temperament shootout!

Yeehah! Welcome one and all to the first
annual Temperament Shootout!

It's five short pieces of music...

() Bach's A-Maj Fugue from the WTC, Book II.
() Byrd's "Wolsey's Wilde".
() My own "Clavichordius Terrificus" (1997).
() My "Guys Dressed Like Girls (1997).
() Izaac's "Rain".

...each in five different 12-tone tunings...

() 12-tone Equal Temperament
() Meantone
() 12-tone Pythagorean Just Intonation
() Werckmeister III
() Young

The Bach, Byrd, and Izaac midi files were
obtained on the web. Credit to Yo Tomita and
and John Sankey for the first two, resp. My
stuff was entered into Encore with a mouse and
exported to midi. All files were stripped of
continuous controllers, re-channeled and
re-timbred in Cakewalk Pro Audio for this demo.

The pieces were retuned with Scala 2.05 with
"example/midi/mts". The Scala files used are
available in a zip archive at...

http://lumma.org/shootout/scales.zip

[Note that the 12-tET tunings were still sent
through this process, to minimize weirdness.]

The resulting midi files were rendered direct
to 16-bit mono wave at 44.1KHz. with Audio
Compositor 4.5. This is *fantastic* software,
and the author, Scott Mitchell, should be
praised for making it, and for fixing a bug so
this project could be completed! AC is $40...

http://home.att.net/~audiocompositor/

The pieces are rendered with patches from the
freely-available "Personal Copy v5r2" soundfont
library, with the exception of the Izaac piece,
which was done with the (also free) "NS Piano"
from NaturalStudio.

The wave files were normalized and adjusted
for DC bias (an AC bug?) in Cool Edit Pro 1.2
and then converted to 128Kbps CBR mp3 with
LAME 3.93.

The five pieces are available per-tuning in
zip archives here...

http://lumma.org/shootout/tuning1.zip
http://lumma.org/shootout/tuning2.zip
http://lumma.org/shootout/tuning3.zip
http://lumma.org/shootout/tuning4.zip
http://lumma.org/shootout/tuning5.zip

[These files are 4.6MB each. You don't have
to download them all to participate. My server
should be fine, but please let me know if you
get any errors.

Your mission, should you choose to accept it,
is to guess the tunings. All you golden-ear
tuners out there give it a shot! This must be
one of the most accurate recorded historical
temperament comparos ever done with real music.

I think posting back to the list in public
should be fine. It's up to you whether to
read any results before you listen!

Any comments on headphones vs. speakers for
this? I don't have speakers aside from the
ones on my keyboard amp, which isn't exactly
sweet-sounding.

-Carl

No response on the temperament shootout? I'm surprised,
given all the discussion about historical temperaments
here in the last couple months.

Here, I've created a poll for it...

