This is an Opt In Archive . We would like to hear from you if you want your posts included. For the contact address see About this archive. All posts are copyright (c).

- Contents - Hide Contents - Home - Section 9

First Previous Next

8000 8050 8100 8150 8200 8250 8300 8350 8400 8450 8500 8550 8600 8650 8700 8750 8800 8850 8900 8950

8000 - 8025 -



top of page bottom of page up down


Message: 8000

Date: Thu, 06 Nov 2003 19:57:38

Subject: Re: Eponyms

From: Paul Erlich

i actually used log(p-q)/log(odd limit), which makes the rankings a 
little different for the big (in cents) intervals but does not affect 
the ranking of the small commas.

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "Gene Ward Smith" 
<gwsmith@s...> 
> wrote:
> > --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" 
> <gdsecor@y...> 
> > wrote:
> > 
> > > I don't have the resources to do that very readily.  Wouldn't 
> > putting 
> > > them in order of product complexity (discarding any with 
factors 
> > > above some particular prime limit) accomplish this? 
> > 
> > Not really; however we can simply take everything below a certain 
> > prime limit and below a limit for what I call "epipermicity",
> 
> i think you mean "epimericity"
> 
> > which 
> > is, for p/q>1 in reduced form, log(p-q)/log(q). It can be shown 
> this 
> > gives a finite list of commas if the epimermicity limit is less 
> than 
> > one.
> 
> i'll try 7-limit some other time, but since i still have my 5-limit 
> list in matlab's memory, here's the top rankings (for intervals < 
600 
> cents) by epimericity -- 1/1 shows up as best but actually its 
> epimericity is 0/0 so is undefined:
> 
>                      numerator                denominator
>                          1                         1
>                         16                        15
>                          6                         5
>                         81                        80
>                          4                         3
>                          9                         8
>                         10                         9
>                          5                         4
>                         25                        24
>                         27                        25
>                        128                       125
>                      32805                     32768
>                        250                       243
>                        135                       128
>                       2048                      2025
>                      15625                     15552
>                        256                       243
>                        648                       625
>                         32                        27
>                       3125                      3072
>                         75                        64
>                      78732                     78125
>                       6561                      6400
>                      20000                     19683
>                        125                       108
>                         27                        20
>                         32                        25
>                         25                        18
>                        625                       576
>                    1600000                   1594323
>                        144                       125
>                     393216                    390625
>                        256                       225
>                      16875                     16384
>                       2187                      2048
>                         81                        64
>                    2109375                   2097152
>                        800                       729
>                       6561                      6250
>                       1125                      1024
>                       3125                      2916
>                        100                        81
>                     531441                    524288
>                         45                        32
>                      20480                     19683
>                       2187                      2000
>                        729                       640
>                       2048                      1875
>                        243                       200
>                      16384                     15625
>                        125                        96
>                        729                       625
>                 1076168025                1073741824
>                       3456                      3125
>                 6115295232                6103515625
>                 1224440064                1220703125
>                    1594323                   1562500
>                    1990656                   1953125
>               274877906944              274658203125
>                     262144                    253125
>                        625                       512
>                10485760000               10460353203
>                       1215                      1024
>                      62500                     59049
>              7629394531250             7625597484987
>                      78125                     73728
>                     273375                    262144
>                       4096                      3645
>                   67108864                  66430125
>                  129140163                 128000000
>                       2500                      2187
>                        162                       125
>                    1638400                   1594323
>                     531441                    512000
>                    4194304                   4100625
>                      82944                     78125
>                      32768                     30375
>                  390625000                 387420489
>                  244140625                 241864704
>                     390625                    373248
>                    9765625                   9565938
>                    1953125                   1889568
>                 4294967296                4271484375
>                      18225                     16384
>                        625                       486
>                   34171875                  33554432
>                       2560                      2187
>                        768                       625
>                     140625                    131072
>                31381059609               31250000000
>                       9375                      8192
> etc.
> 
> considering that i had numerators and denominators well in excess 
of 
> 10^50 in the list, i'm inclined to believe gene that a given 
> epimericity cutoff will yield a finite list. and it's a good list 
> too -- i'm kind of pleased with this as a temperament ranking, with 
> meantone very near the top, augmented, schismic, pelogic, 
> diaschismic, blackwood, kleismic, and diminished forming a 
> consecutive block of interesting and eminently useful systems 
(given 
> their characteristic DE scales), while more unlikely choices for 
> human music making, like semisuper, parakleismic, and ennealimmal, 
as 
> well as many simpler systems with high error, fall further down -- 
> and of course monstrosities like atomic don't appear at all. i 
wonder 
> if even dave could stomach such a ranking -- the very simple 
> temperaments with high error are easy enough to mentally toss out 
for 
> the user seeking a certain goodness of approximation . . .


