DSPRelated.com
Forums

Theory of dissonance - critical bandwidth, frequency discrimination

Started by jon222 May 3, 2007
Hi,

I need help with theory about dissonance. 
I'm looking for a formula for critical bandwidth function and frequency
discrimination function. ??
I have found Plomp theory with graphs, but no exact formulas for counting
it for frequencies.

Thanks
J.

_____________________________________
Do you know a company who employs DSP engineers?  
Is it already listed at http://dsprelated.com/employers.php ?
On 3 Mai, 11:57, "jon222" <s...@kis.fri.uniza.sk> wrote:
> Hi, > > I need help with theory about dissonance.
What is the definition of "dissonance"? The only one I can remember off the top of my head is in the context of audio, "two tones which sound bad when played together." Is this what you are thinking of? Rune
>On 3 Mai, 11:57, "jon222" <s...@kis.fri.uniza.sk> wrote: >> Hi, >> >> I need help with theory about dissonance. > >What is the definition of "dissonance"? The only one I can remember >off the top of my head is in the context of audio, "two tones which >sound bad when played together." Is this what you are thinking of? > >Rune > >
Yes, I mean exactly that . Any help? J. _____________________________________ Do you know a company who employs DSP engineers? Is it already listed at http://dsprelated.com/employers.php ?
Rune Allnor wrote...
> What is the definition of "dissonance"? The only one I can remember > off the top of my head is in the context of audio, "two tones which > sound bad when played together." Is this what you are thinking of?
It's difficult to give a purely objective definition. Best attempt is possibly from Helholtz' theory of 'Kangverwandtschaft' (relationship of sounds which suggests that two tones are consonant if their harmonics (up to the 8th but excluding the 7th) have one or more tones in common. David
>Rune Allnor wrote... >> What is the definition of "dissonance"? The only one I can remember >> off the top of my head is in the context of audio, "two tones which >> sound bad when played together." Is this what you are thinking of? > >It's difficult to give a purely objective definition. > >Best attempt is possibly from Helholtz' theory of 'Kangverwandtschaft'
(relationship of sounds which
>suggests that two tones are consonant if their harmonics (up to the 8th
but excluding the 7th) have
>one or more tones in common. > >David > >
Can you give me some source/book for explanation of the Helmholz' theory? Do you know some theories only about simple tones? Tkanks J. _____________________________________ Do you know a company who employs DSP engineers? Is it already listed at http://dsprelated.com/employers.php ?
On Fri, 04 May 2007 05:15:33 -0500, jon222 wrote:

>>On 3 Mai, 11:57, "jon222" <s...@kis.fri.uniza.sk> wrote: >>> Hi, >>> >>> I need help with theory about dissonance. >> >>What is the definition of "dissonance"? The only one I can remember >>off the top of my head is in the context of audio, "two tones which >>sound bad when played together." Is this what you are thinking of? >> >>Rune >> >> > > Yes, I mean exactly that . > Any help?
Not specifically, and not personally, but regular contributor r b-j has published (here) a beautiful treatise on the relationship between even-tempered scales, harmonies, human hearing and consonance, which (a) google would probably be happy to point to and (b) probably has something to say about the dissonance of non-consonant tones, but I don't remember exactly. Cheers, -- Andrew
David Lee wrote:
> Rune Allnor wrote... >> What is the definition of "dissonance"? The only one I can remember >> off the top of my head is in the context of audio, "two tones which >> sound bad when played together." Is this what you are thinking of? > > It's difficult to give a purely objective definition.
What's more,some intervals once thought dissonant are used by later composers without raising eyebrows.
> Best attempt is possibly from Helholtz' theory of 'Kangverwandtschaft' > (relationship of sounds which suggests that two tones are consonant if > their harmonics (up to the 8th but excluding the 7th) have one or more > tones in common.
I think that's not much different from the observation that consonant fundamentals are related by small whole numbers -- we have learned to make allowances for tempering. 3:2 (a fifth) is consonant, while 27:24 (approximately a whole tone) is dissonant. Jerry -- Engineering is the art of making what you want from things you can get. &macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;
Jerry Avins wrote:

> ... while 27:24 (approximately a whole tone) is dissonant. ...
OOPS! That should be 27:22. the numbers don't need to be that big. 11:9 is also nearly a whole tone. Jerry -- Engineering is the art of making what you want from things you can get. &macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;
Jerry Avins wrote:
> Jerry Avins wrote: > >> ... while 27:24 (approximately a whole tone) is dissonant. ... > > OOPS! That should be 27:22. the numbers don't need to be that big. 11:9 > is also nearly a whole tone.
Bah! I finally pulled out my slide rule. A whole tone step up is 1.1225 or thereabouts. That is indeed near 27:24 (9:8). I thought I was wrong, but I was mistaken. Jerry -- Engineering is the art of making what you want from things you can get. &macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;&macr;
On May 4, 10:55 am, Andrew Reilly <andrew-newsp...@areilly.bpc-
users.org> wrote:
> On Fri, 04 May 2007 05:15:33 -0500, jon222 wrote: > >>On 3 Mai, 11:57, "jon222" <s...@kis.fri.uniza.sk> wrote: > >>> Hi, > > >>> I need help with theory about dissonance. > > >>What is the definition of "dissonance"? The only one I can remember > >>off the top of my head is in the context of audio, "two tones which > >>sound bad when played together." Is this what you are thinking of? > > >>Rune > > > Yes, I mean exactly that . > > Any help? > > Not specifically, and not personally, but regular contributor r b-j has > published (here) a beautiful treatise on the relationship between > even-tempered scales, harmonies, human hearing and consonance,
thanks for the plug, Andrew. i think i know what post you mean, but...
> which (a) google would probably be happy to point to
... unfortunately Google is NOT happy to do that. according to Google Groups search, i haven't posted anything to comp.dsp before Feb. 7 2007. my Google Groups profile seems to know i've been here before, but searching doesn't do it.
> and (b) probably has something > to say about the dissonance of non-consonant tones, but I don't remember exactly.
the theory of the measure of consanace or dissonance of an interval of tones, i think, is to find the ratio of integers that matches the frequency ratio within some limit of tolerance (say within 0.35%, or 1/200 octave, which i seem to remember is about a "just noticable difference" in pitch, like a dB is to loudness). so given a musical interval, find a rational number that equals the frequency ratio of that interval plus or minus 1/200 octave such that the integer denominator of the reduced or simple rational number is minimum. (the numerator value will be between the denominator and twice the denominator if the interval is less wide than an octave.) the more minimum that denominator is, the more consanant the interval is. that's my spin on it. if someone can get Google to dredge up that post (from a few years ago), i would appreciate it. r b-j