000 cents
7-8:____ 96.090 cents____101.955 cents____-5.865 cents

Assuming that C4 is 262 Hz, these would correspond to the frequencies
provided below:

________W________________Y____________D______
A:____437.6543 cps____439.1395 cps____1.4852 cps
B:____492.3611 cps____491.8054 cps____0.5557 cps
C#:___552.0329 cps____552.0329 cps____0.0000 cps
D:____585.5193 cps____586.8432 cps____1.3239 cps
E:____656.4815 cps____657.2232 cps____0.7418 cps
F#:___736.0439 cps____736.0439 cps____0.0000 cps
G#:___828.0494 cps____828.0494 cps____0.0000 cps
A:____875.3086 cps____878.2790 cps____2.9704 cps

Oz.

----- Original Message -----
From: "Aaron Krister Johnson" <aaron@dividebypi.com>
To: <tuning@yahoogroups.com>
Sent: 16 Ocak 2007 Sal� 20:17
Subject: [tuning] Re: Hot seat

>
> Interesting...so I take it that, at least in these files, Johnny
> preferred Young?
>
> -A.
>
> --- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
> >
> > > > > I wouldn't mind seeing a discography of non-12tet Bach.
> > > > >
> > > > > * David Beardsley
> > >
> > > I agree with these guys. It would be worthwhile exploring the
> > > variety of approaches taken to tuning Bach, especially with a view
> > > to finding out which ones really work in performance.
> >
> > Or perhaps we should revisit this post of Aaron Johnson's
> >
> > /tuning/topicId_60103.html#60103
> >
> > I'll spill the beans: nobody has correctly identified the
> > three Bach files yet, though Cameron's second guess was correct.
> >
> > Granted, a single timbre helps to hear the differences.
> > Especially a bright timbre like the harpsichord. And on
> > a real instument resonances between strings might help
> > further. So the buddha isn't dead yet, but it seems to
> > me there's a lot of B.S. written about well temperaments
> > on this list.
> >
> > -Carl
> >
>

It might be worth saying that according to my spreadsheet
anyway, Werck 3 has about the same total error as ET in
A major, with Young a bit better. But the only difference
I can hear in these files, if I'm not fooling myself, is to
listen for the 12-cent difference in starting pitch between
Werck 3 (first file) and ET (second file).

-Carl

You played us then you trickster. ;)

----- Original Message -----
From: "Carl Lumma" <clumma@yahoo.com>
To: <tuning@yahoogroups.com>
Sent: 16 Ocak 2007 Sal� 19:32
Subject: [tuning] Re: Hot seat

> > > > I wouldn't mind seeing a discography of non-12tet Bach.
> > > >
> > > > * David Beardsley
> >
> > I agree with these guys. It would be worthwhile exploring the
> > variety of approaches taken to tuning Bach, especially with a view
> > to finding out which ones really work in performance.
>
> Or perhaps we should revisit this post of Aaron Johnson's
>
> /tuning/topicId_60103.html#60103
>
> I'll spill the beans: nobody has correctly identified the
> three Bach files yet, though Cameron's second guess was correct.
>
> Granted, a single timbre helps to hear the differences.
> Especially a bright timbre like the harpsichord. And on
> a real instument resonances between strings might help
> further. So the buddha isn't dead yet, but it seems to
> me there's a lot of B.S. written about well temperaments
> on this list.
>
> -Carl
>
>
>

[ Attachment content not displayed ]

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > I'll spill the beans: nobody has correctly identified the
> > three Bach files yet, though Cameron's second guess was correct.
>
> It might be worth saying that according to my spreadsheet
> anyway, Werck 3 has about the same total error as ET in
> A major, with Young a bit better. But the only difference
> I can hear in these files, if I'm not fooling myself, is to
> listen for the 12-cent difference in starting pitch between
> Werck 3 (first file) and ET (second file).

I haven't listened to them yet, but harpsichord is a bad timbre to use
if you want to bring out the differences between tunings rather than
gloss them over.

If Carl's 3 examples prove anything, it is that different WTs sound more
alike than either do to ET.

Since WIII sounds different in each key, there is no exact way to pick up
the tuning from a drop the needle approach. I gave my honest impressions from
a general listen.

Ozan -- there is no direct relationship between these MIDI scales, and the
production and apprehension of the movement (and direction) of single cents in
a series. But I did enjoy agreeing with you on which was which.

best, Johnny

Forgot to mention that I like the third one the best. And I was going
to say that my experience with these WTs is limited to a bit of
noodling on the synth, but just realized that that is not true at all-
I've sung in both, and heard a great deal of cembalo played in both,
some years ago. At A-415, come to think of it.

Both WTs are easier and more enjoyable, physically, to sing in than
equal temperament, at any A. This can't be explained, but felt.

-Cameron Bobro

> I haven't listened to them yet, but harpsichord is a bad timbre
> to use if you want to bring out the differences between tunings
> rather than gloss them over.

I disagree. Why do you think so? One of the two people to
get my WT comparison test right last time it came around said
the harpsichord piece in that test was crucial to his success.

-Carl

> If Carl's 3 examples prove anything, it is that different WTs
> sound more alike than either do to ET.

It's true that both you and Ozan correctly identified the ET
file, but with a test like this it probably isn't statistically
significant. So it's hard to draw any real conclusions from
the test, other than that the difference between these three
tunings was not on the order of 'obvious' to those who listened.

-Carl

> Both WTs are easier and more enjoyable, physically, to sing in than
> equal temperament, at any A. This can't be explained, but felt.

I have an explanation: expectation bias.

-Carl

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > Both WTs are easier and more enjoyable, physically, to sing in
than
> > equal temperament, at any A. This can't be explained, but felt.
>
> I have an explanation: expectation bias.
>
> -Carl
>

Hahaha! What kind of tomfool "explanation" is that? You're putting the
cart before the horse- how do you think I got into all this stuff? The
other way around. Boy, this feels nice, and easy to sing! Well, it's
another temperament... A what?

-Cameron Bobro

Just to be certain, you are talking about the 1807 version, right?

*

Thomas Young well temperament (1807), also Luigi Malerbi nr.2 (1794)
|
0: 1/1 0.000 unison, perfect prime
1: 256/243 90.225 limma, Pythagorean minor second
2: 196.090 cents 196.090
3: 32/27 294.135 Pythagorean minor third
4: 392.180 cents 392.180
5: 4/3 498.045 perfect fourth
6: 1024/729 588.270 Pythagorean diminished fifth
7: 698.045 cents 698.045
8: 128/81 792.180 Pythagorean minor sixth
9: 894.135 cents 894.135
10: 16/9 996.090 Pythagorean minor seventh
11: 1090.225 cents 1090.225
12: 2/1 1200.000 octave

*

This little test seems to confirm that the only thing we can readily discern
is the difference between a conventional WT and 12-ET.

So much for the "WIII is the only temp. Bach ever cared for" argument... If
we can't identify the minute difference between one or the other with all
the equipment at our disposal, how do we expect a man of the Baroque age to?

Carl, let us up the ante. Pick 5 well-temperaments from among any you
please, arrange a Cantata of Bach according to these, and let's see how we
score.

Oz.

----- Original Message -----
From: "Carl Lumma" <clumma@yahoo.com>
To: <tuning@yahoogroups.com>
Sent: 17 Ocak 2007 �ar�amba 1:08
Subject: [tuning] Re: Hot seat

> > If Carl's 3 examples prove anything, it is that different WTs
> > sound more alike than either do to ET.
>
> It's true that both you and Ozan correctly identified the ET
> file, but with a test like this it probably isn't statistically
> significant. So it's hard to draw any real conclusions from
> the test, other than that the difference between these three
> tunings was not on the order of 'obvious' to those who listened.
>
> -Carl
>
>

--- In tuning@yahoogroups.com, "Ozan Yarman" <ozanyarman@...> wrote:
>
> Just to be certain, you are talking about the 1807 version, right?

Scala says, "Thomas Young well temperament (1807), also
Luigi Malerbi nr.2 (1794)."

> This little test seems to confirm that the only thing we can
> readily discern is the difference between a conventional WT
> and 12-ET.

It doesn't even tell us that.

> Carl, let us up the ante. Pick 5 well-temperaments from among
> any you please, arrange a Cantata of Bach according to these,
> and let's see how we score.

I'm afraid I don't have the time to do it in the foreseeable
future. However, I would greatly encourage all seekers of
the truth out there -- especially those who think expectation
bias is tomfoolery -- to follow this lead...

http://www.google.com/search?&q=abx+software

...do your own tests, and see if what you think you hear is
really what you hear.

Good luck!

-Carl

While I too enjoyed seeing you agree with my selection, this still doesn't let you off the hook. There IS a direct relationship between these MIDI scales and perceiving interval differences by cents, in that, the same piece is performed in each temperament and certain pitches differ. Wasn't it you who implied that you aimed for cent distinctions when playing the bassoon? Let us face it, the comparative evaluation showed us that we couldn't discern which WT was which. This gives me the impression that you might encounter difficulties in sustaining the WIII tuning throughout a concert. One might also wonder how much you deviate from it in reality...