top of page bottom of page up down


Message: 8001

Date: Thu, 06 Nov 2003 20:05:43

Subject: Re: Eponyms

From: Paul Erlich

there were actually a log of ties in the rankings -- for example log
(1)=0 so all superparticulars got a zero score. here's a list again 
with the scores shown in the third column:

1	1	-Inf
16	15	0
6	5	0
81	80	0
4	3	0
9	8	0
10	9	0
5	4	0
25	24	0
27	25	0.210309918
128	125	0.227535398
32805	32768	0.3472592
250	243	0.35424875
135	128	0.396697481
2048	2025	0.41184295
15625	15552	0.444302056
256	243	0.466943504
648	625	0.487048023
32	27	0.488324507
3125	3072	0.493376213
75	64	0.555391285
78732	78125	0.568834688
6561	6400	0.578161697
20000	19683	0.582442033
125	108	0.586791476
27	20	0.590414583
32	25	0.604530978
25	18	0.604530978
625	576	0.604530978
1600000	1594323	0.605251545
144	125	0.6098276
393216	390625	0.610445976
256	225	0.634033151
16875	16384	0.636604282
2187	2048	0.641650248
81	64	0.644725481
2109375	2097152	0.646280585
800	729	0.646676406
6561	6250	0.653073083
1125	1024	0.65690632
3125	2916	0.663875781
100	81	0.670035965
531441	524288	0.673219541
45	32	0.673805299
20480	19683	0.675686222
2187	2000	0.680222895
729	640	0.680955483
2048	1875	0.683790172
243	200	0.684718377
16384	15625	0.686782398
125	96	0.697406178
729	625	0.704584463
1076168025	1073741824	0.706932191
3456	3125	0.721011768
6115295232	6103515625	0.72260722
1224440064	1220703125	0.723318814
1594323	1562500	0.725946912
1990656	1953125	0.727163637
2.74878E+11	2.74658E+11	0.729258578
262144	253125	0.731984665
625	512	0.7343228
10485760000	10460353203	0.739050418
1215	1024	0.739496502
62500	59049	0.741519042
7.62939E+12	7.6256E+12	0.743614519
78125	73728	0.744596936
273375	262144	0.745006093
4096	3645	0.745199871
67108864	66430125	0.745516574
129140163	128000000	0.746753932
2500	2187	0.747202793
162	125	0.747863148
1638400	1594323	0.748755323
531441	512000	0.749061613
4194304	4100625	0.75181536
82944	78125	0.752731448
32768	30375	0.753804891
390625000	387420489	0.757524855
244140625	241864704	0.757919646
390625	373248	0.758254069
9765625	9565938	0.75830862
1953125	1889568	0.763530363
4294967296	4271484375	0.765348818
18225	16384	0.766324469
625	486	0.76649026
34171875	33554432	0.768629073
2560	2187	0.770007566
768	625	0.770897186
140625	131072	0.773133533
31381059609	31250000000	0.773337708
9375	8192	0.773667414


