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 8

Previous Next Last

7000 7050 7100 7150 7200 7250 7300 7350 7400 7450 7500 7550 7600 7650 7700 7750 7800 7850 7900 7950

7950 - 7975 -



top of page bottom of page up down


Message: 7975

Date: Tue, 04 Nov 2003 22:52:05

Subject: Re: hey Paul

From: Gene Ward Smith

--- In tuning-math@xxxxxxxxxxx.xxxx Carl Lumma <ekin@l...> wrote:
> >> Is there a way for people to look them up?
> >
> >Should I put up a web page? Dave, do you have an objection?
> 
> Graham's catalog is neither complete or up to date, last I
> checked.
> 
> The existence of a single resource is a lot to ask, I know...

I'm perfectly willing to put up web pages of named temperaments.


top of page bottom of page up down


Message: 7976

Date: Tue, 4 Nov 2003 13:36:51

Subject: Re: ennealimmal

From: Manuel Op de Coul

Carl wrote:
>Manuel, is there a convenient way to get MOS-like scales with
>non-octave periods in Scala?

Sure, create them with PYTHAGOREAN specifying the period as
formal octave, and then use EXTEND to change the number of tones.
The SHOW DATA command now also shows whether repeating blocks have
Myhill's property.

>Am I correct that the first ennealimmal scale with an octave
>is simply 9-equal, and the next is this 17-tone one...

You probably mean 18-tone. The generator doesn't need to be
exactly 50 cents, but if I understand your question correctly, yes.

Manuel


top of page bottom of page up down


Message: 7977

Date: Tue, 04 Nov 2003 15:24:14

Subject: Re: hey Paul

From: Carl Lumma

>> >Should I put up a web page? Dave, do you have an objection?
>> 
>> Graham's catalog is neither complete or up to date, last I
>> checked.
>> 
>> The existence of a single resource is a lot to ask, I know...
>
>I'm perfectly willing to put up web pages of named temperaments.

Then don't stop 'til the break of dawn!

-Carl


top of page bottom of page up down


Message: 7978

Date: Wed, 05 Nov 2003 15:44:25

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:
> 
> > Now if Dave says, "Should I use a 23-comma or 23-small-diesis 
(23S) 
> > symbol?", I immediately know what he's talking about: these are 
> > commas that alter tones in a 1/1 pythagorean chain to arrive at 
tones 
> > in a 23/16 and/or a 32/23 pythagorean chain.
> 
> This is fine for your very specialized purposes, but what about the 
> rest of us? Do you have a name for every superparticular ratio up 
to 
> the 23-limit?

I find it very amusing that you would place me, the 
*composer/theorist*, as the one with "very specialized purposes" 
while you, the *mathematician*, would group yourself with the "rest 
of us".  It would seem to me that only those with very specialized 
purposes would actually *need* to have a name for *every* 
superparticular ratio up to the 23-limit.

A systematic naming system should not be something that would make 
things more complicated for the rest of us.  I could hardly imagine a 
professor in a microtonal music course using "minus three, zero, 
zero, zero, one" as a name for 26:27 when "13L-diesis" (which can 
even be shortened to "13L") is so much simpler and clearer.  Monzos 
have their place as a specialized *notation* (which would also be of 
benefit in explaining the names), but not as *names* themselves.

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).

In summary, let's keep the simpler things simple.

--George


top of page bottom of page up down


Message: 7979

Date: Wed, 05 Nov 2003 18:08:51

Subject: Re: Eponyms

From: monz

hi George (and Gene)


--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
wrote:

> A systematic naming system should not be something that
> would make things more complicated for the rest of us.
> I could hardly imagine a professor in a microtonal music
> course using "minus three, zero, zero, zero, one" as a name
> for 26:27 when "13L-diesis" (which can even be shortened
> to "13L") is so much simpler and clearer.  Monzos have
> their place as a specialized *notation* (which would also
> be of benefit in explaining the names), but not as *names*
> themselves.



i see your point, and can agree with that.

... even tho *i* will always think of any rational
interval as its monzo.  

(i guess that's self-evident, given the name of the term?)