Oz.

----- Original Message -----
From: Afmmjr@aol.com
To: tuning@yahoogroups.com
Sent: 17 Ocak 2007 Çarşamba 0:38
Subject: [tuning] Re: Hot seat

If Carl's 3 examples prove anything, it is that different WTs sound more alike than either do to ET.

Since WIII sounds different in each key, there is no exact way to pick up the tuning from a drop the needle approach. I gave my honest impressions from a general listen.

Ozan -- there is no direct relationship between these MIDI scales, and the production and apprehension of the movement (and direction) of single cents in a series. But I did enjoy agreeing with you on which was which.

best, Johnny

--- In tuning@yahoogroups.com, "Ozan Yarman" <ozanyarman@...> wrote:
> Carl, let us up the ante. Pick 5 well-temperaments from among any >
you please, arrange a Cantata of Bach according to these, and
> let's see how we score.

Ozan, you rock...this is a great idea! I'd do it right away if I had
time...I will gleefully participate in this if you or anyone else
pulls it off...hell...why not 1 minute of Bach in 10 WTs?

Best,
Aaron.

Hey all,

Another idea---it would be fun to post say 5 WT samples of the the
same piece as MIDI files---that way, people who object to the tester's
timbre choice(s) can use the timbre(s) that they are comfortable with.
We could also make such a test more challenging and less friendly to
bs by saying "here are 5 MIDI files in 5 temperaments...which is in
WerckIII." By leaving out the info on the other WTs, the listener is
left with purely using their ears, and not their knowledge of
theoretical qualities of the others.

Or, (possibly even harder?)---post a single file in 1 of 5 WTs and do
a multiple choice "which is it?", which would really weed out the
BS---tons of the information we glean about the correct answers in a
set is by comparison, and I think very little of it comes from the
inherent qualities of the WT itself. (It would show how little all the
talk of "etched counterpoint" really means). And in many cases, what
we understand the tuning to be depends on prior knowledge like the key
of the music, etc.

I think it would be interesting to extend this concept to
meantones---how much can we hear the difference between the p/q-comma
temperaments?

Anyway, this stuff is humbling to me. I love science. It cuts through
to the very heart of things in a very straight way!

Best,
Aaron.

Oz, why wonder? Listen to recordings. Besides Bach, there are videos of
Mozart and Haydn that I play on bassoon in Werckmeister on YouTube.
But really, I am more concerned by your insistences. WIII sounds like 12
different major keys and 12 different minor keys. Why does this translate to
you as being cent specific? Without knowing the key, why couldn’t a WIII
selection be many things.
As for Young tuning, I have never played in it. But this confusion seems
natural to me. Too bad, although we liked the Young, that is one of the more
unlikely tunings used by Bach. If I wanted to really analyze, perhaps,
hearing a sixth comma fifth as opposed to a quarter comma fifth. But I just
listened naturally.
Regarding Bach NOT being in meantone, I have included the following:
J. Murray Barbour (1897-1970), in his article “Bach and the Art of
Temperament,” clearly and emphatically demonstrated that the great majority of Bach’s
works require a keyboard tuning capable of playing in a full circle of keys,
a “circular temperament.”
So persistent is the legend that Bach still clung to meantone tuning for
organ that it will be well to cite examples of his freer practice. But,
although the great majority of Bach’s organ works contain notes excluded from the
meantone compass, the diehards are not convinced easily (Barbour, The Musical
Quarterly 33, 1946, p. 81).
Barbour gave numerous examples of works that exceeded the meantone compass.
Barbour wrote:
Conceivably in the third inversion of a dominant-seventh on D flat in an F
major Toccata, there would be room for considerable imperfections in the
tuning. But when Bach calmly modulated to ? sharp minor and made a cadence there,
or wrote a prelude and fugue in F minor, we can be sure he knew that his
organ was tuned so that these rare keys would be ‘possible’ ones (Barbour, MQ
33, 1946, p. 81).
After offering example upon example, Barbour concluded:
But why continue? The evidence is over-whelming that Bach could not have
used the meantone tuning for the organ (Barbour, MQ 33, p. 87).
best, Johnny

While I too enjoyed seeing you agree with my selection, this still doesn't
let you off the hook. There IS a direct relationship between these MIDI scales
and perceiving interval differences by cents, in that, the same piece is
performed in each temperament and certain pitches differ. Wasn't it you who
implied that you aimed for cent distinctions when playing the bassoon? Let us
face it, the comparative evaluation showed us that we couldn't discern which WT
was which. This gives me the impression that you might encounter difficulties
in sustaining the WIII tuning throughout a concert. One might also wonder
how much you deviate from it in reality...
Oz.

Aaron Krister Johnson wrote:
> Hey all,
> > Another idea---it would be fun to post say 5 WT samples of the the
> same piece as MIDI files---that way, people who object to the tester's
> timbre choice(s) can use the timbre(s) that they are comfortable with.
> We could also make such a test more challenging and less friendly to
> bs by saying "here are 5 MIDI files in 5 temperaments...which is in
> WerckIII." By leaving out the info on the other WTs, the listener is
> left with purely using their ears, and not their knowledge of
> theoretical qualities of the others.
> > Or, (possibly even harder?)---post a single file in 1 of 5 WTs and do
> a multiple choice "which is it?", which would really weed out the
> BS---tons of the information we glean about the correct answers in a
> set is by comparison, and I think very little of it comes from the
> inherent qualities of the WT itself. (It would show how little all the
> talk of "etched counterpoint" really means). And in many cases, what
> we understand the tuning to be depends on prior knowledge like the key
> of the music, etc.
> > I think it would be interesting to extend this concept to
> meantones---how much can we hear the difference between the p/q-comma
> temperaments? > > Anyway, this stuff is humbling to me. I love science. It cuts through
> to the very heart of things in a very straight way!

These are all very interesting ideas and I might try one or more of them at some point. Of course, it's likely that the choice may depend greatly on which piece of music is chosen for the comparison. Ultimately, the choice of a tuning for a particular piece depends as much on "whatever pleases the ear" as any other criteria. Whether or not you can recognize a particular tuning from hearing it isn't as critical as whether that tuning sounds right for that music.

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > I haven't listened to them yet, but harpsichord is a bad timbre
> > to use if you want to bring out the differences between tunings
> > rather than gloss them over.
>
> I disagree. Why do you think so? One of the two people to
> get my WT comparison test right last time it came around said
> the harpsichord piece in that test was crucial to his success.

Something with more sustained sounds, so you can hear the chords
better, would make more sense it seems to me.

--- In tuning@yahoogroups.com, "Aaron Krister Johnson" <aaron@...>
wrote:

> I think it would be interesting to extend this concept to
> meantones---how much can we hear the difference between the p/q-comma
> temperaments?

I did that a while back, and ended up finding a bug in Scala when the
103 edo version didn't sound as it was expected it would. The
differences between closely related meantones are subtle.

Of Ligeti's harpsichord pieces, the Passacaglia ungharese requires meantone, and is based entirely on a passacaglia theme of eight pure thirds in meantone, Hungarian Rock can be played in meantone or an unequal tuning, and Continuum and Aventures/Nouvelles Aventures are in 12tet. Unfortunately, on the Sony recording, the Passacaglia is played in ET and on a modern-style instrument. Time and money constraints in the recording studio _are_ a source of evil in the world.

Ligeti was very interested in using alternative tunings to find a path between atonality and functional harmony. Ramifications requires that half ogf the string ensemble be tuned a quartertone off, the Double Concerto and other pieces use quartertones within the micropolyphonic texture as well (Ligeti was not so much interested in a precise quartertone interval as in a marked otherness of pitch, or as a form of melodic diminution. and later works, including the Horn Trio and Hamburg Concerto use a variety of tuning approaches. I find that the Concerto uses the ensemble's horns and natural horns quite beautifully.

Several of Ligeti's students have a serious interest in alternative tunings, among them Manfred Stahnke and Wolfgang von Schweinitz.

Daniel Wolf

>
> "Dante Rosati" danterosati@gmail.com
> <mailto:danterosati@gmail.com?Subject=%20Re%3A%20Hot%20seat> > danterosati <http://profiles.yahoo.com/danterosati>
>
>
> Tue Jan 16, 2007 12:12 pm (PST)
>
> Hi Yaha-
>
> we were discussing Bach tunings, so of course what i meant was "every
> harpsichord recording of Bach". Probably Ligeti harpsichord pieces are
> performed in 12et.

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:

>
> I'm afraid I don't have the time to do it in the foreseeable
> future. However, I would greatly encourage all seekers of
> the truth out there -- especially those who think expectation
> bias is tomfoolery -- to follow this lead...
>
> http://www.google.com/search?&q=abx+software
>
> ...do your own tests, and see if what you think you hear is
> really what you hear.