--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
> i actually used log(p-q)/log(odd limit), which makes the rankings a 
> little different for the big (in cents) intervals but does not 
affect 
> the ranking of the small commas.
> 
> --- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
> wrote:
> > --- In tuning-math@xxxxxxxxxxx.xxxx "Gene Ward Smith" 
> <gwsmith@s...> 
> > wrote:
> > > --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" 
> > <gdsecor@y...> 
> > > wrote:
> > > 
> > > > I don't have the resources to do that very readily.  Wouldn't 
> > > putting 
> > > > them in order of product complexity (discarding any with 
> factors 
> > > > above some particular prime limit) accomplish this? 
> > > 
> > > Not really; however we can simply take everything below a 
certain 
> > > prime limit and below a limit for what I call "epipermicity",
> > 
> > i think you mean "epimericity"
> > 
> > > which 
> > > is, for p/q>1 in reduced form, log(p-q)/log(q). It can be shown 
> > this 
> > > gives a finite list of commas if the epimermicity limit is less 
> > than 
> > > one.
> > 
> > i'll try 7-limit some other time, but since i still have my 5-
limit 
> > list in matlab's memory, here's the top rankings (for intervals < 
> 600 
> > cents) by epimericity -- 1/1 shows up as best but actually its 
> > epimericity is 0/0 so is undefined:
> > 
> >                      numerator                denominator
> >                          1                         1
> >                         16                        15
> >                          6                         5
> >                         81                        80
> >                          4                         3
> >                          9                         8
> >                         10                         9
> >                          5                         4
> >                         25                        24
> >                         27                        25
> >                        128                       125
> >                      32805                     32768
> >                        250                       243
> >                        135                       128
> >                       2048                      2025
> >                      15625                     15552
> >                        256                       243
> >                        648                       625
> >                         32                        27
> >                       3125                      3072
> >                         75                        64
> >                      78732                     78125
> >                       6561                      6400
> >                      20000                     19683
> >                        125                       108
> >                         27                        20
> >                         32                        25
> >                         25                        18
> >                        625                       576
> >                    1600000                   1594323
> >                        144                       125
> >                     393216                    390625
> >                        256                       225
> >                      16875                     16384
> >                       2187                      2048
> >                         81                        64
> >                    2109375                   2097152
> >                        800                       729
> >                       6561                      6250
> >                       1125                      1024
> >                       3125                      2916
> >                        100                        81
> >                     531441                    524288
> >                         45                        32
> >                      20480                     19683
> >                       2187                      2000
> >                        729                       640
> >                       2048                      1875
> >                        243                       200
> >                      16384                     15625
> >                        125                        96
> >                        729                       625
> >                 1076168025                1073741824
> >                       3456                      3125
> >                 6115295232                6103515625
> >                 1224440064                1220703125
> >                    1594323                   1562500
> >                    1990656                   1953125
> >               274877906944              274658203125
> >                     262144                    253125
> >                        625                       512
> >                10485760000               10460353203
> >                       1215                      1024
> >                      62500                     59049
> >              7629394531250             7625597484987
> >                      78125                     73728
> >                     273375                    262144
> >                       4096                      3645
> >                   67108864                  66430125
> >                  129140163                 128000000
> >                       2500                      2187
> >                        162                       125
> >                    1638400                   1594323
> >                     531441                    512000
> >                    4194304                   4100625
> >                      82944                     78125
> >                      32768                     30375
> >                  390625000                 387420489
> >                  244140625                 241864704
> >                     390625                    373248
> >                    9765625                   9565938
> >                    1953125                   1889568
> >                 4294967296                4271484375
> >                      18225                     16384
> >                        625                       486
> >                   34171875                  33554432
> >                       2560                      2187
> >                        768                       625
> >                     140625                    131072
> >                31381059609               31250000000
> >                       9375                      8192
> > etc.
> > 
> > considering that i had numerators and denominators well in excess 
> of 
> > 10^50 in the list, i'm inclined to believe gene that a given 
> > epimericity cutoff will yield a finite list. and it's a good list 
> > too -- i'm kind of pleased with this as a temperament ranking, 
with 
> > meantone very near the top, augmented, schismic, pelogic, 
> > diaschismic, blackwood, kleismic, and diminished forming a 
> > consecutive block of interesting and eminently useful systems 
> (given 
> > their characteristic DE scales), while more unlikely choices for 
> > human music making, like semisuper, parakleismic, and 
ennealimmal, 
> as 
> > well as many simpler systems with high error, fall further down --
 
> > and of course monstrosities like atomic don't appear at all. i 
> wonder 
> > if even dave could stomach such a ranking -- the very simple 
> > temperaments with high error are easy enough to mentally toss out 
> for 
> > the user seeking a certain goodness of approximation . . .


top of page bottom of page up down


Message: 8002

Date: Thu, 06 Nov 2003 19:18:59

Subject: Re: Naming Commas

From: Carl Lumma

> relative to a chain of fifths.

Sacre bleu.

>> Will it?  I recently argued no; Paul seemed to argue yes.
>
>I expect some overlap certainly, but also large regions that are of
>interest to one and not the other. Anything with an absolute
>3-exponent greater than 12 (or at most 18) is not going to be of much
>interest for notational purposes. Are any of these of great interest
>as vanishing in a useful linear temperament?

What I need from you/Paul is a general principle of good chromatic
vectors that differs from the general principle of good commatic
vectors we already have (namely "epimericity", etc.).

-Carl


top of page bottom of page up down


Message: 8003

Date: Fri, 07 Nov 2003 22:38:56

Subject: Re: Eponyms

From: George D. Secor

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> 
> > i.e., 3 is obviously extremely important both historically
> > and theoretically, and thus deserves to be isolated by itself
> > (or grouped with 2, if 2 is included).

All of the new symbols in a JI heptatonic notation will be modifying 
nominals in a pythagorean sequence (if we include sharped/flatted 
tones under a broader concept of the term nominals),  so this is a 
natural place to make a separation.

> > the next comma appears after 11, and earlier in this thread
> > i discussed the idea that 11-limit can be a kind of boundary.
> > Partch thought so too.  (but please, don't anyone make too
> > much of this comment.)

