Mapping MIDI input to scale

Hi-
 
I want to convert MIDI notes to a special scale. (For those who are interested: I want to automatically create Tintinnabuli voices a la Arvo Pärt.)
 
This means for example to map every note of a scale to the lower next chord note. Example A minor:
 
a --> e
b --> a
c --> a
d --> c
e --> c
f --> e
g --> e
 
In Sonar I've found two tools:
 
1. MIDI effect "Transpose". I can create a custom scale that does the trick when I play a midi track. It does not affect the MIDI Input though, that means if I play on a keyboard and route the MIDI track output to a software instrument, the effect doesn't get applied. Also, it's pretty tedious to change each and every entry across seven octaves instead of just defining the map for one octave. And, is there a way to import / export maps? I finally overwrote one of the existing presets just to save the map.
 
2. Snap to Scale (track inspector). Looks very good but it seems it only affects what you draw into the piano roll view. It doesn't affect playback and of course not the MIDI Input.
 
So, any ideas? The biggest improvement for me would be if it worked in real time on the MIDI input.
 
wolf
 
 
 
 
 
 
 
 
 

A drum map may be used to transpose note data too. Drum maps are not just for drums. While it is necessary to define each note relation in the map it may be faster than the transpose tool.

Thanks a lot, I tried the drum map idea and it works great.
Is there an easy way to copy the mapping definitions from one octave to another? It's a little tedious to do click together the same stuff across the keyboard range.

azslow3
For static mapping the way described by brundlefly is better.
 But if you ever written simple programs, there is "Drum map for programmers", see the last example in http://www.azslow.com/index.php/topic,286.0.html

 
I tried it out and got it to work with Sonar X1. This looks more comfortable than the other solutions. Yet, I was wondering whether you need to spell out the mapping across all octaves?
I have no programming skills (save for the programmable pocket calculator I had as a youth) so it would take me a while to do more than just copying code here. But in general, I guess it is possible to add switches to transpose up / down 1 octave and also to define certain scales (by activating one of several Definition blocks). Is there an easy way to get this functionality, that is without really understanding what I do? I mean modifying your example so it does what I need was very easy.

Thanks a lot for your offer.
 
The System behind this is very easy. For each of the 7 note of a scale (melody voice: M), the harmony (Tintinnabuli: T) voice consists of one of the three notes of the respective chord. There are four different Tintinnabuli voices: the upper next chord note (T+1), the 2nd chord note up from the melody note (T+2), the lower next chord note (T-1), and the 2nd lower note (T-2).
 
For the notes of the A minor scale, the T voices are:
 
M    T-2  T-1  T+1  T+2
------------------------------
a4   c3   e3   c4   e4
b4   e3   a4   c4   e4
c4   e3   a4   e4   a5
d4   a4   c4   e4   a5
e4   a4   c4   a5   c5
f4   c4   e4   a5   c5
g4   c4   e4   a5   c5
a5   c4   e4   c5   e6
 
Well I HOPE I there's no mistake in the table.
So, the ideal MIDI effect would look like this: one switch to select the scale (if you want to have it complete, that would be 1 switch or selector for the 11 different root notes, c, c#, d, d#, and so on that define the root of the scale, another switch for Major or minor, and a 3rd Switch to shift the T voice one octave up or down. The latter can be had with the already existing transpose effect in Sonar though. Also, 7 (c, d, e, f, g a b) instead of 11 root notes would be sufficient too. Or, to be even simpler for the start, just use A minor and C major.
 
All right, summary: the simplest version would comprise switching between A minor and C Major and a switch to select one of the 4 T voices, T-2, T-1, T+1, and T+2.
 
With switch I mean you can make the selections in the GUI of the effect, no need to switch while playing.
 
What I want is to get a feeling for the melodic colors  the different T voices add. For real compositions, which is the next step, I guess such a restricted pattern would get boring in no time.  Therefore, it doesn't make sense to put very much work and time in such a tool.
For Arvo Pärt, it was the starting point. "Für Alina" is a composition that is really made up of nothing more than one M and one T voice:
https://www.youtube.com/watch?v=TzIZPZN5K60
 
Also, forgot to ask that yesterday, can you save Lua programs in a file outside of Sonar?
 
 
 

Thanks a lot and again, I'll check it out soon and let you know how it works.
 
Again, how do I load SPP files and how do I save them to disk?

azslow3
Not exactly according to your table, but I think that is what you have asked for:
http://www.azslow.com/index.php/topic,286.msg1479.html#msg1479

The audio card of my music PC had been acting up so I was a little late checking it out. It works great. I need to check out some particular notes that sound a little odd but in general it's a very big help.