I am trying to implement a 3 way switch in csound, but can´t find any easy
solutions. A 2 way switch is relatively easy. I solved that this way:
instr 1

asig1		poscil 0.3,440,gisine
asig2		poscil 0.3,220,gisine
klfo1		lfo 1,2,3
klfo1		port klfo1, 0.01
klfo2		= 1-klfo1
asig1		=asig1*klfo1
asig2		=asig2*klfo2
atot		= asig1 + asig2
out		atot
endin 

But a 3 way (or more) switch is more complicated. I could use a 3 level cv,
and some if-statements, but it seems a bit laboursome. I have tried to find
a switch opcode, but  a search for "switch" returned 0.
I dream about an opcode that looks like this:
asig  NSWITCH ktrig, inumber, asig1, asig2, asig3 etc
where you can specify a trigger, the number of inputs(= inumber), and the
different input signals. It should also have an internal attack-release of
about 0.01 s when you change signal, to avoid clicks. 
Any ideas with existing opcodes, or perhaps someone good at programming
could have a go?



--
View this message in context: http://csound.1045644.n5.nabble.com/cv-switch-tp5727911.html
Sent from the Csound - General mailing list archive at Nabble.com.

Maybe you need an array of signals and some code to cross-fade

> I am trying to implement a 3 way switch in csound, but can´t find any
> easy
> solutions. A 2 way switch is relatively easy. I solved that this way:
> instr 1
>
> asig1		poscil 0.3,440,gisine
> asig2		poscil 0.3,220,gisine
> klfo1		lfo 1,2,3
> klfo1		port klfo1, 0.01
> klfo2		= 1-klfo1
> asig1		=asig1*klfo1
> asig2		=asig2*klfo2
> atot		= asig1 + asig2
> out		atot
> endin
>
> But a 3 way (or more) switch is more complicated. I could use a 3 level
> cv,
> and some if-statements, but it seems a bit laboursome. I have tried to
> find
> a switch opcode, but  a search for "switch" returned 0.
> I dream about an opcode that looks like this:
> asig  NSWITCH ktrig, inumber, asig1, asig2, asig3 etc
> where you can specify a trigger, the number of inputs(= inumber), and the
> different input signals. It should also have an internal attack-release of
> about 0.01 s when you change signal, to avoid clicks.
> Any ideas with existing opcodes, or perhaps someone good at programming
> could have a go?
>
>
>
> --
> View this message in context:
> http://csound.1045644.n5.nabble.com/cv-switch-tp5727911.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
> Send bugs reports to the Sourceforge bug trackers
> csound6:
>             https://sourceforge.net/p/csound/tickets/
> csound5:
>             https://sourceforge.net/p/csound/bugs/
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe
> csound"
>
>
>
>
>

yes, this is not that complicated. actually, if you did not need 
crossfade, the code would be simple:

instr 1
asig1 poscil .2, 770
asig2 poscil .2, 440
asig3 poscil .2, 550

kTrig randomh 1, 4, 1
printk2 int(kTrig)

if kTrig < 2 then
asig = asig1
elseif kTrig < 3 then
asig = asig2
else	
asig = asig3
endif

outs asig, asig
endin

crossfading means that you do more or less the same, but for the volume 
control of each "track". adding port (as you already did) is all you 
need then:

instr 2
;three audio signals
asig1 poscil .2, 770
asig2 poscil .2, 440
asig3 poscil .2, 550

;primitive random choice
kTrig randomh 1, 4, 1
printk2 int(kTrig)

;switch matrix
if kTrig < 2 then
ksig1 = 1
ksig2 = 0
ksig3 = 0
elseif kTrig < 3 then
ksig1 = 0
ksig2 = 1
ksig3 = 0
else	
ksig1 = 0
ksig2 = 0
ksig3 = 1
endif

;apply smooth transitions
ksig1 port ksig1, .1
ksig2 port ksig2, .1
ksig3 port ksig3, .1

;multiply with audio
asig1 = asig1*ksig1
asig2 = asig2*ksig2
asig3 = asig3*ksig3

;send the sum out
outs asig1+asig2+asig3, asig1+asig2+asig3
endin

i have attached a complete example. best -

	joachim



Am 27.09.2013 16:31, schrieb orebronerd:
> I am trying to implement a 3 way switch in csound, but can´t find any easy
> solutions. A 2 way switch is relatively easy. I solved that this way:
> instr 1
>
> asig1		poscil 0.3,440,gisine
> asig2		poscil 0.3,220,gisine
> klfo1		lfo 1,2,3
> klfo1		port klfo1, 0.01
> klfo2		= 1-klfo1
> asig1		=asig1*klfo1
> asig2		=asig2*klfo2
> atot		= asig1 + asig2
> out		atot
> endin
>
> But a 3 way (or more) switch is more complicated. I could use a 3 level cv,
> and some if-statements, but it seems a bit laboursome. I have tried to find
> a switch opcode, but  a search for "switch" returned 0.
> I dream about an opcode that looks like this:
> asig  NSWITCH ktrig, inumber, asig1, asig2, asig3 etc
> where you can specify a trigger, the number of inputs(= inumber), and the
> different input signals. It should also have an internal attack-release of
> about 0.01 s when you change signal, to avoid clicks.
> Any ideas with existing opcodes, or perhaps someone good at programming
> could have a go?
>
>
>
> --
> View this message in context: http://csound.1045644.n5.nabble.com/cv-switch-tp5727911.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
> Send bugs reports to the Sourceforge bug trackers
> csound6:
>              https://sourceforge.net/p/csound/tickets/
> csound5:
>              https://sourceforge.net/p/csound/bugs/
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

very clever ... -- it took some time until i understood how this works. 
would be nice (und much easier) to do this with the new arrays!
best -
	joachim