Partch thought that 7 was implied in a 12-tone octave in that 5:7 or 
7:10 suggest consonant tritones, so 11 is the first prime that is 
completely foreign to conventional harmony.  Since ratios of 13 are 
similar in effect to ratios of 11 (and since 11 harmonically bridges 
to 13 rather easily), this makes a good case for stopping with 11.  
Also, those who advocate achieving new harmonic resources by 
extending the meantone temperament until it forms a closed system of 
31 tones find that they have ended up with an 11-limit (tempered) 
system.

So I think that there are many that would agree with 11 as a boundary.

> > the next comma appears after 19, which i myself used as
> > a limit from about 1988-98.

Mapping a harmonic series consistently into a 12-tone octave (without 
skipping over any primes to reach other primes, and without skipping 
over any odd harmonics to reach other odd harmonics) yields a 19-
limit set that's a favorite of mine:

16:17:18:19:20:21:22:24:25:26:28:30:32

So I also like 19 as a boundary.

> > the next comma appears after 31, which is the highest limit
> > Ben Johnston has used in his music.
> > 
> > interesting.

This complete 5 octaves of harmonics, and there's a big gap between 
31 and 37, the next prime.

> so the primes are arranged as
> 2 3 , 5 7 11 , 13 17 19 , 23 29 31 , 37 41 43 , 47 53 59 , 61 67 71
> 
> looks like the next comma after 31 makes sense too -- isn't 43 the 
> highest limit used by george secor at least in some context?

If you're referring to something that I posted some time ago, I think 
that was 41.  (But I did happen to come across a use for 43 when I 
was subsequently rummaging through some of my old papers.)  Trouble 
with 43 is that 32:43 readily invites confusion with 3:4.

--George


top of page bottom of page up down


Message: 8004

Date: Fri, 07 Nov 2003 22:46:41

Subject: Re: Eponyms

From: George D. Secor

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
> no reply, george?
> 
> --- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
> > --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" 
gdsecor@y...> 
> > wrote:
> > 
> > > BTW, have you ever tried collapsing an 11-limit lattice into 2 
> > > dimensions by mapping 11/8 to <10, 5>?
> > 
> > you're probably referring to the 3-5-11 lattice?
> > 
> > the full 11-limit lattice is at least 3 dimensional if you use 
> > one "xenharmonic bridge" as above. for the 3-5-11 case, this 
choice 
> > (184528125:184549376) is probably a very good one. for the full 
11-
> > limit case, 9800:9801 is probably better for most purposes, since 
> > it's both a little smaller (in cents) and much simpler (i.e., 
shorter 
> > in the lattice).

Sorry, I started something that sudden time constraints didn't permit 
me to finish.  Also, in my haste I also skipped over 7 -- 7/4 would 
also be mapped to the [7, -5] position.

--George


top of page bottom of page up down


Message: 8005

Date: Fri, 07 Nov 2003 22:47:52

Subject: Re: Eponyms

From: Gene Ward Smith

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
>> 
> considering that i had numerators and denominators well in excess 
of 
> 10^50 in the list, i'm inclined to believe gene that a given 
> epimericity cutoff will yield a finite list. 

I've shown that it does, for cutoffs less than 1, using Baker's 
theorem.

i wonder 
> if even dave could stomach such a ranking -- the very simple 
> temperaments with high error are easy enough to mentally toss out 
for 
> the user seeking a certain goodness of approximation . . .

Tossing out powers such as 6561/6400 is what I'd recommend also, 
though Dave might not go for it.


top of page bottom of page up down


Message: 8006

Date: Fri, 07 Nov 2003 22:57:17

Subject: Re: Linear Temperaments

From: Gene Ward Smith

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "Paul G Hjelmstad" 
> <paul.hjelmstad@u...> wrote:
> > A few questions -
> > 
> > I know how to derive generators-to-primes using commas in 
matrices. 
> > How is it done using values? 
> > 
> > Second question - how do you go the other way? That is, derive 
> commas 
> > from generators.
> 
> you need the mapping; then it's straightforward.

I derive commas from the wedgie in my software.


top of page bottom of page up down


Message: 8007

Date: Fri, 07 Nov 2003 23:07:51

Subject: Re: Eponyms

From: Gene Ward Smith

--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
wrote:
> So I think that there are many that would agree with 11 as a 
boundary.