Is Druidism hip once again? This whole debate is fraught with straw
men.

Expectation bias isn't tomfoolery, but you explaining what I said
about singing with WTs with it, is. You must have realized that I
already thought of that obvious explanation, so the order of the
cart and the horse should have been clear: first the experience,
naive and in ignorance, without expectation, then the conclusion.

Yes, folks, do by all means get ABX and spend a lot of time with it,
it'll teach you to scoff at cynics. :-)

-Cameron Bobro

--- In tuning@yahoogroups.com, Afmmjr@... wrote:
> although we liked the Young, that is one of the more unlikely
> tunings used by Bach.

In the 1801 paper Young didn't really seem too emphatic about whether
you should use equal temperament or his, or how to tune them, compared
with other scientific writers. A little bit included in a response to
criticism his article received from one of them:

"In Table XII, for Eb 83810 read Eb 84197 ; for .0011562 read .
0010116. In Fig. 53 the Eb (Q) is too near D, and the Eb (Y) should be
above instead of below it." (dated July 13, 1801; A Journal of Natural
Philosophy, Chemistry and the Arts, vol. 5, 1802. p167)

Clark

> > > I haven't listened to them yet, but harpsichord is a bad timbre
> > > to use if you want to bring out the differences between tunings
> > > rather than gloss them over.
> >
> > I disagree. Why do you think so? One of the two people to
> > get my WT comparison test right last time it came around said
> > the harpsichord piece in that test was crucial to his success.
>
> Something with more sustained sounds, so you can hear the chords
> better, would make more sense it seems to me.

In that case it was Woolsey's Wilde, which basically hammers
you over the head with C major for 3 minutes.

-Carl

--- In tuning@yahoogroups.com, "Gene Ward Smith" <genewardsmith@...>
wrote:
> I did that a while back, and ended up finding a bug in Scala when the
> 103 edo version didn't sound as it was expected it would. The
> differences between closely related meantones are subtle.

I think the conclusion was that there wasn't a bug, but that you used
a notation that didn't match the tuning, so you expected the wrong thing.

Manuel

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > I'll spill the beans: nobody has correctly identified the
> > three Bach files yet, though Cameron's second guess was correct.
>
> It might be worth saying that according to my spreadsheet
> anyway, Werck 3 has about the same total error as ET in
> A major, with Young a bit better. But the only difference
> I can hear in these files, if I'm not fooling myself, is to
> listen for the 12-cent difference in starting pitch between
> Werck 3 (first file) and ET (second file).
>
> -Carl

A major is a tricky one: almost all "well temperament" recipes give
that key an intonation very similar to ET.

There are always going to be two places round the circle where "well
temperament" intersects with ET - usually in the neighbourhood of A
major and Bb/Eb major. Doing a comparison there (with a thin-textured,
fast-moving piece) is not musically very worthwhile - except to
disprove claims that few-cent differences can always be heard ;> - and
indeed I didn't notice any differences I could put my finger on!

One might in principle notice the difference between the fifths in A
maj. if the piece included bare fifths: 1/12 vs 1/6 vs 1/4 comma. But
usually this is masked by all the other stuff going on.

C minor is a little more interesting... but this piece stays mainly in
the minor mode, for which differences of thirds are harder to hear, or
the relative major Eb, for which "WT" tuning is similar to ET.

