back to list

Lists of consonant chords?

🔗Jim Savage <waldpond@xxxxx.xxxx>

11/11/1999 4:52:21 PM

Thanks to all for the information so far. I understand how to do what I
started out on now, with the next step being finding the most (and then
progressively less) consonant "scales" for each of the different types of
chord. A few of the easiest examples:

MM7 1 5/4 3/2 15/8
mm7 1 6/5 3/2 9/5

In this light I understand John Link's use of (the dominant 7th of a major
scale as in his previous example)

Mm7 1 5/4 3/2 7/4

as the most consonant. I'm not saying this is the most desirable to use in
all dominant 7th situations, only that it seems to be the most consonant
chord taken in total isolation. Thus the desire for listing further, less
consonant (in some sort of rough, increasingly dissonant order) for each
chord type, such as Paul Ehrlich's use of 9/5 or 16/9 for the 7th in the
above Mm7.

So, is

dd7 1 6/5 7/5 5/3

a good one? Or is it significantly more consonant to have the smaller
intervals to the 7th more consonant and the interval from the root to the
7th less consonant? I don't want to get into arguments about which is "THE
most consonant" since, unless there's a precise (mathematical) definition of
consonance there's no real way to decide the issue. (Perhaps some weighted
measure of beats, harmonic entropy, timbral aspects, etc., with different
weights for different styles?) Rather, I'm interested in listing the main
possibilities so it's easy to try the different possibilities out in
different situations, with a rough ordering, or better yet a tag for each
possibility indicating the style or context it's most appropriate for.

I want to extend this procedure to the various 9, 11, 13, 6 and whatever
chords, until there is essentially a list of "scales" for each of the triad
types, each scale in the list being labelled as to their properties, use,
whatever.

I get the impression that John Link is doing something similar to this, no?
Are there lists like this available - at least so I can check mywork against
them to make sure I don't miss anything important, or is this an exercise
for the student? :)

Addendum re previous posts:

John and Paul: Thanks for all the examples, etc.

Margo: Thanks for the info on the major third wolf. I knew there were
various wolf intervals around, but had only specificaly read of a fifth
wolf. I found the actual uses of this major third wolf interesting, I'ld
gotten the impression before that people just stayed away from them.

Jim Savage

🔗Paul H. Erlich <PErlich@xxxxxxxxxxxxx.xxxx>

11/11/1999 9:43:03 PM

Jim Savage wrote,

>I understand how to do what I
>started out on now, with the next step being finding the most (and then
>progressively less) consonant "scales"

You mean tunings?

>for each of the different types of
>chord.

Remember that some chords may have their most consonant tuning outside of
JI.

>So, is

>dd7 1 6/5 7/5 5/3

>a good one? Or is it significantly more consonant to have the smaller
>intervals to the 7th more consonant and the interval from the root to the
>7th less consonant? I don't want to get into arguments about which is "THE
>most consonant" since, unless there's a precise (mathematical) definition
of
>consonance there's no real way to decide the issue. (Perhaps some weighted
>measure of beats, harmonic entropy, timbral aspects, etc., with different
>weights for different styles?) Rather, I'm interested in listing the main
>possibilities so it's easy to try the different possibilities out in
>different situations, with a rough ordering, or better yet a tag for each
>possibility indicating the style or context it's most appropriate for.

If you're considering the beating/roughness (Helmoltz/Plomp/Sethares)
components of consonance, it is most important to list all the intervals in
your chord. In this case, you have

6:5 7:5 5:3 7:6 25:18 25:21

The last two ratios are complex and are formed by the 5/3. So the 5/3 would
not be too stable and might want to move, but where? Say you move it to
(6/5)*(7/5)=42/25. The chord is now

dd7 1/1 6/5 7/5 42/25

Then the list of intervals becomes

6:5 7:5 42:25 7:6 7:5 6:5

This may be an improvement in some timbres, but in others the best way to
minimize roughness may be to tune the diminshed seventh chord to something
very close to 12-tET (or 4-tET). This is because in many timbres a
16-cent-out-of-tune 6:5 is still less rough than a pure 7:6 (remember that
the critical bandwidth kicks in at about 7:6 in most registers).

But if you're considering the combination tone/virtual pitch components of
consonance,

dd7 1/1 6/5 7/5 17/10

is by far the best, since it's just harmonics 10, 12, 14, and 17 over a
common fundamental (while the last two JI options come out as 15:18:21:25
and 25:30:35:42, respectively) and there seems to be a real drop in our
ability to understand high harmonics somewhere between 17 and 25.