I've recently done some pieces with full 13-limit chords as harmony; 
the results may convince people that stopping earlier is a good plan. 
So far as complete otonal and utonal chords go, however, 5-limit has 
a natural 3-et triadic nature, 7-limit 4-et tetradic, 9-limit 5-et 
quintadic, and 13-limit 7-et septadic. For 11-limit complete harmony, 
we are stuck with a 6 val, which is a little ungainly.


top of page bottom of page up down


Message: 8008

Date: Fri, 07 Nov 2003 23:10:38

Subject: Re: Eponyms

From: George D. Secor

--- In tuning-math@xxxxxxxxxxx.xxxx "Gene Ward Smith" <gwsmith@s...> 
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" 
<gdsecor@y...> wrote:
> 
> > I don't have the resources to do that very readily.  Wouldn't 
putting 
> > them in order of product complexity (discarding any with factors 
> > above some particular prime limit) accomplish this? 
> 
> Not really; however we can simply take everything below a certain 
> prime limit and below a limit for what I call "epipermicity", which 
> is, for p/q>1 in reduced form, log(p-q)/log(q). It can be shown 
this 
> gives a finite list of commas if the epimermicity limit is less 
than 
> one.
> 
> >  One would only 
> > need to determine at what point two commas were forced to share 
the 
> > same name and then decide if they were both "important".
> 
> NEVER! Two commas with the same name makes no sense at all.

I wasn't suggesting giving them the same name, as I said in msg. 
#7420:

<<  As for the answer to your second question: I don't know if there 
will be unique names for *all* of them (probably not), but I believe 
that there are enough names to cover all of those that would be of 
use to most theorists.  Those with very specialized purposes (such as 
yourself) could devise a modification to Dave's naming system (such 
as appending letters a, b, c, etc. in order of ratio complexity) to 
distinguish equivocal names -- I think you're creative enough to come 
up with something that would be meaningful (or perhaps Dave might 
have some ideas).  >>

I really haven't had very much time lately to participate in this 
sort of discussion, but when it seemed that Dave had dropped out I 
saw that there were things that hadn't been resolved, I felt that I 
had to jump in and say something.  If you want a couple of specific 
examples to pursue this further, I can mention a couple of instances 
from which we were trying to set the kleisma-comma boundary:

152:153 (~11.4c) is definitely a 17:19-kleisma, but 1114112:1121931 
(~12.1c) will either be either a (subordinate) 17:19-kleisma or the 
17:19-comma, depending on where the boundary is set.

135:136 (~12.8c) will either be the 5:17-kleisma or the 5:17-comma, 
but 327680:334611 (~36.2c) also claims the name 5:17-comma with our 
present comma/S-diesis boundary.

These are issues that we still need to work out.

--George


top of page bottom of page up down


Message: 8009

Date: Fri, 07 Nov 2003 00:30:14

Subject: Re: Eponyms

From: Dave Keenan

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> wrote:
...
> 
> considering that i had numerators and denominators well in excess of 
> 10^50 in the list, i'm inclined to believe gene that a given 
> epimericity cutoff will yield a finite list. and it's a good list 
> too -- i'm kind of pleased with this as a temperament ranking, with 
> meantone very near the top, augmented, schismic, pelogic, 
> diaschismic, blackwood, kleismic, and diminished forming a 
> consecutive block of interesting and eminently useful systems (given 
> their characteristic DE scales), while more unlikely choices for 
> human music making, like semisuper, parakleismic, and ennealimmal, as 
> well as many simpler systems with high error, fall further down -- 
> and of course monstrosities like atomic don't appear at all. i wonder 
> if even dave could stomach such a ranking -- the very simple 
> temperaments with high error are easy enough to mentally toss out for 
> the user seeking a certain goodness of approximation . . .

It's not bad. It seems to sufficiently penalise excessive complexity,
but as you observe, it does not sufficiently penalise excessive error.


top of page bottom of page up down


Message: 8010

Date: Fri, 07 Nov 2003 01:25:03

Subject: Naming Commas

From: Dave Keenan

--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...>
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> > hi George,
> > 
> > --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" 
> <gdsecor@y...> wrote:
> > 
> > > Loocs like you kaught something from Dave Ceenan.  ;-)
> > 
> > i usually try to follow suggestions for standardization
> > ... unless i very strongly disagree, as i did with Sims
> > 72edo notation.

Thanks Monz, but I actually never proposed it as any kind of standard.
I first just did it as an expedient in one particular post, where I
thought there was potential for confusion, and I explained the usage
at the start of that post.
Yahoo groups: /tuning-math/message/6875 * [with cont.] 
But apparently a few people failed to interpret this sentence
correctly, and instead assumed I was inventing a new term, and they
didn't like it so they didn't read any further.

But I didn't learn this until much later, so I carried on using it. It
seemed to me to be a convenient way to keep the two meanings distinct,
at least in writing. Had I known that this was actually a _barrier_ to
understanding the proposed naming system, I would have stopped using
it sooner.

