Analogue-sequencer

On Thu, 16 Sep 2004 23:27:59 -0000, ch.³l <oozi@oozi.demon.nl> wrote:

>hi Paul,

hey hey..

>hm..maybe you can give me a hand as well, with the following; on 
>switching to another bank, i want to send 3 program changes to my 
>nord modular, on 3 channels which are P3 tracks 5,6 and 7, but i only 

Ah, the Nord Modular. This aspect of it was the single biggest pain
for me live. Thank heavens the G2 has a proper multitimbral setup
mode! My solution was to seek out an Akai ME25S which can send 4
program changes from a single program change. Seem to remember it had
some issues too. The P3 has a means of sending a single program change
on the channel of your choice when the bank is loaded. I'm sure you
know this. In order to send several program changes then mute those
events automatically... ugh, complication. 

>want to send them once, as i'd like to be able to tweak the knobs 
>without the voice resetting every time the program change message 
>comes around. as i use voices from different banks in the NM in 
>different P3 banks, i also need to send 3 'bank select' messages to 
>the NM, which is cc#32. so what i'm doing is using an unused track on 

Is it a combination of 32 and 0, I forget. Too many companies do bank
select differently... 8-(

>the P3, which i can turn off as soon as the program changes are sent. 
>on this track, i set aux A to 'auxes to track N', aux B to cc#32 and 
>aux C to the 'send midi pgm' event. on aux A, B and C the first 3 
>steps are active, with all the values set correctly; A1=5, A2=6, 
>A3=7, B1=5, B2=5, B3=5 (all voices in NM bank 6 in this case), C1=12, 
>C2=16, C3=26. the B and C values are correct, as they work when i put 
>them in the tracks they're being sent to directly. nothing happens 
>though; none of the messages seem to get through, and the NM pretty 
>much freezes up. i've also tried staggering the B and C steps so you 
>get A1&2=5, A3&4=6, A5&6=7, B1=5, B3=5, B5=5, C2=12, C4=16, C6=26, 
>with the same lack of result. lastly, i've tried setting up aux B 
>as 'aux D = cc#n' with the n value set to 32 and steps 1, 2 and 3 
>active, aux C set to 'aux D = event n' with n set to 127(= send midi 
>pgm) and steps 4, 5 and 6 active and aux D left at cc#0 and 
>programmed to the correct values...also tried this with aux B&C 
>staggered, but in both cases still no effect. 
>any ideas? 

Only that this is one of those things I'd probably avoid. I haven't
experimented with sending bank select messages, let alone several at
once on several channels. I'll have a potter about later and see if I
can think of anything. It could be a timing thing - and you're sure
this version of bank select doesn't require CC0?

>the best option would of course be (a user-config option?) to 
>automatically send a program change for each track on switching P3 
>banks...

Or designate an "initial event" to be sent rather than the current
single program change. The initial event could be complex up to a
certain limit and include chains of MIDI events on whatever channel
you wanted. Of course this would take space and require a user
interface and.... er.... best wait until Colin reads this ... 8-)

>...and while i'm at it, here's another one; is there a way to get a 
>negative 'random' value with the 'Rndmz note/velo/len/delay/auxB/C/D' 
>events? as they all seem to be adding to the value, not subtracting 
>from it. it's be nice to have a middle value programmed and then be 

I thought it would be cool to have an "Add value to Aux x" where the
value added could be positive, negative, random, a mathematical
operation...  but I tend to float this kind of thing by Colin and he
just sighs wearily, but patiently... we can ask...

>able to get a live random floating around this middle by setting aux 
>A as 'knob N to auxC' where aux C is a 'Rnd mask auxD' event so a 
>knob N is in control of the randomness being active or not, and 
>setting aux B as 'knob N to auxD' where aux D is, say, a 'Rndmz 
>length' event so another knob N is in control of the measure of 
>randomness. 

Nice idea.

>i'm guessing i'd have to do something with the accumulators to get 
>negative random values here, but i haven't figured it out yet..
>once again, any ideas?

Only that this is where I'd look too. I would love a means of adding
the accumulator (even if negative) to any aux value. 

Sorry, as usual I was no help... 8-)

Paul
---
Paul Nagle / Soft Room Music
Email: paul@softroom.co.uk www.softroom.co.uk
                           www.BogusFocus.com