How does 7 bit MIDI CC map to percentage in Renoise?


(Artie Fufkin) #1

I’ve got a 7 bit [0-127] MIDI CC assigned to a Hydra Input and the Hydra output is assigned to a Send’s Receiver parameter. There are 128 send tracks.

Here’s is the mapping I’m getting:

CC: 0, 1, 2, 3, 4, … 127

Percent: 0, 0.787, 1.575, 2.362, 3.15, … 100

Track: S1, S1, S2, S4, S4, … S128

Is there a way I can use the formula device and/or other map modes(Relative Signed bit, Relative signed bit 2, etc) to get an arithmetic sequence for the send receiver, to match the MIDI value?


#2

One of your questions is: How does 7 bit MIDI CC map to percentage in Renoise?

Off the top of my head Artie (I have no idea how Renoise does it internally btw), I suppose I’d scale the MIDI CC number by 127 and multiply by 100?

Percentage = (CC / 127) * 100

Producing the following table:

CC % CC % CC %
---------------------------------------------
000 - 0.000 048 - 37.795 096 - 75.591  
001 - 0.787 049 - 38.583 097 - 76.378  
002 - 1.575 050 - 39.370 098 - 77.165  
003 - 2.362 051 - 40.157 099 - 77.953  
004 - 3.150 052 - 40.945 100 - 78.740  
005 - 3.937 053 - 41.732 101 - 79.528  
006 - 4.724 054 - 42.520 102 - 80.315  
007 - 5.512 055 - 43.307 103 - 81.102  
008 - 6.299 056 - 44.094 104 - 81.890  
009 - 7.087 057 - 44.882 105 - 82.677  
010 - 7.874 058 - 45.669 106 - 83.465  
011 - 8.661 059 - 46.457 107 - 84.252  
012 - 9.449 060 - 47.244 108 - 85.039  
013 - 10.236 061 - 48.031 109 - 85.827  
014 - 11.024 062 - 48.819 110 - 86.614  
015 - 11.811 063 - 49.606 111 - 87.402  
016 - 12.598 064 - 50.394 112 - 88.189  
017 - 13.386 065 - 51.181 113 - 88.976  
018 - 14.173 066 - 51.969 114 - 89.764  
019 - 14.961 067 - 52.756 115 - 90.551  
020 - 15.748 068 - 53.543 116 - 91.339  
021 - 16.535 069 - 54.331 117 - 92.126  
022 - 17.323 070 - 55.118 118 - 92.913  
023 - 18.110 071 - 55.906 119 - 93.701  
024 - 18.898 072 - 56.693 120 - 94.488  
025 - 19.685 073 - 57.480 121 - 95.276  
026 - 20.472 074 - 58.268 122 - 96.063  
027 - 21.260 075 - 59.055 123 - 96.850  
028 - 22.047 076 - 59.843 124 - 97.638  
029 - 22.835 077 - 60.630 125 - 98.425  
030 - 23.622 078 - 61.417 126 - 99.213  
031 - 24.409 079 - 62.205 127 - 100.000 
032 - 25.197 080 - 62.992		 
033 - 25.984 081 - 63.780
034 - 26.772 082 - 64.567
035 - 27.559 083 - 65.354
036 - 28.346 084 - 66.142
037 - 29.134 085 - 66.929
038 - 29.921 086 - 67.717
039 - 30.709 087 - 68.504
040 - 31.496 088 - 69.291
041 - 32.283 089 - 70.079
042 - 33.071 090 - 70.866
043 - 33.858 091 - 71.654
044 - 34.646 092 - 72.441
045 - 35.433 093 - 73.228
046 - 36.220 094 - 74.016
047 - 37.008 095 - 74.803

Likewise (with a spot of Algebra) you could rearrange the formula to calculate a rough CC from a given percentage:

RoughCC = (Percentage / 100) * 127

From -> https://forum.renoise.com/t/midi-footcontroller-for-live-guitar-multi-fx/49657

And this table ->

CC Percent Track
0 0 1
1 0.787 1
2 1.575 2
3 2.362 4
4 3.15 4
5 3.937 6
6 4.724 7
7 5.512 8
8 6.299 8
9 7.087 9
10 7.874 11
11 8.661 12
12 9.449 13
13 10.236 13
14 11.024 15

I think you are asking why you are getting those Track values in the Send Receiver when you set the Hydra Min/Max range to 0 and 127? Just out of my own curiosity: do you get a more consistent result if you set the Min/Max range in Hydra to Min:0 and Max:128?