Am 27.09.2013 19:14, schrieb Iain McCurdy:
> Here's another option using a table with wrapping. The advantage of this
> method is that you could expand it to a 20 way switch without the code
> needing to be vastly more complicated.
>
> 
>
> 
> -odac
> 
>
> 
> sr = 44100
> ksmps = 64
> nchnls = 1
> 0dbfs = 1
>
> instr    1
> iOnOff    ftgen 0,0,4,-2,1
> kswitch = p4
> k1 table kswitch-1 ,iOnOff,0,0,1
> k2 table kswitch-2,iOnOff,0,0,1
> k3 table kswitch-3,iOnOff,0,0,1
> a1 vco2 0.3*k1,400,4,0.5 ; switch pos. 1
> a2 vco2 0.3*k2,500,4,0.5 ; switch pos. 2
> a3 vco2 0.3*k3,600,4,0.5 ; switch pos. 3
> out a1+a2+a3
> endin
>
> 
> 
> ; p4 = switch position 1,2 or 3
> i 1 0 0.5 1
> i 1 + 0.5 2
> i 1 + 0.5 3
> 
>
> 
>
>
>
>  > Date: Fri, 27 Sep 2013 07:31:48 -0700
>  > From: martino.flodino@gmail.com
>  > To: csound@lists.bath.ac.uk
>  > Subject: [Csnd] cv switch
>  >
>  > I am trying to implement a 3 way switch in csound, but can´t find any
> easy
>  > solutions. A 2 way switch is relatively easy. I solved that this way:
>  > instr 1
>  >
>  > asig1 poscil 0.3,440,gisine
>  > asig2 poscil 0.3,220,gisine
>  > klfo1 lfo 1,2,3
>  > klfo1 port klfo1, 0.01
>  > klfo2 = 1-klfo1
>  > asig1 =asig1*klfo1
>  > asig2 =asig2*klfo2
>  > atot = asig1 + asig2
>  > out atot
>  > endin
>  >
>  > But a 3 way (or more) switch is more complicated. I could use a 3
> level cv,
>  > and some if-statements, but it seems a bit laboursome. I have tried
> to find
>  > a switch opcode, but a search for "switch" returned 0.
>  > I dream about an opcode that looks like this:
>  > asig NSWITCH ktrig, inumber, asig1, asig2, asig3 etc
>  > where you can specify a trigger, the number of inputs(= inumber), and the
>  > different input signals. It should also have an internal
> attack-release of
>  > about 0.01 s when you change signal, to avoid clicks.
>  > Any ideas with existing opcodes, or perhaps someone good at programming
>  > could have a go?
>  >
>  >
>  >
>  > --
>  > View this message in context:
> http://csound.1045644.n5.nabble.com/cv-switch-tp5727911.html
>  > Sent from the Csound - General mailing list archive at Nabble.com.
>  >
>  >
>  > Send bugs reports to the Sourceforge bug trackers
>  > csound6:
>  > https://sourceforge.net/p/csound/tickets/
>  > csound5:
>  > https://sourceforge.net/p/csound/bugs/
>  > Discussions of bugs and features can be posted here
>  > To unsubscribe, send email sympa@lists.bath.ac.uk with body
> "unsubscribe csound"
>  >
>  >

Thank you for your suggestions. They all look like workable solutions. I will
try them out tomorrow, it´s getting late in Sweden. I do miss a switch
opcode though. Perhaps something for the next update?



--
View this message in context: http://csound.1045644.n5.nabble.com/cv-switch-tp5727911p5727916.html
Sent from the Csound - General mailing list archive at Nabble.com.

I had another go, (it wasn´t so late after all) working with table lookups
and if statements, but then I have the problem with the clicks when it jumps
in amplitude. 





sr = 44100
ksmps = 1
nchnls = 1
0dbfs  = 1

gisine	ftgen	0,0,2^12,10,1
gitrig	ftgen	0,0,-3,-2,1,2,3	

instr 1
kndx		phasor	1
ktrig		table kndx*3,gitrig
asig1		poscil 0.3,440,gisine
asig2		poscil 0.3,220,gisine
asig3		poscil 0.3,880,gisine
if 			ktrig == 1 then
asig1		= asig1
asig2		= asig2*0
asig3		= asig3*0
elseif		ktrig== 2 then
asig1		= asig1*0
asig2		= asig2
asig3		= asig3*0
elseif		ktrig== 3 then
asig1		= asig1*0
asig2		= asig2*0
asig3		= asig3
endif

atot		= asig1 + asig2 + asig3
out		atot*0.3
endin 



i 1 0 20







--
View this message in context: http://csound.1045644.n5.nabble.com/cv-switch-tp5727911p5727917.html
Sent from the Csound - General mailing list archive at Nabble.com.