;-)



-monz


top of page bottom of page up down


Message: 7980

Date: Wed, 05 Nov 2003 18:32:59

Subject: Re: Eponyms

From: monz

hi George,


--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
wrote:

> The monzo comma-naming system is so cumbersome 
> (i.e., unfriendly), that I can't imagine how anyone
> could follow a name above the 11-limit unless it's
> written down.  Imagine trying to mention to someone 
> in spoken conversation that you're trying to decide
> whether to use the symbol for "<-6, 0, 0, 0, 0, 0, 0, 1>
> or <6, 0, 0, 0, 0, 0, 0, 1>" in a composition -- are
> you expecting me to be mentally prepared to count all
> those zeros so I know what prime number you mean when 
> you finally get to the "1" that matters?  



this is an interesting and good point.

in fact, over the years of working in extended-JI,
i've pretty much come to the conclusion that there are
so many notes available in even a rather compact 11-limit
euler-genus, that via xenharmonic bridging, that system
can imply many higher-prime ratios.

in particular, i've noticed that lots of 11-limit ratios
sound very similar to nearby 13-limit ratios.

if one accepts this aspect of my theory, then monzos
can easily be used to name all the relevant 11-limit
kommas.



on the other hand, the main reason i came up with the
idea of using monzos to represent prime-factored ratios
was that i wanted to avoid both having to always specify
the primes, and also to avoid superscripts. 

at the time i originally thought of the monzo idea,
i was working in 19-limit, and it seemed a easier to me 
to specify, to pick a totally random example, 133:72 as 
[-3, -2, 0, 1, 0, 0, 0, 1]  (with the prime-factors 
2, 3, 5, 7, 11, 13, 17, 19 understood) than to write
is out as 2^-3 * 3^-2 * 7^1 * 19^1.


> The problem is, the "name" (if you can call it that)
> emphasizes *powers* rather than *primes*, so it tends
> to get rather cryptic.


when one works with the same set of prime-factors
over and over again, one very easily gets used to
remembering the primes which underly the monzo.

and as i've pointed out before, the monzo allows direct
visualization of the lattice, which in turn helps in
comprehension of the structure of the entire tuning system.



i came up with the monzo idea based on analogy with
our regular numbering system.

it doesn't take too long for one to learn, whether in school 
or in everyday life, that, for example, the number 133 is a 
monzo-like representation of (10^2)*1 + (10^1)*3 + (10^0)*3.

the regular arabic numeral is a nice compact way of expressing
what would look like a rather complicated mathematical expression
if it were written out in full.  but once one learns how
it works, the long version is never need anymore.



anyway, i'll get off my soapbox now.  i've already agreed
that for purposes of naming beyond 11-limit, the prime system
is better than the monzo system.

but i will maintain that for 3-, 5-, 7-, and 11-limit,
the monzo system works just fine.




-monz


top of page bottom of page up down


Message: 7981

Date: Wed, 05 Nov 2003 18:34:19

Subject: Re: Eponyms

From: George D. Secor

--- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> 
> ... even tho *i* will always think of any rational
> interval as its monzo.  
> 
> (i guess that's self-evident, given the name of the term?)
> 
> ;-)

Oh, dear!  Now I fear that if someone brings up the word "secor" in a 
tuning context, they'll too readily associate that with the 
term "irrational".  ;-)

--George


top of page bottom of page up down


Message: 7982

Date: Wed, 05 Nov 2003 20:18:15

Subject: Re: Eponyms

From: monz

hi George,

--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
wrote:

> --- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> > 
> > ... even tho *i* will always think of any rational
> > interval as its monzo.  
> > 
> > (i guess that's self-evident, given the name of the term?)
> > 
> > ;-)
> 
> Oh, dear!  Now I fear that if someone brings up the
> word "secor" in a tuning context, they'll too readily
> associate that with the term "irrational".  ;-)
> 
> --George



oh, not necessarily!


in a series of old posts, most informatively this one:

Yahoo groups: /tuning/message/23195 * [with cont.] 

i presented a "rational canasta" tuning.  the express
purpose of this was to be able to map the canasta scale
to the computer keyboard in the old (JustMusic) version
of my software, which was not able to accept irrational
pitches.


in my "rational canasta" tuning, a secor is:

<3,5,7,11,13>-monzo = <-4, -1, 0, 1, -1>
ratio = 5632:5265 
= ~116.657 cents

this is only ~0.0101 cent less than 2^(7/72), the
"standard" secor.

and of course many other rational secors could be found.


i mentioned in one of those old posts the irony of
having to find a rational tuning which approximated
the subset of the irrational 72edo MIRACLE, which in
turn provides manifold approximations of rational 
JI intervals ... and even mentioned how it conjured
up Escher images in my mind.



and in fact, with regard to my original comment you
quoted, i also think of many irrational intervals
in terms of their monzos ... even tho many folks here
find that to be pointless since irrational intervals
can be prime-factored in an infinite number of ways.

but i find it useful, for example, to see the generator
"5th" of 1/4-comma meantone as the [2,3,5]-monzo [0, 0, 1/4],
or that of 2/7-comma meantone as [1/7, -1/7, 2/7].

i just wrote "rational" to avoid having to go into
details like this ... but now you've gone and forced
to do it anyway!   :)



-monz


top of page bottom of page up down


Message: 7983

Date: Wed, 05 Nov 2003 20:52:43

Subject: Re: Eponyms

From: Gene Ward Smith

--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
wrote:

> 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.

Could you list what you think of as the important commas?


top of page bottom of page up down


Message: 7984

Date: Wed, 05 Nov 2003 21:01:14

Subject: Re: Eponyms

From: Paul Erlich

--- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:

> in my "rational canasta" tuning, a secor is:
> 
> <3,5,7,11,13>-monzo = <-4, -1, 0, 1, -1>
> ratio = 5632:5265 
> = ~116.657 cents
> 
> this is only ~0.0101 cent less than 2^(7/72), the
> "standard" secor.

though i know you wanted to use the 72-equal secor, the "standard" 
secor is (as you correctly state on The Proxomitron Reveals... * [with cont.]  (Wayb.)
arts.org/dict/secor.htm) (18/5)^(1/19) = ~116.7156 cents.


top of page bottom of page up down


Message: 7985

Date: Wed, 05 Nov 2003 21:22:20

Subject: Re: Eponyms

From: George D. Secor

--- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> hi George,
> 
> --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" 
<gdsecor@y...> wrote:
> 
> > --- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> > > 
> > > ... even tho *i* will always think of any rational
> > > interval as its monzo.  
> > > 
> > > (i guess that's self-evident, given the name of the term?)
> > > 
> > > ;-)
> > 
> > Oh, dear!  Now I fear that if someone brings up the
> > word "secor" in a tuning context, they'll too readily
> > associate that with the term "irrational".  ;-)
> > 
> > --George
> 
> oh, not necessarily!
> 
> in a series of old posts, most informatively this one:
> 
> Yahoo groups: /tuning/message/23195 * [with cont.] 
> 
> i presented a "rational canasta" tuning.  ...
> 
> and of course many other rational secors could be found.

Probably distant relatives of mine.  ;-)

> ...
> i just wrote "rational" to avoid having to go into
> details like this ... but now you've gone and forced
> to do it anyway!   :)

Hey, you're taking my reply much too seriously.

--George


top of page bottom of page up down


Message: 7986

Date: Wed, 05 Nov 2003 21:38:35

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:
> 
> > 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.
> 
> Could you list what you think of as the important commas?

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?  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".  Dave 
already discussed this:

Yahoo groups: /tuning-math/messages/7320 * [with cont.] 

--George


top of page bottom of page up down


Message: 7987

Date: Wed, 05 Nov 2003 21:46:05

Subject: Re: Eponyms

From: George D. Secor

--- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> hi George,
> 
> --- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" 
<gdsecor@y...> wrote:
> 
> > The monzo comma-naming system is so cumbersome 
> > (i.e., unfriendly), that I can't imagine how anyone
> > could follow a name above the 11-limit unless it's
> > written down.  Imagine trying to mention to someone 
> > in spoken conversation that you're trying to decide
> > whether to use the symbol for "<-6, 0, 0, 0, 0, 0, 0, 1>
> > or <6, 0, 0, 0, 0, 0, 0, 1>" in a composition -- are
> > you expecting me to be mentally prepared to count all
> > those zeros so I know what prime number you mean when 
> > you finally get to the "1" that matters?  
> 
> this is an interesting and good point.
> 
> in fact, over the years of working in extended-JI,
> i've pretty much come to the conclusion that there are
> so many notes available in even a rather compact 11-limit
> euler-genus, that via xenharmonic bridging, that system
> can imply many higher-prime ratios.
> 
> in particular, i've noticed that lots of 11-limit ratios
> sound very similar to nearby 13-limit ratios.