-Carl

At 01:38 PM 4/19/2003, you wrote:
>Yeehah! Welcome one and all to the first
>annual Temperament Shootout!
>
>It's five short pieces of music...
>
>() Bach's A-Maj Fugue from the WTC, Book II.
>() Byrd's "Wolsey's Wilde".
>() My own "Clavichordius Terrificus" (1997).
>() My "Guys Dressed Like Girls (1997).
>() Izaac's "Rain".
>
>...each in five different 12-tone tunings...
>
>() 12-tone Equal Temperament
>() Meantone
>() 12-tone Pythagorean Just Intonation
>() Werckmeister III
>() Young
>
>The Bach, Byrd, and Izaac midi files were
>obtained on the web. Credit to Yo Tomita and
>and John Sankey for the first two, resp. My
>stuff was entered into Encore with a mouse and
>exported to midi. All files were stripped of
>continuous controllers, re-channeled and
>re-timbred in Cakewalk Pro Audio for this demo.
>
>The pieces were retuned with Scala 2.05 with
>"example/midi/mts". The Scala files used are
>available in a zip archive at...
>
>http://lumma.org/shootout/scales.zip
>
>[Note that the 12-tET tunings were still sent
>through this process, to minimize weirdness.]
>
>The resulting midi files were rendered direct
>to 16-bit mono wave at 44.1KHz. with Audio
>Compositor 4.5. This is *fantastic* software,
>and the author, Scott Mitchell, should be
>praised for making it, and for fixing a bug so
>this project could be completed! AC is $40...
>
>http://home.att.net/~audiocompositor/
>
>The pieces are rendered with patches from the
>freely-available "Personal Copy v5r2" soundfont
>library, with the exception of the Izaac piece,
>which was done with the (also free) "NS Piano"
>from NaturalStudio.
>
>The wave files were normalized and adjusted
>for DC bias (an AC bug?) in Cool Edit Pro 1.2
>and then converted to 128Kbps CBR mp3 with
>LAME 3.93.
>
>The five pieces are available per-tuning in
>zip archives here...
>
>http://lumma.org/shootout/tuning1.zip
>http://lumma.org/shootout/tuning2.zip
>http://lumma.org/shootout/tuning3.zip
>http://lumma.org/shootout/tuning4.zip
>http://lumma.org/shootout/tuning5.zip
>
>[These files are 4.6MB each. You don't have
>to download them all to participate. My server
>should be fine, but please let me know if you
>get any errors.
>
>Your mission, should you choose to accept it,
>is to guess the tunings. All you golden-ear
>tuners out there give it a shot! This must be
>one of the most accurate recorded historical
>temperament comparos ever done with real music.
>
>I think posting back to the list in public
>should be fine. It's up to you whether to
>read any results before you listen!
>
>Any comments on headphones vs. speakers for
>this? I don't have speakers aside from the
>ones on my keyboard amp, which isn't exactly
>sweet-sounding.
>
>-Carl

Carl,

--- In tuning@yahoogroups.com, Carl Lumma <ekin@l...> wrote:
> No response on the temperament shootout?

This may be a somewhat oblique response, but its sincere: I'm really impressed by the amount of thought, effort, and careful documentation that went into this 'shootout'. I meant to say as much when I saw the first notice fly by, but it has been one of 'those' weeks.

You may notice I haven't even downloaded/listened to any yet, but I still think you need to be commended for the project. I hope to get to it at some point, but we should all note that I don't stand a ghost of a chance on id-ing the temps - I'm just going to enjoy listening to the differences for now!

Good on you,
Best,
Jon

--- In tuning@yahoogroups.com, "Jon Szanto" <JSZANTO@A...> wrote:
> Carl,
>
> --- In tuning@yahoogroups.com, Carl Lumma <ekin@l...> wrote:
> > No response on the temperament shootout?
>
> This may be a somewhat oblique response, but its sincere: I'm
really impressed by the amount of thought, effort, and careful
documentation that went into this 'shootout'. I meant to say as much
when I saw the first notice fly by, but it has been one of 'those'
weeks.
>
> You may notice I haven't even downloaded/listened to any yet, but I
still think you need to be commended for the project. I hope to get
to it at some point, but we should all note that I don't stand a
ghost of a chance on id-ing the temps - I'm just going to enjoy
listening to the differences for now!
>
> Good on you,
> Best,
> Jon

i, too, will have to wait a few days to get to this . . . can't be
playing music at work, you know (my cubicle is not an effective sound
shield . . .)

[Jon wrote...]
>You may notice I haven't even downloaded/listened to any yet, but I
>still think you need to be commended for the project. I hope to get
>to it at some point, but we should all note that I don't stand a ghost
>of a chance on id-ing the temps - I'm just going to enjoy listening
>to the differences for now!

Sure thing, hope you enjoy listening!

[Paul wrote...]
>i, too, will have to wait a few days to get to this . . . can't be
>playing music at work, you know (my cubicle is not an effective
>sound shield . . .)

Been there. We used to use headphones. Used to listen to shoutcast
while working. The shootout might require more attention... perhaps
you can put up some sort of phoney graph on your monitor. :)

Just kidding. Hope you find the right time to listen.