Ab major, C# major and B major would be excellent tests. Of course,
the fuller the harmony and the slower the piece, the more likely we
are to differentiate between tunings of chords.

~~~T~~~

--- In tuning@yahoogroups.com, "Manuel Op de Coul"
<manuel.op.de.coul@...> wrote:
>
> --- In tuning@yahoogroups.com, "Gene Ward Smith" <genewardsmith@>
> wrote:

> > I did that a while back, and ended up finding a bug in Scala when
the
> > 103 edo version didn't sound as it was expected it would. The
> > differences between closely related meantones are subtle.
>
> I think the conclusion was that there wasn't a bug, but that you used
> a notation that didn't match the tuning, so you expected the wrong
thing.

Well, it was clear you didn't think it was a bug, but everyone else
seemed to. I should choose a different word, but I don't know one for
that nuance.

--- In tuning@yahoogroups.com, "Aaron Krister Johnson" <aaron@...>
wrote:
>
> --- In tuning@yahoogroups.com, "Ozan Yarman" <ozanyarman@> wrote:
> > Carl, let us up the ante. Pick 5 well-temperaments from among any
>
> you please, arrange a Cantata of Bach according to these, and
> > let's see how we score.
>
> Ozan, you rock...this is a great idea! I'd do it right away if I
had
> time...I will gleefully participate in this if you or anyone else
> pulls it off...hell...why not 1 minute of Bach in 10 WTs?
>
> Best,
> Aaron.

Hey, guys, It occurred to me that what we really need is a short
demonstration piece, specially written so that it contains *all* of
the major and minor triads, so you could hear the intonation change
from one part of the fifth-circle to another. A score would need to
be available so that you could follow the chord progressions.

I've already composed such a piece in my head (a few months back),
but I just haven't had the time (or sufficient incentive) to put it
into a midi file. If someone else would volunteer to add the pitch-
bends for various WT's (and would promise to include a couple of
mine), perhaps I can be persuaded to produce a 12-ET file (if I
haven't underestimated how long this might take).

If you think the timbre I used for the pajara comparisons is
desirable, I could then make mp3 files from the retuned midis.

We'd also need someone to volunteer to produce a score from the midi
file. (Actually, I might be able to do that without too much
trouble, since no microtonal symbols are required.)

--George

I shall excuse myself from this conversation now that I have exhausted all I have to say.

Oz.
----- Original Message -----
From: Afmmjr@aol.com
To: tuning@yahoogroups.com
Sent: 17 Ocak 2007 Çarşamba 5:35
Subject: re: [tuning] Re: Hot seat

Oz, why wonder? Listen to recordings. Besides Bach, there are videos of Mozart and Haydn that I play on bassoon in Werckmeister on YouTube.

But really, I am more concerned by your insistences. WIII sounds like 12 different major keys and 12 different minor keys. Why does this translate to you as being cent specific? Without knowing the key, why couldn’t a WIII selection be many things.

As for Young tuning, I have never played in it. But this confusion seems natural to me. Too bad, although we liked the Young, that is one of the more unlikely tunings used by Bach. If I wanted to really analyze, perhaps, hearing a sixth comma fifth as opposed to a quarter comma fifth. But I just listened naturally.

Regarding Bach NOT being in meantone, I have included the following:

J. Murray Barbour (1897-1970), in his article “Bach and the Art of Temperament,� clearly and emphatically demonstrated that the great majority of Bach’s works require a keyboard tuning capable of playing in a full circle of keys, a “circular temperament.�

So persistent is the legend that Bach still clung to meantone tuning for organ that it will be well to cite examples of his freer practice. But, although the great majority of Bach’s organ works contain notes excluded from the meantone compass, the diehards are not convinced easily (Barbour, The Musical Quarterly 33, 1946, p. 81).

Barbour gave numerous examples of works that exceeded the meantone compass. Barbour wrote:

Conceivably in the third inversion of a dominant-seventh on D flat in an F major Toccata, there would be room for considerable imperfections in the tuning. But when Bach calmly modulated to ? sharp minor and made a cadence there, or wrote a prelude and fugue in F minor, we can be sure he knew that his organ was tuned so that these rare keys would be ‘possible’ ones (Barbour, MQ 33, 1946, p. 81).

After offering example upon example, Barbour concluded:

But why continue? The evidence is over-whelming that Bach could not have used the meantone tuning for the organ (Barbour, MQ 33, p. 87).

best, Johnny

That is surely overkill at this stage. One step at a time would be a better
approach. Let us digest the new challange first.

Oz.

----- Original Message -----
From: "Aaron Krister Johnson" <aaron@dividebypi.com>
To: <tuning@yahoogroups.com>
Sent: 17 Ocak 2007 �ar�amba 4:18
Subject: [tuning] Re: Hot seat

> --- In tuning@yahoogroups.com, "Ozan Yarman" <ozanyarman@...> wrote:
> > Carl, let us up the ante. Pick 5 well-temperaments from among any >
> you please, arrange a Cantata of Bach according to these, and
> > let's see how we score.
>
> Ozan, you rock...this is a great idea! I'd do it right away if I had
> time...I will gleefully participate in this if you or anyone else
> pulls it off...hell...why not 1 minute of Bach in 10 WTs?
>
> Best,
> Aaron.
>
>

Hello all you tuners!

It has been interesting reading the fallout of Carl's "test." A real test would define things more than this.

However, when everyone fails a test, something is usually wrong with the test. (I don't believe it was intentional.)

In A major WIII comes out to the following in cents:

A+ WIII:
0 108 204 312 402 504 606 702 810 900 1008 1104

But Young, comes in 2 versions: Young I and Young II. Of course, we don't know what Carl chose. According to Barbour, and transposed to A major, we have:

A+ Young I :
0 106 198 306 400 502 604 698 806 898 1004 1102

A+ Young II:
0 102 196 306 396 502 600 698 804 894 1004 1098

Half of the Young II intervals are actually also WIII intervals, but in other keys. They are underlined.

Based on the speed of the MIDI, confusion between these two is more understandable. Without providing the key, the correct label of the Young tuning, non-involvement with Young tunings, and no perfect pitch, even besides what others have said about the tone and the speed, this is really no test at all.

best, Johnny

p.s. Ligetti's harpsichord solo in meantone is gorgeous. Manfred Stahnke sent me a copy. I don't know if it is commercially available, but I will ask him.

________________________________________________________________________
Check out the new AOL. Most comprehensive set of free safety and security tools, free access to millions of high-quality videos from across the web, free AOL Mail and more.

So, Carl's wasn't a "real" test because you along with everyone else failed to distinguish between WIII and any of the two Youngs?

Sad state of affairs IMNSHO, if not a botched up effort to cover up the discrepancy between the perceived and actual.

This is a recurring phenomenon among the Maqam Music elite too. Simply dreadful...

Oz.

----- Original Message -----
From: Afmmjr@aol.com
To: tuning@yahoogroups.com
Sent: 18 Ocak 2007 Perşembe 5:52
Subject: [tuning] Re: Hot seat

Hello all you tuners!

It has been interesting reading the fallout of Carl's "test." A real test would define things more than this.

However, when everyone fails a test, something is usually wrong with the test. (I don't believe it was intentional.)

In A major WIII comes out to the following in cents:

A+ WIII:
0 108 204 312 402 504 606 702 810 900 1008 1104

But Young, comes in 2 versions: Young I and Young II. Of course, we don't know what Carl chose. According to Barbour, and transposed to A major, we have:

A+ Young I :
0 106 198 306 400 502 604 698 806 898 1004 1102

A+ Young II:
0 102 196 306 396 502 600 698 804 894 1004 1098

Half of the Young II intervals are actually also WIII intervals, but in other keys. They are underlined.

Based on the speed of the MIDI, confusion between these two is more understandable. Without providing the key, the correct label of the Young tuning, non-involvement with Young tunings, and no perfect pitch, even besides what others have said about the tone and the speed, this is really no test at all.

best, Johnny

p.s. Ligetti's harpsichord solo in meantone is gorgeous. Manfred Stahnke sent me a copy. I don't know if it is commercially available, but I will ask him.

> In A major WIII comes out to the following in cents:
>
> A+ WIII:
> 0 108 204 312 402 504 606 702 810 900 1008 1104
>
> But Young, comes in 2 versions: Young I and Young II. Of
> course, we don't know what Carl chose.

Actually we do: I cited the exact version I used.

> Based on the speed of the MIDI, confusion between these two
> is more understandable. Without providing the key,

The key of the piece is A. The key of the tuning is given
in the Scala file I cited. Here it is again:

!
Thomas Young well temperament (1807).
12
!
256/243
196.09000
32/27
392.18000
4/3
1024/729
698.04500
128/81
894.13500
16/9
1090.22500
2/1
!

>the correct label of the Young tuning, non-involvement with
>Young tunings,

Non-involvement?

>and no perfect pitch,

??

-Carl

By the way, I only included the Young file to reduce
the chances of success with the other two (ET and WIII)
below 50%. These new tests with 10 files are bit
daunting to me.

-Carl

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> By the way, I only included the Young file to reduce
> the chances of success with the other two (ET and WIII)
> below 50%. These new tests with 10 files are bit
> daunting to me.
>
> -Carl

The number of tunings is intentional, though I did at least decide not
to use all 30-40 temperaments easily available to me (or even the 11th
that I rendered). The purpose of the distractors is precisely the same
as the ordinary people in a police identity parade.

Apologies if I made it a bit too daunting to you ;)

Peace,
Patty

--- In tuning@yahoogroups.com, Afmmjr@... wrote:

> But Young, comes in 2 versions: Young I and Young II. Of course, we
don't know what Carl chose. According to Barbour, and transposed to A
major, we have:
>
> A+ Young I :
> 0 106 198 306 400 502 604 698 806 898 1004 1102
>
> A+ Young II:
> 0 102 196 306 396 502 600 698 804 894 1004 1098
>

Oh, I forgot to quote the part from the 1801 letter in JNPC&A from
Young to Robison (who, he says, "justly observes, that different
persons differ exceedingly in their estimation of the effect of the
same temperament on different concords, and that much of this arises
from their different dispositions") that says "Errata in the paper on
sound and light."

Clark

Gene,

> Well, it was clear you didn't think it was a bug, but everyone else
> seemed to. I should choose a different word, but I don't know one for
> that nuance.

Feature? I'm not annoyed either but would like to get you to understand
how it works, because you probably didn't get it completely right.
Remember a while ago I asked you to make a change to your webpage
"Composing in Meantone", composing.htm? I like this page and think
it's very useful, but it's also slightly misleading. So now I've
changed it a bit myself, and will send it back to you in the hope that
you'll replace it on your website and read it carefully. And sure the
Scala help file could have improvement in this area too.

Manuel

--- In tuning@yahoogroups.com, "Manuel Op de Coul"
<manuel.op.de.coul@...> wrote:

> Remember a while ago I asked you to make a change to your webpage
> "Composing in Meantone", composing.htm?

It probably got buried under spam.

Hi George,

George D. Secor wrote:
>
[snip]
>
> Hey, guys, It occurred to me that what we really need is a short
demonstration piece, specially written so that it contains *all* of
the major and minor triads, so you could hear the intonation change
from one part of the fifth-circle to another. A score would need to
be available so that you could follow the chord progressions.
>
> I've already composed such a piece in my head (a few months back),
but I just haven't had the time (or sufficient incentive) to put it
into a midi file. If someone else would volunteer to add the pitch-
bends for various WT's (and would promise to include a couple of
mine), perhaps I can be persuaded to produce a 12-ET file (if I
haven't underestimated how long this might take).
>
> If you think the timbre I used for the pajara comparisons is
desirable, ...

Were they the ones I complained about? ;-)

> ... I could then make mp3 files from the retuned midis.
>
> We'd also need someone to volunteer to produce a score from the
midi file. (Actually, I might be able to do that without too much
trouble, since no microtonal symbols are required.)

Nothing easier, provided you don't need any subtleties of dynamics.
(Putting my hand up.)

Regards,
Yahya

--- In tuning@yahoogroups.com, "yahya_melb" <yahya@...> wrote:
>
> Hi George,
>
> George D. Secor wrote:
> >
> [snip]
> >
> > Hey, guys, It occurred to me that what we really need is a short
> demonstration piece, specially written so that it contains *all* of
> the major and minor triads, so you could hear the intonation change
> from one part of the fifth-circle to another. A score would need to
> be available so that you could follow the chord progressions.
> >
> >

That reminds me: Diapason Press do a publication with 'musical circles' :
http://diapason.xentonic.org/dp/dp001.html

due to Sorge, Heinichen and Kirnberger. I still don't know how to
order from them, though.

~~~T~~~

--- In tuning@yahoogroups.com, "yahya_melb" <yahya@...> wrote:
>
> Hi George,
>
> George D. Secor wrote:
> >
> [snip]
> >
> > Hey, guys, It occurred to me that what we really need is a short
> demonstration piece, specially written so that it contains *all* of
> the major and minor triads, so you could hear the intonation change
> from one part of the fifth-circle to another. A score would need
to
> be available so that you could follow the chord progressions.
> >
> > I've already composed such a piece in my head (a few months
back),
> but I just haven't had the time (or sufficient incentive) to put it
> into a midi file. If someone else would volunteer to add the pitch-
> bends for various WT's (and would promise to include a couple of
> mine),

I now think I should be able to do this fairly easily with Scala.

> perhaps I can be persuaded to produce a 12-ET file (if I
> haven't underestimated how long this might take).
> >
> > If you think the timbre I used for the pajara comparisons is
> desirable, ...
>
> Were they the ones I complained about? ;-)

Yes, but, as before, my purpose is to make the intonational
differences as obvious as possible. I'm almost half done with the
piece, and I now think that it might be best to use two or three
different timbres.

> > ... I could then make mp3 files from the retuned midis.
> >
> > We'd also need someone to volunteer to produce a score from the
> midi file. (Actually, I might be able to do that without too much
> trouble, since no microtonal symbols are required.)
>
> Nothing easier, provided you don't need any subtleties of
dynamics.
> (Putting my hand up.)

Okay, but it'll be at least a few days before I finish it, which is
just as well, since at the moment we seem to be "up to our ears" in
well-temperament examples for comparison.

