posted 08 April 2006 14:06            

but I just don't get why the compiler doesn't let me do this!!!

| trig rnd |
rnd := (1 bpm: !BPM) nextRandomWithSeed: 55.
trig := (( 1 toggleWhen: (rnd gt: 0.9)) countTriggersMod: 2) of: #(1 -1).
trig negative
IfTrue: [
trig := (( 1 toggleWhen: (rnd gt: 0.3)) countTriggersMod: 2) of: #(0 1)
].

(default * trig) smooth: 0.5s

posted 08 April 2006 16:36            

| r |

r := (1 bpm: !BPM) nextRandom abs.
default * (((r gt: 0.3) true: ((r gt: 0.9) true: -1 false: 1) false: 0) smooth: 0.5 s

This generates a new random number (-1,1) on each beat and takes the absolute value so the range is (0,1).

Depending on the value of the random number, choose a multiplier for default:

posted 08 April 2006 16:38            

but!

i have now refined it to this, which seems to be working syntax wise and I understand it..

BUT... it doesn't ever seem to enter into the ifTrue block ( i have put some exagerrated values in the fwd: array , but its never entering that block... ermm...

| bkwd rnd rnd2 pulse|
pulse := (1 bpm: (!BPM * 8)).
rnd := pulse nextRandomWithSeed: 55.
rnd2 := pulse nextRandom.
bkwd := (( pulse toggleWhen: (rnd ge: 0.90)) countTriggersMod: 2) of: #(1 -1) .

(bkwd = -1)
ifTrue: [ | fwd |
fwd := (( pulse toggleWhen: (rnd2 ge: 0.10)) countTriggersMod: 2) of: #(2 -2) .
(default * fwd) smooth: 0.2s .

]

ifFalse:[
(default * bkwd) smooth: 0.2s
].

posted 08 April 2006 16:41            

edit
"| r |
r := (1 bpm: !BPM) nextRandom abs.
default * (((r gt: 0.3) true: ((r gt: 0.9) true: -1 false: 1) false: 1) smooth: 0.5 s)"

gives me the sort of result i was after, thanks

posted 08 April 2006 16:44            

-=-
| direction |

direction := ((1 bpm: !BPM) nextRandom abs * 10) of: #(0 0 0 1 1 1 1 1 1 -1).
default * direction smooth: 0.5 s.

-=-

posted 08 April 2006 16:52            

quote:

---

Originally posted by SSC:
An alternative approach is to create a "dartboard" where some numbers take up a larger "area" of the dartboard and are therefore easier to hit with your random dart.

---

A-ha! Random Darts and a Drunk DJ - Sounds like a quality night out to me!

posted 09 April 2006 05:31            

there's still something's unclear

I'll go back to my array indexing example - it has a different behavior to random darts style... once its going backwards, the probabilities that it will go forward again is different.

So, in this example, the compiler works fine if the conditional is

(direction = -1 )

also works fine if

(direction ~= -1 )

but gives me the 'Hot variables cannot perform..'

with

(direction < 0 )

or ( direction <=0 ) or ( direction negative )

this is the part I really want to understand , before going further - why doesn't the compiler want to compile theses SmallTalk messages?

posted 09 April 2006 12:20            

eq: (for =)
lt: (for < )
le: (for <=)
ne: (for ~=)

etc...

posted 11 April 2006 06:26            

So, just to clarify...

When you want to evaluate code after the sound has started running, you must think in terms of Capytalk, the dynamic variable language.

In Capytalk, there aren't blocks like in Smalltalk, but expressions, right?

posted 11 April 2006 08:56            

!sw01 true: #(1 2 3) false: #(3 2 1)

or you could make each element of the array depend on some condition, eg.
#({!sw01 true: 1 false: 3} 2 {!sw01 true: 3 false: 1})