No question about it!  Many years ago I couldn't help noticing that 
the ratio 351:352 vanishes in most of the good edos below 100.

More recently I also found that complex 7-limit ratios tend to *very 
closely* approximate ratios of 13.  The 13-schismina, 4095:4096 or <-
2, -1, -1, 0, -1>, vanishes in most of the best edos between 200 and 
800.  This was the first step in establishing an economy of symbol-
elements (flags) in Sagittal notation development, since it allows 
one to notate a 13M-diesis (1024:1053) as the (approximate) sum 
(product) of a 5-comma (80:81) and 7-comma (63:64) in both medium and 
high-precision JI.  (Only extreme-precision Sagittal JI has separate 
symbols for 13-limit consonances.)

> if one accepts this aspect of my theory, then monzos
> can easily be used to name all the relevant 11-limit
> kommas.

Loocs like you kaught something from Dave Ceenan.  ;-)

> on the other hand, the main reason i came up with the
> idea of using monzos to represent prime-factored ratios
> was that i wanted to avoid both having to always specify
> the primes, and also to avoid superscripts. 
> 
> at the time i originally thought of the monzo idea,
> i was working in 19-limit, and it seemed a easier to me 
> to specify, to pick a totally random example, 133:72 as 
> [-3, -2, 0, 1, 0, 0, 0, 1]  (with the prime-factors 
> 2, 3, 5, 7, 11, 13, 17, 19 understood) than to write
> is out as 2^-3 * 3^-2 * 7^1 * 19^1.

Sure.

> > The problem is, the "name" (if you can call it that)
> > emphasizes *powers* rather than *primes*, so it tends
> > to get rather cryptic.
> 
> when one works with the same set of prime-factors
> over and over again, one very easily gets used to
> remembering the primes which underly the monzo.
> 
> and as i've pointed out before, the monzo allows direct
> visualization of the lattice, which in turn helps in
> comprehension of the structure of the entire tuning system.

It looks as if Dave's and your method of designating ratios are 
complementary.  Dave's (which is at the moment restricted to ratios 
no larger than ~69 cents) helps more in comprehending the size of the 
comma and especially in identifying the Sagittal symbol used to 
notate ratios having the same combination of primes >3.

BTW, have you ever tried collapsing an 11-limit lattice into 2 
dimensions by mapping 11/8 to <10, 5>?

> i came up with the monzo idea based on analogy with
> our regular numbering system.
> 
> it doesn't take too long for one to learn, whether in school 
> or in everyday life, that, for example, the number 133 is a 
> monzo-like representation of (10^2)*1 + (10^1)*3 + (10^0)*3.
> 
> the regular arabic numeral is a nice compact way of expressing
> what would look like a rather complicated mathematical expression
> if it were written out in full.  but once one learns how
> it works, the long version is never need anymore.

Yes, I can readily appreciate this sort of shorthand. However, its 
weakness as a naming system lies in the fact that you need a good way 
to verbalize what you're seeing.  Decimal numbers work as names 
because, in saying a number such as 2,010,030 (as "two million ten 
thousand thirty"), we include words to indicate the placement of the 
numerals.  But this is not as easy for monzos, since primes are not 
related to one another as are powers of ten, since exponents may be 
negative numbers, and since numbers may not be rounded off by 
dropping the lower prime exponents (as with 2.01 million).

Something that you might want to consider is replacement of the comma 
following the powers of 3, 11, and 19 (and every 3rd prime 
thereafter) by a semicolon (to serve as a place marker, similar in 
function to a decimal point and commas in decimal numbers), so that 
133:72 could be written as either [-3, -2; 0, 1, 0; 0, 0, 1] or [-2; 
0, 1, 0; 0, 0, 1].