> It's very nice that you brought this up, because that's the next 
> thing I was going to suggest.  Why not modify my suggestion above by 
> dropping the commas entirely, then changing the semicolons that 
> remain back to commas, so that the above example (133:72) would be 
> done this way:
> [-3 -2, 0 1 0, 0 0 1] or 
> [-2, 0 1 0, 0 0 1].
> This makes the grouping by threes more obvious (and the higher primes 
> much easier to locate), and angle brackets would no longer be 
> necessary.

I like this very much. It is also MATLAB/Octave compatible. Since
commas are optional, but semicolons indicate the end of a row (and so
would make a matrix, not a vector). Isn't that right Paul?

The angle-bracket thing was OK too. But I was going to suggest that
the octave-specific vectors should use the angle-brackets, since I
think octave-equivalent ones using square brackets have been around a
lot longer. But with this scheme they can all use square brackets.

And as pointed out by Monz, it will be easy to remember that the
commas come after the exponents of 3, 11, 19 and 31 since these have
indeed seemed to be natural stopping (resting?) places.

So a Pythagorean comma(generic sense) could be given in
octave-specific form as [x y] or [x y,] without ambiguity, but in
octave equivalent form it would have to be [y,] although I don't
recall ever having seen a vextor with only octaves in it, so if you
saw [y] you'd be pretty sure it was meant to be [y,].

Then there's the possibility of 2-and-3-free monzos being used to name
very complex commas in George's and my system. These would have to
_start_ with a comma. For example, the atom of Kirnberger is also the
[,12]-schismina. Although I don't know how you pronounce that so it's
clear you're giving a prime exponent, and not a factor.
"five-to-the-twelve-schismina" works just as well for me in this case.

And indeed, how should we pronounce the commas(punctuation sense) so
they don't get confused with commas(generic sense) or commas(specific
size range sense)?

I'm guessing these won't be very comma-n in spoken conversation
between musicians, so we can ignore this problem. But I think we've
comma long way.


top of page bottom of page up down


Message: 8011

Date: Fri, 07 Nov 2003 01:31:25

Subject: Naming Commas

From: Dave Keenan

George made a good point about who "the rest of us" might be.

I then realised that mathematical types searching for temperaments
want names for the commas(generic sense) that vanish, while musicians
using those temperaments will need names for the commas(generic sense)
that _don't_ vanish. Why would they need a name for something that
isn't there?

This dichotomy will also lead to quite different rankings of "the most
important commas(generic sense)".


top of page bottom of page up down


Message: 8013

Date: Fri, 07 Nov 2003 02:14:16

Subject: Re: Naming Commas

From: Carl Lumma

>I then realised that mathematical types searching for temperaments
>want names for the commas(generic sense) that vanish, while musicians
>using those temperaments will need names for the commas(generic 
sense)
>that _don't_ vanish. Why would they need a name for something that
>isn't there?

The names can be the same.

>This dichotomy will also lead to quite different rankings of "the 
most
>important commas(generic sense)".

Will it?  I recently argued no; Paul seemed to argue yes.

-Carl


top of page bottom of page up down


Message: 8014

Date: Fri, 07 Nov 2003 02:38:08

Subject: Re: Naming Commas

From: Dave Keenan

--- In tuning-math@xxxxxxxxxxx.xxxx "Carl Lumma" <ekin@l...> wrote:
> >I then realised that mathematical types searching for temperaments
> >want names for the commas(generic sense) that vanish, while musicians
> >using those temperaments will need names for the commas(generic 
> sense)
> >that _don't_ vanish. Why would they need a name for something that
> >isn't there?
> 
> The names can be the same.

Certainly. But these two different purposes may lead to diffferent
ideas about what would constitute a good name. George and I think that
a good name will indicate the simplest ratios that can be notated with
that comma, relative to a chain of fifths.

> 
> >This dichotomy will also lead to quite different rankings of "the 
> most
> >important commas(generic sense)".
> 
> Will it?  I recently argued no; Paul seemed to argue yes.

I expect some overlap certainly, but also large regions that are of
interest to one and not the other. Anything with an absolute
3-exponent greater than 12 (or at most 18) is not going to be of much
interest for notational purposes. Are any of these of great interest
as vanishing in a useful linear temperament?


top of page bottom of page up down


Message: 8015

Date: Fri, 07 Nov 2003 03:00:29

Subject: Re: Eponyms

From: Dave Keenan

--- In tuning-math@xxxxxxxxxxx.xxxx "Gene Ward Smith" <gwsmith@s...>
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
> wrote: 
> > One would only 
> > need to determine at what point two commas were forced to share the 
> > same name and then decide if they were both "important".
> 
> NEVER! Two commas with the same name makes no sense at all.

Gene, perhaps before you assume you are the last remaining defender of
good sense, and leap, with battle-cry, down the throat of the imagined
enemy, you might check to see if they really meant what you assumed
they meant.

