Indeed, I remember now that there are different standard for this.
Hopefully this is right now. My brain sweats.
Edit: I made my post editable so that you can adjust values if you find mistakes.
1 Like
Middle C is MIDI note 60 but it’s 261.6Hz, not 264Hz.
The numbers definitely aren’t MIDI notes though, that’s probably just a coincidence. MIDI notes would be 12 numbers per octave.
1 Like
Maybe we got slightly off track here.
There are 128 CC “units”, and these are used all over the machine for different purposes and value ranges. For EQ they are spread across 0 to 20KHz range. Of course this somehow translates to octaves as well but that’s secondary and also not to be expected to match perfectly to 10, 20, etc. The listed notes are good enough and useful for orientation, and that’s fine.
1 Like
Yeah, but somebody at electron must have decided how they were going to work.
I’m just trying to figure it out (like a puzzle).
The fact that it neatly doubles every ten values instead of every twelve is interesting. Ten is neater but twelve seems more useful, musically.
I suspect the mathematical value that popped out of the equation for “127 mapped to human hearing range” was somewhere around eleven and they decided to round it down instead of up.
Personally I am mostly annoyed about losing nearly a third of that 128 range just to roll off the bass, but I’ll have to live with that 
Apparently what’s wrong is a failure in the arithmetic unit of my brain.
1 Like
This is something a lot of filter devs do - there’s a most likely a mathematical curve to where a CC/pot value lands on the filter frequence according to what sounds good/musical with that particular filtermodel
Get out your calculator/excel/slide rule and run the numbers, the filter cutoff frequency is 440 Hz * 2^((n - 69)/12) where n is whatever value the filter cutoff is set to in the range of 0 to 127.
@LyingDalai if your table is supposed to be all Cs then it should go in steps of 12 not 10.
2 Likes
Totally agree with you. Something is wrong here. But my brain is melting so I set my post in edit mode so that anyone can get this table right. Please verify on the device.
And let’s not forget that each octave the frequency is multiplied by 2, which necessarily bring its lot of decimal frequencies for each of the 11 notes in-between
Please get this table full if you can !
@Elektron can you provide a list, please.
FWIW my reading of igor’s post with the list (and it’s guesswork I’ll admit) is that it was something someone discovered by experiment and shared. Unless we have concrete information from elektron, or experimental evidence otherwise, I don’t see the reasoning behind ‘it must be steps of 12 not 10’ ?
Though it has to be said that its interesting the frequency of middle C (261.6 concert, 256.0 scientific) is pretty close to 264 in the chart which has an assigned number which happens to be the midi encoding of middle C.
EDIT: Hmmm I just discovered that the comb filters on DT2 , which are known to jump in octaves with FUNC , go from 48, 60, 72, 84, all tuned to a C. It would actually be odd if the above table is correct with its steps of 10.
The filter frequencies correspond directly to MIDI note values. If you turn on keyboard tracking it becomes a little more complex, it seems to tilt around 48. That means if you set the filter to 48 and keyboard tracking to 100% the filter cutoff is exactly the frequency of the currently played not, set the cutoff to 60 and you get twice the note frequency, 72 you get four times the note frequency and so on.
1 Like