Gene, weren't you asking about samples of music in various historical
temperaments? Johnny? Ibo? I've tried my best to break the "it
doesn't work over the internet" barrier. If it can work in a
recording, it ought to work in this demo.

It took about an hour and a half to upload all five files over a
dialup connection (and not a very good one, at that).

-Carl

--- In tuning@yahoogroups.com, Carl Lumma <ekin@l...> wrote:

> No response on the temperament shootout? I'm surprised,
> given all the discussion about historical temperaments
> here in the last couple months.

How many of us did you expect to already know how to tell which was
which by ear? I'd need to listen to a lot of examples before even
attempting such a feat. That was one reason I complained when it
seemed to me I was listening to stuff that wasn't 12-et, but the notes
failed to indicate what tuning was actually used.

>> No response on the temperament shootout? I'm surprised,
>> given all the discussion about historical temperaments
>> here in the last couple months.
>
>How many of us did you expect to already know how to tell
>which was which by ear?

I'm trying to find out; I have no idea.

-Carl

>>How many of us did you expect to already know how to tell
>>which was which by ear?
>
>I'm trying to find out; I have no idea.

Well, some have made strong claims that they can already
do this. For those that need training, I've included the
scala files -- you can run midis through Scala and listen
till your heart's content.

The poll is set up so you can provide only as many answers
as you want. I expect everybody to get meantone. The
others, however...

-Carl

In a message dated 4/22/03 6:32:15 PM Eastern Daylight Time, ekin@lumma.org
writes:

> Gene, weren't you asking about samples of music in various historical
> temperaments? Johnny? Ibo? I've tried my best to break the "it
> doesn't work over the internet" barrier. If it can work in a
> recording, it ought to work in this demo.
>
>

Would you agree that it is important to name the key of the selection? If in
Werckmeister III it makes a whole lot of difference.

best, Johnny Reinhard

>Would you agree that it is important to name the key of the
>selection? If in Werckmeister III it makes a whole lot of
>difference.

Absolutely, an important omission on my part.

We'll have to ask Manuel about Scala's exact behavior absent
a keyboard map. I believe it takes the 1/1 from the .scl
file and tunes it to the "base frequency" specified in the
settings. In this case, that is C = 261.6Hz. The user can
then view the .scl files (in a text editor, say) to see the
tunings (all wolves occur between Eb and G#)...

http://lumma.org/shootout/scales.zip

All the music is played with this setup -- I don't change
the "key" of the tuning, as per the practice of your
ensemble(s). I also thought it would make things easier,
by increasing the chances of hearing the 'worse' keys in
the tunings that have them.

The only snag is that I'm holding C, not A, which may throw
off absolute-pitch cues (which can be important even to
musicians and listeners without 'perfect pitch'). However,
even with a fixed A, which pitch standard should I choose
for, say, Byrd and Bach? So I don't think this is too much
of a problem.

The music is in the following keys...

Bach - A Maj
Byrd - C Maj
Carl/Clavichordius - G Maj
Carl/GuysGirls - F Maj
Izaac - G Maj

-Carl

Carl,

> [These files are 4.6MB each. You don't have
> to download them all to participate.
> Your mission, should you choose to accept it,
> is to guess the tunings.

I guess I was not understanding correctly: I wanted to listen to the Bach and Byrd (the first 2 pieces you list), and downloaded the first two files. Alas, each zip file is the five pieces in 1 tuning.

If you don't have to download all 5 files, what kind of feedback did you expect to get? If I only downloaded the first file, should one be able to tell the tuning in use for those five examples without hearing any of the other tunings?

Well, now that I see what is going on, I'll probably download them tonight when other things are happening...

Cheers,
Jon

>> [These files are 4.6MB each. You don't have
>> to download them all to participate.
>> Your mission, should you choose to accept it,
>> is to guess the tunings.
>
>I guess I was not understanding correctly: I wanted to listen
>to the Bach and Byrd (the first 2 pieces you list), and downloaded
>the first two files.