In the very next sentence George wrote:

> > Dave already discussed this:
> > 
> > Yahoo groups: /tuning-math/messages/7320 * [with cont.] 

In that, you will find:
"To be certain that your comma actually deserves the
name, you have to run the process in reverse (as I've described
already) trying 3-exponents in the series 0, 1, -1, 2, -2, 3, -3, ...
and octave reducing, until you get a hit on the correct size-category.
Then see if you've got your original comma ratio back again."

So only the comma with the lowest absolute 3-exponent gets the simple
systematic name.

I think what George intended was, if only one is "important" then it
gets the name and the other little piggy has none. But if both are
important we have to figure a way to give them different names.

I suggest simply adding the adjective "complex" to the start of the
one with the second lowest 3-exponent, then if you need to go beyond
that, which seems very unliklely, then "hypercomplex" or some such.

For example, we have 
[12 19] as the Pythagorean-comma 23.5 c
and so
[41 65] might be called the complex-Pythagorean-comma 19.8 c.


top of page bottom of page up down


Message: 8016

Date: Fri, 07 Nov 2003 03:51:16

Subject: Re: Naming Commas

From: Dave Keenan

Oops!
I serously screwed up the monzos for those commas(specific size-range
sense). I should have written:

For example, we have 
[-19 12] as the Pythagorean-comma 23.5 c
and so
[65 -41] might be called the complex-Pythagorean-comma 19.8 c.


top of page bottom of page up down


Message: 8017

Date: Fri, 07 Nov 2003 03:58:08

Subject: Re: Naming Commas

From: Dave Keenan

--- In tuning-math@xxxxxxxxxxx.xxxx Carl Lumma <ekin@l...> wrote:
> What I need from you/Paul is a general principle of good chromatic
> vectors that differs from the general principle of good commatic
> vectors we already have (namely "epimericity", etc.).

I'll leave that to Paul, since that is slightly different again from
good commas for a general-purpose notation system.


top of page bottom of page up down


Message: 8018

Date: Fri, 07 Nov 2003 05:23:51

Subject: Re: Naming Commas

From: Paul Erlich

--- In tuning-math@xxxxxxxxxxx.xxxx "Dave Keenan" <d.keenan@b...> 
wrote:

> > [-3 -2, 0 1 0, 0 0 1] or 
> > [-2, 0 1 0, 0 0 1].
> > This makes the grouping by threes more obvious (and the higher 
primes 
> > much easier to locate), and angle brackets would no longer be 
> > necessary.
> 
> I like this very much. It is also MATLAB/Octave compatible. Since
> commas are optional, but semicolons indicate the end of a row (and 
so
> would make a matrix, not a vector). Isn't that right Paul?

i checked and you're right!


top of page bottom of page up down


Message: 8019

Date: Fri, 07 Nov 2003 05:25:09

Subject: Re: Naming Commas

From: Paul Erlich

--- In tuning-math@xxxxxxxxxxx.xxxx "Dave Keenan" <d.keenan@b...> 
wrote:
> George made a good point about who "the rest of us" might be.
> 
> I then realised that mathematical types searching for temperaments
> want names for the commas(generic sense) that vanish, while 
musicians
> using those temperaments will need names for the commas(generic 
sense)
> that _don't_ vanish. Why would they need a name for something that
> isn't there?

because it rules harmonic scale construction.


top of page bottom of page up down


Message: 8020

Date: Fri, 07 Nov 2003 05:26:48

Subject: Re: Linear Temperaments

From: Paul Erlich

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul G Hjelmstad" 
<paul.hjelmstad@u...> wrote:
> A few questions -
> 
> I know how to derive generators-to-primes using commas in matrices. 
> How is it done using values? 
> 
> Second question - how do you go the other way? That is, derive 
commas 
> from generators.

you need the mapping; then it's straightforward.

> Are Linear Temperaments always based on commas?

sure, it always can be, as long as it has a mapping.

> Any information, even partial is appreciated.

the mapping represents the primes in terms of the generators.