--George

--- In tuning@yahoogroups.com, "yahya_melb" <yahya@...> wrote:
>
> Hi George,
>
> George D. Secor wrote:
> >
> [snip]
> >
> > Hey, guys, It occurred to me that what we really need is a short
> demonstration piece, specially written so that it contains *all* of
> the major and minor triads, so you could hear the intonation change
> from one part of the fifth-circle to another. A score would need
to
> be available so that you could follow the chord progressions.
> >
> > ... I could then make mp3 files from the retuned midis.
> >
> > We'd also need someone to volunteer to produce a score from the
> midi file. (Actually, I might be able to do that without too much
> trouble, since no microtonal symbols are required.)
>
> Nothing easier, provided you don't need any subtleties of
dynamics.
> (Putting my hand up.)
>
> Regards,
> Yahya

Thanks! Why don't you see if you can do something with this (it's on
4 tracks, but would be read most easily on 2 staves, if you could
manage that):

http://xenharmony.wikispaces.com/space/showimage/Circle.mid
(harpsichord patch)
http://xenharmony.wikispaces.com/space/showimage/Circle1.mid
(timbre I used for pajara comparisons)

It turned out to be more of a 4-part harmony exercise than a real
composition, which will better serve to illustrate the differences
between well-temperaments (once pitch-bends are added). The first
file has a harpsichord timbre, but I've included another file: in
case anyone wants to add pitch-bends for a particular WT and send the
file back to me, I'll make an mp3 from it (Circle1.mid) for inclusion
in the WT set I'll be submitting soon.

--George

> http://xenharmony.wikispaces.com/space/showimage/Circle.mid
> (harpsichord patch)
> http://xenharmony.wikispaces.com/space/showimage/Circle1.mid
> (timbre I used for pajara comparisons)
>
> It turned out to be more of a 4-part harmony exercise than a real
> composition, which will better serve to illustrate the differences
> between well-temperaments (once pitch-bends are added). The first
> file has a harpsichord timbre, but I've included another file: in
> case anyone wants to add pitch-bends for a particular WT and send
> the file back to me, I'll make an mp3 from it (Circle1.mid) for
> inclusion in the WT set I'll be submitting soon.

It would make a good ground. FWIW, the second timbre sounds
better on the "MS GS Wavetable Synth" on my machine.

-Carl

George D. Secor wrote:
[snip]
> > > Hey, guys, It occurred to me that what we really need is a
short demonstration piece, specially written so that it contains
*all* of the major and minor triads, so you could hear the intonation
change from one part of the fifth-circle to another. A score would
need to be available so that you could follow the chord progressions.
> > >
> > > ... I could then make mp3 files from the retuned midis.
> > >
> > > We'd also need someone to volunteer to produce a score from the
midi file. (Actually, I might be able to do that without too much
trouble, since no microtonal symbols are required.)
> >
> > Nothing easier, provided you don't need any subtleties of
dynamics.
> > (Putting my hand up.)
> >
> > Regards,
> > Yahya
>
> Thanks! Why don't you see if you can do something with this (it's
on 4 tracks, but would be read most easily on 2 staves, if you could
manage that):
>
> http://xenharmony.wikispaces.com/space/showimage/Circle.mid
> (harpsichord patch)

That's using the General MIDI patch number 7, Harpsichord.

> http://xenharmony.wikispaces.com/space/showimage/Circle1.mid
> (timbre I used for pajara comparisons)

This is the General MIDI patch number 31, Overdriven guitar.
Apparently the same notes are played, so one score should suffice.
I've posted the score, notated for harpsichord or SATB, on eSnips at:
http://preview.tinyurl.com/357t22

> It turned out to be more of a 4-part harmony exercise than a real
composition, which will better serve to illustrate the differences
between well-temperaments (once pitch-bends are added). The first
file has a harpsichord timbre, but I've included another file: in
case anyone wants to add pitch-bends for a particular WT and send the
file back to me, I'll make an mp3 from it (Circle1.mid) for inclusion
in the WT set I'll be submitting soon.

Regards,
Yahya

--- In tuning@yahoogroups.com, "George D. Secor" <gdsecor@...> wrote:
>
> --- In tuning@yahoogroups.com, "yahya_melb" <yahya@> wrote:
> > George D. Secor wrote:
> > >
> > [snip]> > ... We'd also need someone to volunteer to produce a
score from the
> > midi file.
> > ...
> > (Putting my hand up.)
> > ...

After making and listening to some retuned midi files, I now think
that a score won't be necessary. All you need to know is the chord
sequence, which is very easy to follow, since there are only 2 chords
in each measure (except for the last) and 4 measures in each line:

| D A | C G | Bb F | Em F# |
| Bm F#m | Am Em | Gm Dm | Gm F |
| Bb F | Ab Eb | Gb Db | Cm D |
| Gm Dm | Fm Cm | Ebm Bbm | Ebm Db |
| F# C# | E B | D A | G#m A# |
| D#m A#m | C#m G#m | Bm F#m | Bm A |
| D |

> http://xenharmony.wikispaces.com/space/showimage/Circle.mid
> (harpsichord patch)
> http://xenharmony.wikispaces.com/space/showimage/Circle1.mid
> (timbre I used for pajara comparisons)
>
> It turned out to be more of a 4-part harmony exercise than a real
> composition, which will better serve to illustrate the differences
> between well-temperaments (once pitch-bends are added). The first
> file has a harpsichord timbre, but I've included another file: in
> case anyone wants to add pitch-bends for a particular WT and send
the
> file back to me, I'll make an mp3 from it (Circle1.mid) for
inclusion
> in the WT set I'll be submitting soon.

Since I've found that this is so easy to do in Scala, I'll simply
take requests for WT's to be included in the examples. I already
have Werckmeister III, Valotti-Young, Kellner-Bach, Lehman-Bach, and
AKJ-GS-RWT and would appreciate other suggestions.

I've also been thinking of ways to deal with the "listening fatigue"
that I know many have experienced in this present "explosion" of WT
listening tests, and I advise the following:

1) The examples should not be overly long;
2) Listen to each example at least several times, over a period of at
least several days;
3) Once you have a general idea what a tuning sounds like, determine
specific things to listen for in specific measures;
4) Don't attempt to evaluate too many things in one pass.
5) Don't attempt to listen to too much in one session.

--George

George wrote...
> Valotti-Young,

Didn't these two give different key centers or something?

> Kellner-Bach, Lehman-Bach, and
> AKJ-GS-RWT and would appreciate other suggestions.

I'd like to see Barnes-Bach, Lumma-moh-ha-ha, Secor-VRWT,
and whichever you prefer of Secor-24a and Secor-24b.

Moh-ha-ha is:

!
Rational well temperament.
12
!
19/18
323/288
19/16
323/256
171/128
361/256
551/368
19/12
323/192
57/32
513/272
2/1
!

> I've also been thinking of ways to deal with the "listening
> fatigue"

My suggestion is to listen to three tunings a day and pick
a favorite each time, then iterate until you have a single
favorite, then seed it back in with two tunings that were
eliminated in the first round and see if you can identify it.
:)

-Carl

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:

> Moh-ha-ha is:

I didn't have this already in my Scala scl directory. Why is it moh-ha-
ha?

I don't see much bang for the buck in this--there's not much of a sweet
spot, and for some reason it's located at A major.

> > Moh-ha-ha is:
>
> I didn't have this already in my Scala scl directory. Why is
> it moh-ha-ha?

Simply the first thing that popped into my head when I first
saved the file.

> I don't see much bang for the buck in this--there's not much of
> a sweet spot, and for some reason it's located at A major.

I do that because A is a good key to play the halberstadt in.
And in this case, it looks like, to put the worst 3rd on C#.

The scale passes the 'at least as good on average as ET' test,
and has lots of 19-limit intervals. This mode makes it more
obvious how I constructed it:

17/16
9/8
19/16
29/23
4/3
17/12
3/2
27/17
32/19
16/9
17/9
2/1

Everything except 29/23 and 27/17 is a 19-limit primary
interval.

-Carl

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> George wrote...
> > Valotti-Young,
>
> Didn't these two give different key centers or something?

You could say that the best keys are centered halfway between C and G.

> > Kellner-Bach, Lehman-Bach, and
> > AKJ-GS-RWT and would appreciate other suggestions.
>
> I'd like to see Barnes-Bach,

Okay, found it!

> Lumma-moh-ha-ha,

Sounds like a belly laugh. ;-) This one should be easy to pick out
from all the others.

> Secor-VRWT,

That one's a must; it's Ellis #2 modified to create proportional
beating in 9 major and 6 minor triads. There are no just 5ths (and
therefore no pythagorean triads) in this one.

> and whichever you prefer of Secor-24a and Secor-24b.

I'll use 24a, since it'll be easier to distinguish from everything
else. And I'll want to know if anyone agrees with me about what I
think are its two types of flaws.

I was also going to throw in one or two of my other tunings --
including one that's new & improved -- as of yesterday.