Which is why the files are called "tuning1", "tuning2", etc.

>If I only downloaded the first file, should one be able to
>tell the tuning in use for those five examples without hearing
>any of the other tunings?

Quite possibly. The poll allows incomplete votes.

-Carl

> Which is why the files are called "tuning1", "tuning2", etc.

I know, you think I could have figured that out myself! :) Oh well, I have to download all 5 then, as I'm interested in comparing the same piece in the 5 temperaments.

> Quite possibly. The poll allows incomplete votes.

Well, I'll be *real* interested to see who can identify a temperament this way!

Cheers,
Jon

>Well, I'll be *real* interested to see who can identify a temperament
>this way!

I haven't spent much time listening to these, but it seems to me that
meantone has a characteristic sound, and I can usually spot the
harshness of Pythagorean major thirds.

Others claim that temperaments impart such a strong fingerprint as to
be unmistakable, without reference to any other tuning, to the point
where it is a gross artistic error to use the wrong one.

-Carl

Many thanks to Carl Lumma for the "Illummanating" Temperament
Shootout, which I'm *finally* getting to. Great fun.

Well, I wonder if I got them all wrong... I'm a bit behind on my
list reading, so this is entirely fresh here:

I would say that T1, T3 and T5 seem the most oriented to just for
me. T2 seems to beat all over the place, and I suspect Pythagorean,
and T4 sounds like our "beloved" 12-equal.

So this is what I get:

T1 = Werckmeister III

T2 = Pythagorean

T3 = Young

T4 = 12 Equal

T5 = Meantone

Now, please embarass me by telling me how many are "wrong..." :)

I enjoyed using the Byrd particularly for this. I found that some of
the pieces with faster tempos made the exercise rather
indistinguisable...

Thanks!

J. Pehrson

>So this is what I get:
>
>T1 = Werckmeister III
>T2 = Pythagorean
>T3 = Young
>T4 = 12 Equal
>T5 = Meantone
>
>Now, please embarass me by telling me how many are "wrong..." :)

I see someone else has voted on the poll page. Is that you?
Did you look at the poll page?

I'm not giving the answers away anymore than I already have until
Paul gets a chance to try it.

>I enjoyed using the Byrd particularly for this. I found that some of
>the pieces with faster tempos made the exercise rather
>indistinguisable...

That's what I found.

-Carl

--- In tuning@yahoogroups.com, "Jon Szanto" <JSZANTO@A...> wrote:

/tuning/topicId_43429.html#43490

> Carl,
>
> > [These files are 4.6MB each. You don't have
> > to download them all to participate.
> > Your mission, should you choose to accept it,
> > is to guess the tunings.
>
> I guess I was not understanding correctly: I wanted to listen to
the Bach and Byrd (the first 2 pieces you list), and downloaded the
first two files. Alas, each zip file is the five pieces in 1 tuning.
>
> If you don't have to download all 5 files, what kind of feedback
did you expect to get? If I only downloaded the first file, should
one be able to tell the tuning in use for those five examples without
hearing any of the other tunings?
>
> Well, now that I see what is going on, I'll probably download them
tonight when other things are happening...
>
> Cheers,
> Jon

***Well, for purposes of this exercise, it might have been better to
have had *one* piece in each .zip file, tuned the five different
ways, rather than each .zip a different tuning of the five pieces.
But too late now. :)

JP

--- In tuning@yahoogroups.com, Carl Lumma <ekin@l...> wrote:

/tuning/topicId_43429.html#43832

> >So this is what I get:
> >
> >T1 = Werckmeister III
> >T2 = Pythagorean
> >T3 = Young
> >T4 = 12 Equal
> >T5 = Meantone
> >
> >Now, please embarass me by telling me how many are "wrong..." :)
>
> I see someone else has voted on the poll page. Is that you?
> Did you look at the poll page?
>

***Yes, that was me. I've been very busy this last month or so, and
have been Tuning List "skimming." Finally catching up a bit...

JP