KVR Audio

Dunks · Post by **Dunks** » Wed Jan 23, 2008 4:26 am

I'm trying to find out what the values of the notes from c#1 > B-1 would be if C-1 = 0 and C-2 = 100. I have looked on wiki and in this forum but still not sure are all the semitones equal in length or different, what do modern day hosts use?

bmrzycki · Post by **bmrzycki** » Wed Jan 23, 2008 5:54 am

Are you talking about MIDI note numbers? MIDI note numbers are designed around 12-Tet Equal Temperment with "Middle C" defined as number 60. This has caused confusion between hardware and software vendors as described here.

In equal temperament 12-Tet the octave is split into 12 pieces called semitones. This is a good approximation but isn't perfect. You can read more here to dive into this.

I hope this helps and at least one link answers your questions.

Dunks · Post by **Dunks** » Wed Jan 23, 2008 7:14 am

bmrzycki wrote:Are you talking about MIDI note numbers?

I was talking about software. 12-Tet Equal Temperment is that what most audio software use, like tracktion, renoise ect ect...? Thanks for helping.

Dunks · Post by **Dunks** » Wed Jan 23, 2008 9:02 pm

Bump.

nuffink · Post by **nuffink** » Wed Jan 23, 2008 9:06 pm

I'm not entirely sure, but...

Are you talking about cents (the hundredth division of a semitone) and the degree of difference equal temperament shows from just temperament and all that jazz?

If so... http://en.wikipedia.org/wiki/Equal_temperament

Edit: Oops, that's the same link as bmrzycki so I guess not.

jancivil · Post by **jancivil** » Wed Jan 23, 2008 9:31 pm

Dunks wrote:I'm trying to find out what the values of the notes from c#1 > B-1 would be if C-1 = 0 and C-2 = 100. I have looked on wiki and in this forum but still not sure are all the semitones equal in length or different, what do modern day hosts use?

in software, all the semitones will be equal, unless you have hacked something to where it doesn't; in theory equal temperament is just that, equal. however, many piano tuners use a technic called stretch tuning, at extreme ends of range, bass/treble. (the only 'pure' interval in 12-tone et is the octave, the rest are compromised)

in acoustically derived intonations, the ratii are integers, an octave (this holds true in 12-et) is 2:1, a fifth, this is a pure fifth, is 3:2; pythagorean intonations derive a Tone, 9:8 and a Bitone, 81:64 (which is a syntonic comma 81:80 sharp from a pure third (major) 5:4.) by process of multiplication. (3:2 squared = 9:4, fith + a fith, w. that 'octave + a tone' result transposed down to = a Tone, 9:8; squared = 81:64)
Tempered intonations are designed to allow modulations and hose ratii are 'irrational', derived from square-rooting 12.

I guess as to your '100' (your interval appears to be a major 9th (octave + a tone) distance, which is 1/6th greater than what appears to be an octave)?? I don't know where you start with '100' - except

every semitone is called 100 cents distant from its neighbor.
[in 24-tone et, that quartertone would be 50 cents in value.]

so, if c#1 is 0, c#0 is -1200, b-1 is -1400

it's useful in a sequencer to know that 1 represents a semitone and 12 represents an octave in terms of modulation in 12-et. (major third = 4; perfect fifth = 7)

Dunks · Post by **Dunks** » Wed Jan 23, 2008 9:49 pm

I think I will use equal temperament.

I think I described this wrong, if a octave = 100 them a semitone would = 8.33333333 100 divided by 12 is this correct?

jancivil · Post by **jancivil** » Wed Jan 23, 2008 9:59 pm

sumthin like that

but I don't see where that gets you, I know of no convention that uses '100' for an 8ve...
intonations use ratio to compare intervals.

Twelve-tone equal temperament:

/ 12 equal parts, the ratio of frequencies between two adjacent semitones is the twelfth root of two:

r = \sqrt[12]{2} \approx 1.05946309:1

compared with 16:15 ['minor second' in 'just' intonation], or 256:243 (or 135:128) in Hindustani systems...

Dunks · Post by **Dunks** » Wed Jan 23, 2008 11:16 pm

It may seem strange, I am trying to map semitones onto a pitch slider. The slider has 100 steps 0-100, 100 being twice as fast as 0.

jancivil · Post by **jancivil** » Thu Jan 24, 2008 12:05 am

ok now I'm hooked.

would this be --
a resolution of 100 for a semitone, now that I could use.

jancivil · Post by **jancivil** » Thu Jan 24, 2008 12:10 am

although, now I think of it, the MIDI convention would seems like it'd be 0-127 for a semitone.

all I know is that the list editor in cubase has some godawful high numbers when I've done a pitch bend over, say, a fourth.

Dunks · Post by **Dunks** » Thu Jan 24, 2008 12:20 am

Basically I need to fit 12 notes on a slider that go's from 0-100% 0 = basenote 100 is 1 octave higher.

Meffy · Post by **Meffy** » Thu Jan 24, 2008 12:24 am

Ah! That clarifies things a great deal.

Were this my project I'd use just the values from 0 to 96, so each step could be one eighth of a semitone. If you must map the entire 0 to 100 range, the intermediate values aren't going to hit exact notes in the conventional Western (TET) scale. Or any other scale that's in real-world use, for that matter.

Dunks · Post by **Dunks** » Thu Jan 24, 2008 12:33 am

It has to be 100, but I can put a few decimal places to get as close as possible.

jancivil · Post by **jancivil** » Thu Jan 24, 2008 12:44 am

I am having a tough time seeing the musical application of 100-tone et. it's far too coarse for pitch bend..

is this a programming sort of exercise, strictly speaking?

Semitones help please.