> ...
> > I've also been thinking of ways to deal with the "listening
> > fatigue"
>
> My suggestion is to listen to three tunings a day and pick
> a favorite each time, then iterate until you have a single
> favorite, then seed it back in with two tunings that were
> eliminated in the first round and see if you can identify it.
> :)

After listening to around 20 different WT's using my test chord
progressions, I've come to the conclusion that a blind test isn't
necessary the most productive way to compare WT's. I prefer first to
look at the numbers in order to spot the significant differences
between tunings. Then I'll listen closely to those particular points
in the progressions where those differences are expected to show up,
to see how that compares with my expectations. A lot of these chords
on the far side of the WT circle are pythagorean, so those won't
differ at all. Most of the differences (and flaws) are on the near
side, and you're liable to miss them if you're not prepared in
advance.

If you would still prefer a blind test, once I'm ready, I could put
the sound files out there a few at a time without identifying the
tunings, except that I would identify the places that, judging by the
numbers, will probably contain significant differences, but I won't
reveal which file has the "better" numbers in which place. You could
first listen to them before reading my comments, and then listen
again for the specific places I mention. (This procedure could even
involve including the same file in different sessions.)

I think that both Tom & Brad have stated that the ideal way to test
tunings is to listen as you play on a keyboard, and I completely
agree. What I'm trying to do is to figure out the next best thing.

--George

> > > Valotti-Young,
> >
> > Didn't these two give different key centers or something?
>
> You could say that the best keys are centered halfway
> between C and G.

I mean, aren't Valotti and Young different scales, in the
sanse that they are intended to be mapped to the keyboard
differently?

> > and whichever you prefer of Secor-24a and Secor-24b.
>
> I'll use 24a, since it'll be easier to distinguish from everything
> else. And I'll want to know if anyone agrees with me about what I
> think are its two types of flaws.
>
> I was also going to throw in one or two of my other tunings --
> including one that's new & improved -- as of yesterday.

Great. Say, is 24a or b closer to ET (or are they both about
the same in this respect)?

> > ...
> > > I've also been thinking of ways to deal with the "listening
> > > fatigue"
> >
> > My suggestion is to listen to three tunings a day and pick
> > a favorite each time, then iterate until you have a single
> > favorite, then seed it back in with two tunings that were
> > eliminated in the first round and see if you can identify it.
//
> If you would still prefer a blind test, once I'm ready, I could put
> the sound files out there a few at a time without identifying the
> tunings, except that I would identify the places that, judging by
> the numbers, will probably contain significant differences, but I
> won't reveal which file has the "better" numbers in which place.
> You could first listen to them before reading my comments, and
> then listen again for the specific places I mention. (This
> procedure could even involve including the same file in different
> sessions.)

I'd like to get all the files at once, numbered 1-20 (or
whatever) and listen piecemeal (by threes) on my own. Then
when I'm ready, learn about the points of interest to listen
for. And finally, when I'm ready, learn the names of the
tunings. This could be done by putting text files in the
zip package: one with the points of interest and another
with the answers (names of the tunings).

Whaddya say?

-Carl

A remark on (Bach) instruments and players again...

I've noticed on some recordings that meantone has a pretty different
effect on Italian versus Flemish harpsichords - somehow much brighter
and livelier on the Italians. Of course it's something to do with the
partial spectrum, but I can't tell exactly what.

On the issue of performance style and articulation, how quick the
damping is has a strong influence. Pianos (and most electronic
instruments) have virtually instantaneous damping, whereas many
harpsichords (or clavichords) take a certain fraction of a second, or
longer, for the tail of the sound to disappear after the finger has
been withdrawn.

I should take this opportunity to be more precise about what Lindley
meant, and played, with the Ab Prelude: on the piano, he aims to hold
the chords for no more or less than a quarter note. The idea is that
they are effectively staccato half-notes, and give an impression of
staccato because they only occupy 1 beat of 3 in the measure.
Certainly not whipping the hands away as quickly as possible, and I
regret it if people got that impression.

One point I think John Barnes made back in 1979 is that the G major
Prelude, when played fast on a harpsichord, effectively creates chords
due to the 'tail' on the sound overlapping from one note to another -
even if the player doesn't attempt to use legato. This was one of the
caveats he admitted for his method of counting (chordal) thirds, and
is a possible pitfall of electronic simulations that cut off the sound
immediately at the official end of a note - let alone those that fade
into silence *before* the end!

~~~T~~~

--- In tuning@yahoogroups.com, "George D. Secor" <gdsecor@...> wrote:
>
> I think that both Tom & Brad have stated that the ideal way to test
> tunings is to listen as you play on a keyboard, and I completely
> agree. What I'm trying to do is to figure out the next best thing.
>
> --George
>

--- In tuning@yahoogroups.com, "Carl Lumma" <clumma@...> wrote:
>
> > > > Valotti-Young,
> > >
> > > Didn't these two give different key centers or something?
> >
> > You could say that the best keys are centered halfway
> > between C and G.
>
> I mean, aren't Valotti and Young different scales, in the
> sanse that they are intended to be mapped to the keyboard
> differently?

My understanding is that Valotti-Young and Young's #2 are the same
tuning. If they were intended to be mapped to different starting
keys, then that's something I wasn't aware of. (Lately I haven't had
time to read each & every one of the messages on this list, so I
could have missed that.)

> > > and whichever you prefer of Secor-24a and Secor-24b.
> >
> > I'll use 24a, since it'll be easier to distinguish from
everything
> > else. And I'll want to know if anyone agrees with me about what
I
> > think are its two types of flaws.

Now that I think about it, there are actually 3 types of flaws in
24a. Of course, these are IMO, and you might not agree.

(Here, when I'm using the word "you", I'm addressing anyone who
happens to be reading this -- not just Carl.)

> > I was also going to throw in one or two of my other tunings --
> > including one that's new & improved -- as of yesterday.
>
> Great. Say, is 24a or b closer to ET (or are they both about
> the same in this respect)?

24b is significantly closer to ET (and it overcomes 2-1/2 out of the
3 types of flaws of 24a), but both are significantly farther from ET
than most of the WT's in my list. If proportional beating is the
only thing important to you, then the numbers in 24a look a bit
prettier than 24b (particularly in the minor triads); however, looks
aren't everything, as a listening test will indicate.

At the risk of totally wearing your ears out, I suppose that I should
include both 24a & 24b (and a couple more of my own tunings than I
had previously intended, for reasons given below), since you seem to
have your heart set on finding the "best" WT (or WT's) -- or at least
eliminating those that, for one good reason or other, could be
excluded in favor of certain others.

I think that some of the differences between WT's will not be a
matter of good vs. not so good, but rather how well does the WT meet
its intended (or your desired) objective(s) -- so it's possible to
have multiple "bests."

> > > ...
> > > > I've also been thinking of ways to deal with the "listening
> > > > fatigue"
> > >
> > > My suggestion is to listen to three tunings a day and pick
> > > a favorite each time, then iterate until you have a single
> > > favorite, then seed it back in with two tunings that were
> > > eliminated in the first round and see if you can identify it.
> //
> > If you would still prefer a blind test, once I'm ready, I could
put
> > the sound files out there a few at a time without identifying the
> > tunings, except that I would identify the places that, judging by
> > the numbers, will probably contain significant differences,

Hereafter, I'll refer to these as "listening points".

> > but I
> > won't reveal which file has the "better" numbers in which place.
> > You could first listen to them before reading my comments, and
> > then listen again for the specific places I mention. (This
> > procedure could even involve including the same file in different
> > sessions.)
>
> I'd like to get all the files at once, numbered 1-20 (or
> whatever) and listen piecemeal (by threes) on my own. Then
> when I'm ready, learn about the points of interest to listen
> for. And finally, when I'm ready, learn the names of the
> tunings. This could be done by putting text files in the
> zip package: one with the points of interest and another
> with the answers (names of the tunings).
>
> Whaddya say?

Okay, how about this? I'll begin by releasing all of the files I
have at once (without identifying the tunings), along with a listing
of the chord progressions (same for each file). You'll have the
opportunity to identify your own points of interest and make your own
comments.

After a day or two, I'll make a list of some listening points,
without singling out any particular files. Others could add to
these, as they see fit.

After that, I could single out specific files to compare in
connection with the various listening points (without identifying the
tunings), and your comments ^

Don't worry about identifying the tunings -- we're not trying to
determine *who* has the keenest ears, but rather *which tunings* are
among the "best", and for what reasons. If we happen to agree to
exclude any along the way, then maybe I'll identify those when that
occurs.

I expect that at least 1/3 of the tunings will be ones I've designed,
for the following reasons:

1) I've designed literally dozens of WT's (and subtle variations
thereof), and I've already narrowed my choices down to the very best
ones (based on both listening comparisons & numerical evaluation of
interval errors and beat interactions within triads). Any further
exclusions would depend on subjective "judgment calls" regarding how
much (and/or what type of) error or key contrast is acceptable and/or
desirable, for which I might not reach the same conclusions as others.

2) I've made a deliberate effort to avoid certain flaws in most of my
WT's, so differences in key contrast (resulting from variations in
the distribution of parts of a comma) can be compared between two of
these WT's without having to make allowances for such flaws.

3) If both WT's in a particular comparison have already been
identified as mine, then my feelings won't be hurt if you happen to
prefer one vs. the other.

4) I'd appreciate the opportunity to get some specific feedback --
it's a good way to learn.

--George

> > I mean, aren't Valotti and Young different scales, in the
> > sanse that they are intended to be mapped to the keyboard
> > differently?
>
> My understanding is that Valotti-Young and Young's #2 are the same
> tuning. If they were intended to be mapped to different starting
> keys, then that's something I wasn't aware of. (Lately I haven't
> had time to read each & every one of the messages on this list, so
> I could have missed that.)

Oh, I didn't know there was a scale "Valotti-Young".

> At the risk of totally wearing your ears out, I suppose that
> I should include both 24a & 24b

I'll be listening by 3s, so no worries there.

> > I'd like to get all the files at once, numbered 1-20 (or
> > whatever) and listen piecemeal (by threes) on my own. Then
> > when I'm ready, learn about the points of interest to listen
> > for. And finally, when I'm ready, learn the names of the
> > tunings. This could be done by putting text files in the
> > zip package: one with the points of interest and another
> > with the answers (names of the tunings).
> >
> > Whaddya say?
>
> Okay, how about this? I'll begin by releasing all of the
> files I have at once (without identifying the tunings),
> along with a listing of the chord progressions (same for each
> file). You'll have the opportunity to identify your own
> points of interest and make your own comments.
>
> After a day or two, I'll make a list of some listening points,
> without singling out any particular files. Others could add to
> these, as they see fit.
>
> After that, I could single out specific files to compare in
> connection with the various listening points (without
> identifying the tunings), and your comments ^

That would be awesome!

> Don't worry about identifying the tunings

Don't worry, I won't!

-Carl

--- In tuning@yahoogroups.com, "George D. Secor" <gdsecor@...> wrote:
>
> > > > > Valotti-Young,
> > > >
> >
> > I mean, aren't Valotti and Young different scales, in the
> > sanse that they are intended to be mapped to the keyboard
> > differently?
>
> My understanding is that Valotti-Young and Young's #2 are the same
> tuning. If they were intended to be mapped to different starting
> keys, then that's something I wasn't aware of.

It would be simpler to remove the term 'Vallotti-Young' entirely. It
is historically meaningless, since the two gentlemen never met and
described slightly different tunings.

Vallotti has the six pure fifths from B through F, Young #2 has them
from F# through C.

Note that Young intended this to represent the common practice of past
(English) tuners; he actually proposed #1 instead.

> Now that I think about it, there are actually 3 types of flaws in
> 24a. Of course, these are IMO, and you might not agree

> 2) I've made a deliberate effort to avoid certain flaws in most of my
> WT's, so differences in key contrast (resulting from variations in
> the distribution of parts of a comma) can be compared between two of
> these WT's without having to make allowances for such flaws.

It would be interesting to know what George considers to be 'flaws',
and how he goes about eliminating them. Open up the black box!