--George


top of page bottom of page up down


Message: 7988

Date: Wed, 05 Nov 2003 22:24:37

Subject: Re: Eponyms

From: Paul Erlich

--- 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 if this duplicates a previous message; that one didn't seem to 
show up . . .


top of page bottom of page up down


Message: 7989

Date: Wed, 05 Nov 2003 22:02:04

Subject: Re: Eponyms

From: Paul Erlich

--- 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>?

a single "xenharmonic bridge" like this would only collapse the 4-
dimensional 11-limit lattice into 3 dimensions, or a 5-dimensional 
version (with factors of 2 shown) into 4 dimensions, wouldn't it?

the unison vector in this case would be 184528125:184549376, which is 
a good one, but 9800:9801 is both smaller and simpler . . .

maybe you're specifically talking about the lattice with no 7 axis?


top of page bottom of page up down


Message: 7990

Date: Wed, 05 Nov 2003 22:37:28

Subject: Re: Eponyms

From: monz

hi paul,

--- In tuning-math@xxxxxxxxxxx.xxxx "Paul Erlich" <perlich@a...> 
wrote:
> --- In tuning-math@xxxxxxxxxxx.xxxx "monz" <monz@a...> wrote:
> 
> > in my "rational canasta" tuning, a secor is:
> > 
> > <3,5,7,11,13>-monzo = <-4, -1, 0, 1, -1>
> > ratio = 5632:5265 
> > = ~116.657 cents
> > 
> > this is only ~0.0101 cent less than 2^(7/72), the
> > "standard" secor.
> 
> though i know you wanted to use the 72-equal secor, 
> the "standard" secor is (as you correctly state on 
> Definitions of tuning terms: secor, (c) 2001 b... * [with cont.]  (Wayb.)) 
> (18/5)^(1/19) = ~116.7156 cents.



OK, thanks for pointing that out.  so my "rational secor"
is only a little more than half a cent smaller than that.


-monz


top of page bottom of page up down


Message: 7991

Date: Wed, 05 Nov 2003 22:52:35

Subject: Re: Eponyms

From: monz

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.




> BTW, have you ever tried collapsing an 11-limit lattice into 2 
> dimensions by mapping 11/8 to <10, 5>?


no, i never did that.  but i did do this:

Yahoo groups: /tuning/message/1372 * [with cont.] 
Yahoo groups: /tuning/message/1380 * [with cont.] 


(if you're viewing on the stupid Yahoo web interface, 
you'll have to forward them to your email account to 
view the lattices properly.)



> Yes, I can readily appreciate this sort of shorthand.
> However, its weakness as a naming system lies in the
> fact that you need a good way to verbalize what you're
> seeing.


which is the main reason why i agree with you about names
in general.

but i do think that for 11-limit, using only 4 exponents,
it's not so bad.



> Something that you might want to consider is replacement
> of the comma following the powers of 3, 11, and 19 (and 
> every 3rd prime thereafter) by a semicolon (to serve as 
> a place marker, similar in function to a decimal point 
> and commas in decimal numbers), so that 133:72 could be
> written as either [-3, -2; 0, 1, 0; 0, 0, 1] or
> [-2; 0, 1, 0; 0, 0, 1].



well, since that last monzo doesn't use 2, it should be
written (following the convention proposed by Gene and
accepted by me) with angle brackets instead of square:
<-2; 0, 1, 0; 0, 0, 1>.

anyway, yes, that's an excellent idea ... except that
i was never crazy about adding the commas in the first place.
i still think i prefer the nice clean look of a single space
and nothing else, separating the numbers in the monzo.



-monz


top of page bottom of page up down


Message: 7992

Date: Wed, 05 Nov 2003 22:39:33

Subject: Re: Eponyms

From: monz

--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
wrote:

> > i just wrote "rational" to avoid having to go into
> > details like this ... but now you've gone and forced
> > to do it anyway!   :)
> 
> Hey, you're taking my reply much too seriously.



i know ... but still, i thought of it before i wrote
my original post, so i figured that since you brought
it up (albeit as a joke) i might as well give a nice
fat response.  

:)


... isn't it so much nicer when communication here
is this pleasant?