Here's another possible solution. kswitch sig is any signal. It gets
truncated to just the integer part, then the amplitude for that index
is set to 1 or 0.  Port is used to smooth the transition from those
two states.

You could also do a recursive UDO version that could be extended to n-
oscillators at run time.

Cheers!
steven





sr=44100
ksmps=1
0dbfs=1


instr 1

kswitchsig linseg 0, p3 * .5, 3, p3 *.5, 0
kswitchsig = int(kswitchsig)

kamp1 = (kswitchsig == 0) ? 1 : 0
kamp1 port kamp1, 0.1
a1    vco2 kamp1, 440


kamp2 = (kswitchsig == 1) ? 1 : 0
kamp2 port kamp2, 0.1
a2    vco2 kamp2, 660

kamp3 = (kswitchsig == 2) ? 1 : 0
kamp3 port kamp3, 0.1
a3    vco2 kamp3, 880

aout sum a1, a2, a3
aout = aout * 0.4

outs aout, aout

endin




i1 0 4




On Fri, Sep 27, 2013 at 4:29 PM, orebronerd  wrote:
> I had another go, (it wasn´t so late after all) working with table lookups
> and if statements, but then I have the problem with the clicks when it jumps
> in amplitude.
>
> 
> 
> 
> 
> sr = 44100
> ksmps = 1
> nchnls = 1
> 0dbfs  = 1
>
> gisine  ftgen   0,0,2^12,10,1
> gitrig  ftgen   0,0,-3,-2,1,2,3
>
> instr 1
> kndx            phasor  1
> ktrig           table kndx*3,gitrig
> asig1           poscil 0.3,440,gisine
> asig2           poscil 0.3,220,gisine
> asig3           poscil 0.3,880,gisine
> if                      ktrig == 1 then
> asig1           = asig1
> asig2           = asig2*0
> asig3           = asig3*0
> elseif          ktrig== 2 then
> asig1           = asig1*0
> asig2           = asig2
> asig3           = asig3*0
> elseif          ktrig== 3 then
> asig1           = asig1*0
> asig2           = asig2*0
> asig3           = asig3
> endif
>
> atot            = asig1 + asig2 + asig3
> out             atot*0.3
> endin
>
> 
> 
> i 1 0 20
>
> 
> 
>
>
>
>
> --
> View this message in context: http://csound.1045644.n5.nabble.com/cv-switch-tp5727911p5727917.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
> Send bugs reports to the Sourceforge bug trackers
> csound6:
>             https://sourceforge.net/p/csound/tickets/
> csound5:
>             https://sourceforge.net/p/csound/bugs/
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>

yes, i have shown in my previous mail how you can prevent clicks (in the 
second example).

	joachim


Am 27.09.2013 22:29, schrieb orebronerd:
> I had another go, (it wasn´t so late after all) working with table lookups
> and if statements, but then I have the problem with the clicks when it jumps
> in amplitude.
>
> 
> 
> 
> 
> sr = 44100
> ksmps = 1
> nchnls = 1
> 0dbfs  = 1
>
> gisine	ftgen	0,0,2^12,10,1
> gitrig	ftgen	0,0,-3,-2,1,2,3	
>
> instr 1
> kndx		phasor	1
> ktrig		table kndx*3,gitrig
> asig1		poscil 0.3,440,gisine
> asig2		poscil 0.3,220,gisine
> asig3		poscil 0.3,880,gisine
> if 			ktrig == 1 then
> asig1		= asig1
> asig2		= asig2*0
> asig3		= asig3*0
> elseif		ktrig== 2 then
> asig1		= asig1*0
> asig2		= asig2
> asig3		= asig3*0
> elseif		ktrig== 3 then
> asig1		= asig1*0
> asig2		= asig2*0
> asig3		= asig3
> endif
>
> atot		= asig1 + asig2 + asig3
> out		atot*0.3
> endin
>
> 
> 
> i 1 0 20
>
> 
> 
>
>
>
>
> --
> View this message in context: http://csound.1045644.n5.nabble.com/cv-switch-tp5727911p5727917.html
> Sent from the Csound - General mailing list archive at Nabble.com.
>
>
> Send bugs reports to the Sourceforge bug trackers
> csound6:
>              https://sourceforge.net/p/csound/tickets/
> csound5:
>              https://sourceforge.net/p/csound/bugs/
> Discussions of bugs and features can be posted here
> To unsubscribe, send email sympa@lists.bath.ac.uk with body "unsubscribe csound"
>
>
>

Thank you once more for all of your ideas. Sorry, I missed the port solution,
I guess it was late after all. :-)



--
View this message in context: http://csound.1045644.n5.nabble.com/cv-switch-tp5727911p5727928.html
Sent from the Csound - General mailing list archive at Nabble.com.