~~~T~~~

--- In tuning@yahoogroups.com, "Tom Dent" <stringph@...> wrote:
>
> --- In tuning@yahoogroups.com, "George D. Secor" <gdsecor@> wrote:
> >
> > > > > > Valotti-Young,
> > > > >
> > >
> > > I mean, aren't Valotti and Young different scales, in the
> > > sanse that they are intended to be mapped to the keyboard
> > > differently?
> >
> > My understanding is that Valotti-Young and Young's #2 are the
same
> > tuning. If they were intended to be mapped to different
starting
> > keys, then that's something I wasn't aware of.
>
> It would be simpler to remove the term 'Vallotti-Young' entirely. It
> is historically meaningless, since the two gentlemen never met and
> described slightly different tunings.
>
> Vallotti has the six pure fifths from B through F, Young #2 has them
> from F# through C.

Thanks, Tom, for clarifying that.

> Note that Young intended this to represent the common practice of
past
> (English) tuners; he actually proposed #1 instead.
>
> > Now that I think about it, there are actually 3 types of flaws in
> > 24a. Of course, these are IMO, and you might not agree
>
> > 2) I've made a deliberate effort to avoid certain flaws in most
of my
> > WT's, so differences in key contrast (resulting from variations
in
> > the distribution of parts of a comma) can be compared between two
of
> > these WT's without having to make allowances for such flaws.
>
> It would be interesting to know what George considers to be 'flaws',
> and how he goes about eliminating them.

One flaw is failure to observe harmonic balance, e.g., excessive
favoring of triads in either the sharp or flat direction over the
other. This may be influenced by whether or not the error of the
minor 3rd (or major 6th) has been taken into account, per my
observation that beating major 6ths are just as prominent as beating
major 3rds. I would judge Young #2 to be superior to Valotti in this
respect, but I recognize that this is open to debate.

Another flaw would be excessive tempering of one or more fifths,
which IMO would be more than 1/4-comma. This is one of the flaws in
my 24a, which has two fifths narrower than 6.3 cents. My 24b has one
fifth ~6.1 cents narrow, and all the rest have <4.3 cents error, so I
would consider that an improvement. As to whether proportional-
beating triads are worth the trouble (or the lack thereof should be a
flaw) is, of course, a matter of conjecture; I'd say that they're of
more significance in the major than the minor triads, and of much
less significance in triads with rapidly beating 3rds.

Yet another flaw would be fifths that zig-zag considerably in size as
one goes around the circle.

In brief, a flaw is anything that causes one or more triads to "stick
out" or to have a sound that could be judged inappropriate, relative
to those adjacent in the circle of fifths.

> Open up the black box!

All in good time -- I still need more time to get all of the files
made.

--George

--- In tuning@yahoogroups.com, "George D. Secor" <gdsecor@...> wrote:

> Another flaw would be excessive tempering of one or more fifths,
> which IMO would be more than 1/4-comma.

> Yet another flaw would be fifths that zig-zag considerably in size as
> one goes around the circle.

Grail has both properties to an extreme degree, and yet people
generally had a positive response to it in my test. Is it time to give
what the test tunings were? I haven't had any new activity with it in a
while.

--- In tuning@yahoogroups.com, "George D. Secor" <gdsecor@...> wrote:
>
>
> One flaw is failure to observe harmonic balance, e.g., excessive
> favoring of triads in either the sharp or flat direction over the
> other. This may be influenced by whether or not the error of the
> minor 3rd (or major 6th) has been taken into account, per my
> observation that beating major 6ths are just as prominent as beating
> major 3rds. I would judge Young #2 to be superior to Valotti in this
> respect, but I recognize that this is open to debate.

OK ... though I would say that Young 2 favours A major as compared to
Eb major. Perhaps Young recognized that many 'ordinaire'-like tunings
did favour the sharps. (Which you might want for playing the dominants
of relative minor keys.)

> Another flaw would be excessive tempering of one or more fifths,
> which IMO would be more than 1/4-comma.

This makes sense - particularly if the major third on the same root is
not pure. One of my objections to the 'Werckmeister' formulae is that
they have (major) triads where the fifth is 1/4 comma flat and the
third is 1/2 or even 3/4 comma sharp. It sounds, how shall I put it,
conflict-ridden. Although the minor third is relatively bad, that
doesn't quite account for the level of aural distress.

I can only accept such flat fifths if the third is virtually pure.

> As to whether proportional-
> beating triads are worth the trouble (or the lack thereof should be a
> flaw) is, of course, a matter of conjecture

I wouldn't take it myself (i.e. the trouble) - I think there were some
aural tests that showed synchronized beating was actually obtrusive in
most listeners' estimation.

> Yet another flaw would be fifths that zig-zag considerably in size as
> one goes around the circle.

This connects up with my 'Werckmeister' critique, since if this
happens, you'll get a combination of an impure third and a fifth which
is unnecessarily out-of-tune given the size of the third.

> In brief, a flaw is anything that causes one or more triads to "stick
> out" or to have a sound that could be judged inappropriate, relative
> to those adjacent in the circle of fifths.

And it's good to have a definite idea of what you think could be
inappropriate!