top of page bottom of page up down


Message: 8021

Date: Fri, 07 Nov 2003 05:27:58

Subject: Re: Naming Commas

From: Paul Erlich

--- In tuning-math@xxxxxxxxxxx.xxxx "Dave Keenan" <d.keenan@b...> 
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "Carl Lumma" <ekin@l...> wrote:
> > >I then realised that mathematical types searching for 
temperaments
> > >want names for the commas(generic sense) that vanish, while 
musicians
> > >using those temperaments will need names for the commas(generic 
> > sense)
> > >that _don't_ vanish. Why would they need a name for something 
that
> > >isn't there?
> > 
> > The names can be the same.
> 
> Certainly. But these two different purposes may lead to diffferent
> ideas about what would constitute a good name. George and I think 
that
> a good name will indicate the simplest ratios that can be notated 
with
> that comma, relative to a chain of fifths.
> 
> > 
> > >This dichotomy will also lead to quite different rankings 
of "the 
> > most
> > >important commas(generic sense)".
> > 
> > Will it?  I recently argued no; Paul seemed to argue yes.
> 
> I expect some overlap certainly, but also large regions that are of
> interest to one and not the other. Anything with an absolute
> 3-exponent greater than 12 (or at most 18) is not going to be of 
much
> interest for notational purposes. Are any of these of great interest
> as vanishing in a useful linear temperament?

depends what you mean by useful, but i'd say no.


top of page bottom of page up down


Message: 8022

Date: Fri, 07 Nov 2003 00:47:24

Subject: Re: Naming Commas

From: Carl Lumma

>What I need from you/Paul is a general principle of good chromatic
>vectors that differs from the general principle of good commatic
>vectors we already have (namely "epimericity", etc.).

I could see just 1/complexity, or somehow weakening the size term,
as size doesn't seem important beyond preserving propriety.

-Carl


top of page bottom of page up down


Message: 8023

Date: Fri, 07 Nov 2003 10:15:07

Subject: Comma size categories extended

From: Dave Keenan

I'm sorry, but I can't stop playing with the idea of systematic comma
names. But please note that these are intended to be _in_addition_to_
(not replacing) common or historical names in the case of well known
commas - just like in the naming of chemical compounds or biological
organisms.

Here's a proposal to extend the systematic comma size categories out
to 115 cents. And again I would like to acknowledge Joe Monzo's
pioneering work in this area.

The following tables will be much easier to read in email. If you're
reading on Yahoo's web interface, you can "Forward" it to yourself.

Size category name     Boundary        Alternative name
----------------------------------------------------------
                               0    c
schismina
             [-84 53]/2   ~=   1.81 c
schisma
             [317 -200]/2 ~=   4.50 c
kleisma
             [-19 12]/2   ~=  11.73 c
comma
             [-57 36]/2   ~=  35.19 c
minor-diesis                           or small-diesis
             [8 -5]/2     ~=  45.11 c
diesis                                 or medium-diesis
             [-11 7]/2    ~=  56.84 c
major-diesis                           or large-diesis
             [-30 19]/2   ~=  68.57 c
chromatic-semitone                     or small-semitone
             [35 -22]/2   ~=  78.49 c
limma                                  or medium-semitone
             [-3 2]/2     ~= 101.96 c
diatonic-semitone                      or large-semitone
             [62 -39]/2   ~= 111.88 c
apotome
             [-106 67]/2  ~= 115.49 c

These will give many systematic names that are the same as the
historical or common names used in Scala (with the substitution of
"Pythagorean" for 3, "classic" for 5, "septimal" for 7, etc.)

They also have the useful property that each category has an exact
apotome-complement category, except for schisma and kleisma which must
be combined to give a complement to the diatonic-semitone category.
But that's all right because the distinction between schisma and
kleisma isn't necessary for making names unique, but only for matching
historical usage.

Category         Complementary category
---------------------------------------
schismina        apotome
schisma/kleisma  diatonic-semitone
comma            limma
minor-diesis     chromatic-semitone
diesis           major-diesis

In case anyone has just joined the discussion: The purpose of these
precise boundaries is to make it possible to uniquely and
unambiguously name many commas without having to refer to the power of
3 (or 2) contained in the comma, since this is effectively encoded in
the size category.

George,
Please remind me why we didn't use the terms minor-diesis, diesis,
major-diesis?


top of page bottom of page up down


Message: 8024

Date: Sat, 08 Nov 2003 10:50:05

Subject: Re: Eponyms

From: Carl Lumma

>To give a simple example, in the 5-limit, (5+1)/2 = 3, and we may 
>start from the 3-chord [5/4, 3/2, 2]. If we solve for a val [a, b, c]
>such that 5/4, or [-2, 0, 1] is mapped to 1, 3/2 is mapped to 2, and 
>2 is mapped to 3 we get the equations a5 - 2 a2 = 1, a3 - a2 = 2, and
>a2 = 3, the solution of which is a2 = 3, a3 = 5, and a5 = 7, so the 
>val in question is uniquely determined to be [3, 5, 7], the standard 
>3-val for the 5-limit.

What's the definition of "standard val"?

-Carl


top of page bottom of page up

First Previous Next

8000 8050 8100 8150 8200 8250 8300 8350 8400 8450 8500 8550 8600 8650 8700 8750 8800 8850 8900 8950

8000 - 8025 -

top of page