-monz


top of page bottom of page up down


Message: 7993

Date: Wed, 05 Nov 2003 02:36:29

Subject: Some 13-limit TM et bases

From: Gene Ward Smith

9: [21/20, 26/25, 27/25, 33/32, 45/44]

10: [25/24, 28/27, 35/33, 40/39, 49/48]

26: [45/44, 50/49, 65/64, 78/77, 81/80]

29: [49/48, 55/54, 65/64, 91/90, 100/99]

46: [91/90, 121/120, 126/125, 169/168, 176/175]

58: [126/125, 144/143, 176/175, 196/195, 364/363]

111: [176/175, 351/350, 540/539, 676/675, 1331/1323]

130: [243/242, 351/350, 364/363, 441/440, 3136/3125]


top of page bottom of page up down


Message: 7994

Date: Thu, 06 Nov 2003 15:48:41

Subject: Re: Eponyms

From: George D. Secor

--- 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.

I agree with Gene that using the conventional spelling for the new 
(defined-range) meaning of comma and the unconventional spelling for 
the commonly accepted (i.e., generic) meaning for a comma would not 
be a good idea, so I advise that the new spelling be dropped.  (It's 
not required for a comma-naming system to work, anyway.)

> > ...
> > Something that you might want to consider is replacement
> > of the comma following the powers of 3, 11, and 19 (and 
> > every 3rd prime thereafter) by a semicolon (to serve as 
> > a place marker, similar in function to a decimal point 
> > and commas in decimal numbers), so that 133:72 could be
> > written as either [-3, -2; 0, 1, 0; 0, 0, 1] or
> > [-2; 0, 1, 0; 0, 0, 1].
> 
> well, since that last monzo doesn't use 2, it should be
> written (following the convention proposed by Gene and
> accepted by me) with angle brackets instead of square:
> <-2; 0, 1, 0; 0, 0, 1>.

I used the square brackets to make the point that you would no longer 
need the angle brackets in order to show that the exponent of 2 is 
omitted.  But perhaps the semicolon looks too much like a comma, so 
that the angle brackets would be more distinctive in indicating this 
(however, see the following).

> anyway, yes, that's an excellent idea ... except that
> i was never crazy about adding the commas in the first place.
> i still think i prefer the nice clean look of a single space
> and nothing else, separating the numbers in the monzo.

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.

--George


top of page bottom of page up down


Message: 7995

Date: Thu, 06 Nov 2003 16:21:40

Subject: Re: Eponyms

From: monz

hi George,

--- In tuning-math@xxxxxxxxxxx.xxxx "George D. Secor" <gdsecor@y...> 
wrote:

> I agree with Gene that using the conventional spelling for
> the new (defined-range) meaning of comma and the unconventional
> spelling for the commonly accepted (i.e., generic) meaning for
> a comma would not be a good idea, so I advise that the new
> spelling be dropped.  (It's not required for a comma-naming
> system to work, anyway.)


OK.


>
> <snip>
>
> [regarding the format of monzos:]
>
> ... 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 that a lot!

in fact, i find it very interesting that group the primes
by threes like this also keeps them in bunches that make
sense to me in terms of how i've used them and theorized
about them myself!

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).

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.)

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

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

interesting.



-monz


top of page bottom of page up down


Message: 7996

Date: Thu, 06 Nov 2003 17:19:24

Subject: Re: Eponyms

From: Paul Erlich

--- 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).
> 
> 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.)
> 
> the next comma appears after 19, which i myself used as
> a limit from about 1988-98.
> 
> the next comma appears after 31, which is the highest limit
> Ben Johnston has used in his music.
> 
> interesting.

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?


top of page bottom of page up down


Message: 7997

Date: Thu, 06 Nov 2003 17:25:44

Subject: Re: Eponyms

From: Paul Erlich

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 if this duplicates a previous message; that one didn't seem 
to 
> show up . . .


top of page bottom of page up down


Message: 7998

Date: Thu, 06 Nov 2003 19:25:22

Subject: Re: Eponyms

From: Gene Ward Smith

--- 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.


top of page bottom of page up down


Message: 7999

Date: Thu, 06 Nov 2003 19:54:55

Subject: Re: Eponyms

From: Paul Erlich

--- 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

Previous Next Last

7000 7050 7100 7150 7200 7250 7300 7350 7400 7450 7500 7550 7600 7650 7700 7750 7800 7850 7900 7950

7950 - 7975 -

top of page