OK, I thought of this one quite recently, a mash-up of Youngy and
Neidhardty features:

C -2 G -2 D -2 A -2 E -1 B -1 F# -1 C# 0 G# 0 D# 0 Bb 0 F -1 C

usual 1/12 PC units.
~~~T~~~

Hi all,

[George D. Secor]:
> > One flaw is failure to observe harmonic balance, e.g., excessive
favoring of triads in either the sharp or flat direction over the
other. This may be influenced by whether or not the error of the
minor 3rd (or major 6th) has been taken into account, per my
observation that beating major 6ths are just as prominent as beating
major 3rds. I would judge Young #2 to be superior to Valotti in this
respect, but I recognize that this is open to debate.
>
[Tom Dent]:
> OK ... though I would say that Young 2 favours A major as compared
to Eb major. Perhaps Young recognized that many 'ordinaire'-like
tunings did favour the sharps. (Which you might want for playing the
dominants of relative minor keys.)

[Yahya]:
Eg, from C major to A minor to E major; from 0 to 4 sharps. I see
your point.

[George D. Secor]:
> > Another flaw would be excessive tempering of one or more fifths,
which IMO would be more than 1/4-comma.

[Yahya]:
Others might draw the line more narrowly, eg at 1/3 or 2/7 comma.
For music at a moderate tempo, I have found 2/7-comma tempered fifths
quite unobjectionable.

[Tom Dent]:
> This makes sense - particularly if the major third on the same root
is not pure. One of my objections to the 'Werckmeister' formulae is
that they have (major) triads where the fifth is 1/4 comma flat and
the third is 1/2 or even 3/4 comma sharp. It sounds, how shall I put
it, conflict-ridden. Although the minor third is relatively bad, that
doesn't quite account for the level of aural distress.
>
> I can only accept such flat fifths if the third is virtually pure.

[Yahya]:
Presumably, Tom, you mean the major third? For my part, if the minor
third between the third and fifth is awful, it won't wash. So an
excessively flat fifth is more bearable when the damage is spread
over its two constituent thirds.

[George D. Secor]:
> > As to whether proportional- beating triads are worth the trouble
(or the lack thereof should be a flaw) is, of course, a matter of
conjecture
>
[Tom Dent]:
> I wouldn't take it myself (i.e. the trouble) - I think there were
some aural tests that showed synchronized beating was actually
obtrusive in most listeners' estimation.

[Yahya]:
Can you point us to a reference for this? And didn't we have a few
proportional-beating temperaments onlist just a while ago (George,
Gene)? If so, it would be worth including some in our current round
of listening tests, so that list members can evaluate the effect of
synchronised beating for themselves.

Of course, any tuning whose proportional beats can only be heard in
notes held for a quarter-second or more will need to be represented
by music slow enough for that effect to be noticed. I think beats
need two full cycles or more to be heard clearly. For example, a
tuning with an interval that beats two times a second would want
notes of at least a second long, eg a minim (half note) in common
time at 120bpm. This suggests that our test pieces need to be slow,
with chords (preferably dyads) held for at least a second, and with
all representative dyads of the tuning so represented.

[George D. Secor]:
> > Yet another flaw would be fifths that zig-zag considerably in
size as one goes around the circle.

[Tom Dent]:
> This connects up with my 'Werckmeister' critique, since if this
happens, you'll get a combination of an impure third and a fifth
which is unnecessarily out-of-tune given the size of the third.

[George D. Secor]:
> > In brief, a flaw is anything that causes one or more triads
to "stick out" or to have a sound that could be judged inappropriate,
relative to those adjacent in the circle of fifths.

[Yahya]:
Seems a reasonable criterion.

[Tom Dent]:
> And it's good to have a definite idea of what you think could be
inappropriate!

[Yahya]:
Indeed! It's always so much better when we can sharpen our vague
notions into definite and testable sensa.

[Tom Dent]:
> OK, I thought of this one quite recently, a mash-up of Youngy and
Neidhardty features:
>
> C -2 G -2 D -2 A -2 E -1 B -1 F# -1 C# 0 G# 0 D# 0 Bb 0 F -1 C
>
> usual 1/12 PC units.
> ~~~T~~~

[Yahya]:
OK, Tom, would you like to make us a sample piece in this Denty
tuning? ;-)

Regards,
Yahya

--- In tuning@yahoogroups.com, "Tom Dent" <stringph@...> wrote:
>
> --- In tuning@yahoogroups.com, "George D. Secor" <gdsecor@> wrote:
> >
> > One flaw is failure to observe harmonic balance, e.g., excessive
> > favoring of triads in either the sharp or flat direction over the
> > other. This may be influenced by whether or not the error of the
> > minor 3rd (or major 6th) has been taken into account, per my
> > observation that beating major 6ths are just as prominent as
beating
> > major 3rds. I would judge Young #2 to be superior to Valotti in
this
> > respect, but I recognize that this is open to debate.
>
> OK ... though I would say that Young 2 favours A major as compared
to
> Eb major.

Yes, very definitely.

> Perhaps Young recognized that many 'ordinaire'-like tunings
> did favour the sharps. (Which you might want for playing the
dominants
> of relative minor keys.)

I don't consider a dissonant dominant in a minor key to be a defect,
but rather an opportunity to take advantage of its effective
resolution to a more consonant tonic chord (a kind of expressive
intonation). So that isn't my reason for favoring the sharp keys.

Rather, it's that I would not want any of the 8 good major triads in
the Eb-to-G# 1/4-comma meantone gamut to be significantly more
dissonant than 12-ET (and most of them significantly better), which
means that I would expect the 4 worst ones to be B, F#, C#, and G#
major. (And, yes, this would be especially applicable to
an 'ordinaire'.)

> > Another flaw would be excessive tempering of one or more fifths,
> > which IMO would be more than 1/4-comma.
>
> This makes sense - particularly if the major third on the same root
is
> not pure. One of my objections to the 'Werckmeister' formulae is
that
> they have (major) triads where the fifth is 1/4 comma flat and the
> third is 1/2 or even 3/4 comma sharp. It sounds, how shall I put it,
> conflict-ridden. Although the minor third is relatively bad, that
> doesn't quite account for the level of aural distress.

Yes, indeed! I hadn't intended to bad-mouth WkIII before the
listening test, in which the G major triad (which should be one of
the best) is clearly worse than F. G is also worse than Bb, which
(although less obvious than the contrast with F) is readily confirmed
by looking at the numbers. So this is also a case of bad harmonic
balance, or two flaws for the price of one.

> I can only accept such flat fifths if the third is virtually pure.
>
> > As to whether proportional-
> > beating triads are worth the trouble (or the lack thereof should
be a
> > flaw) is, of course, a matter of conjecture
>
> I wouldn't take it myself (i.e. the trouble) - I think there were
some
> aural tests that showed synchronized beating was actually obtrusive
in
> most listeners' estimation.

I remember that, and I made a point to stay out the discussion,
because I didn't think that the testing was extensive enough to
validate that conclusion. The purpose of proportional beating is
twofold:

1) To avoid an undesirable interaction that's sometimes encountered
between 5ths & 3rds that have slow beat rates that are not
proportional (this is particularly noticeable when the 5th is
tempered by more than 3 cents);

2) To mask the beating of a major 3rd by giving it the same beat rate
as (or multiple of) the 5th.

An example this 2nd point is that a 5/23-comma-meantone triad (with
M3rd 2.8 cents wide) will sound as consonant as a 1/4-comma-meantone
one (with just M3rd), due to its slower-beating 5th (4.68c vs.5.38c
narrow; note that both triads in this example are proportional-
beating). This is of significance inasmuch as I happen to recall
that another conclusion of that brief test was that the preferred
triad was generally the one in which the 5th had a slower beat rate.
That 0.7-cents-per-fifth saving gives you at least a couple more
cents to play with on the far side of the WT circle.

> > Yet another flaw would be fifths that zig-zag considerably in
size as
> > one goes around the circle.
>
> This connects up with my 'Werckmeister' critique, since if this
> happens, you'll get a combination of an impure third and a fifth
which
> is unnecessarily out-of-tune given the size of the third.

Yep! Poor Wreckmeister! (typo intended ;-)

> > In brief, a flaw is anything that causes one or more triads
to "stick
> > out" or to have a sound that could be judged inappropriate,
relative
> > to those adjacent in the circle of fifths.
>
> And it's good to have a definite idea of what you think could be
> inappropriate!
>
> OK, I thought of this one quite recently, a mash-up of Youngy and
> Neidhardty features:
>
> C -2 G -2 D -2 A -2 E -1 B -1 F# -1 C# 0 G# 0 D# 0 Bb 0 F -1 C
>
> usual 1/12 PC units.
> ~~~T~~~

Can't say that I see anything grossly inappropriate there. It is a
bit unusual in that you've favored the flat keys on the near side of
the circle (taking the minor 3rds into account) and the sharp keys on
the far side, but only very slightly. You might be surprised to
learn that most of your triads (both major & minor) come very close
to being proportional-beating. This would probably be a good one to
include in the listening test.

Golly, you know what? The brats are very similar to this one:

! GS2_11WT.scl
!
George Secor's rational 2/11-comma well-temperament
12
!
560/531
2643/2360
70/59
74/59
315/236
2240/1593
883/590
280/177
890/531
105/59
443/236
2/1

BTW, do you happen to know whether anyone (other than myself) ever
advocated the following WT: Make the fifths from C to B narrow 1/5-
Didymus comma and just from C to Gb; B-F# gets the leftover (~1.9
cents narrow). This one also gets most of the triads (both major &
minor) very close to being proportional-beating. (I also have some
rational versions of this.)

--George