(Zer0 Fly) #3

Try this formula in the *formula device:

(A*127)/255

it should translate 7bit cc to send channel selection in #send device


(Artie Fufkin) #4

Ah ok I had an off-by-one error with the divisor.

4Tey, your table is exactly what Renoise is producing.

I think the order in which I created send tracks may have affected which ones were assigned(I was hoping to be able to switch through the sends, wrapping around at 128, effectively ignoring sends beyond 128 so that I could use them differently), or perhaps the Receiver list wasn’t updating…?.. So I started over and created the 128 tracks and nowHydra Min/Max range to 0 and 127 is working!

I was switching the max on the Hydra and getting the “Linked Send Track Does Not Exist” even for a track that existed, which is why I speculated that the Receiver list might not update on certain changes.

Zer0 Fly, that formula is producing the skips and repeats I was getting before.

(A*128)/256 works though. Thank you

The last part I would want to do with this is scale it down to less tracks(4 to 8) and still have it wrap around.

I would think I’d have to do some modular arithmetic, but I haven’t quite figured out what to do or how yet.


#5

I think the order in which I created send tracks may have affected which ones were assigned(I was hoping to be able to switch through the sends, wrapping around at 128, effectively ignoring sends beyond 128 so that I could use them differently), or perhaps the Receiver list wasn’t updating…?.. So I started over and created the 128 tracks and nowHydra Min/Max range to 0 and 127 is working!

I was switching the max on the Hydra and getting the “Linked Send Track Does Not Exist” even for a track that existed, which is why I speculated that the Receiver list might not update on certain changes.

Just a bit of speculation: But I think you’ve uncovered a subtle ‘bug’ (bug is too strong a word though) Artie in Renoise due to some internal rounding error when it comes to assigning say some Hydra output to that send receiver. Let’s say it ‘skips’ send tracks with 127 assigned in max Hydra. If you momentarily increase the max value from 127 to 128, you’ll find that it probably ‘works’ (albeit with that ‘send track doesn’t exist (even though it does)’ warning. Then if you knock the max value in Hydra back down to 127, it’ll now suddenly ‘work’ more as expected. Basically you can’t fully trust Renoise here and is kinda confusing the situation. You are using a floating point value to control/select from a (integer) list. You see what you want Artie is to put in the number say integer 5 from a MIDI CC, go through Hydra with its floating point division and multiplication and scaling from 0 to 255 etc, and output the number exactly 5 to the send receiver list. Unfortunately that doesn’t fully happen, maybe sometimes you get the value 4.46543 internally, in which case it’ll probably convert to the number 4, hence the ‘skips’ and ‘repeats’. Sometimes though you can get it so that the rounding plays ball by working more evenly with numbers like 128 and 256 rather than 127 and 255. IDK Artie, but this is just some quick speculation on my part :slight_smile:


(Zer0 Fly) #6

I am very sorry for the sloppy job.

Try the following formula:

(A*127-0.5)/255

It should produce no more skips and jumps or the like.

The problem is (x+1)/255 defines not the selected channel for the send device. It defines the thresholds between the selected states for the analog “x” floating point value. The first formula would target directly at the threshold, and generate flaky results because of rounding errors. The “0.5” factor corrects it by placing the results right in the middle between the thresholds.

Fun fact: you can make renoise have more send channels than are selectable with the send device. duh…

Edit, for your wraparound with the first 8 channels do something like:

(mod(A*127,8)-0.5)/255

for wrapping around channels 4-10 (6 channels selected starting from channel 4):

(mod(A*127,6)+4-0.5)/255