hi -

if i have something like
	kVal OPCODE iVal1, iVal2
the k-version of OPCODE is chosen.

but if i define UDO both at i and k, then in
	kVal UDO iVal1, iVal2
the i-version of UDO is chosen.

i think this is not correct, and should be in the same way as it works 
for OPCODE.

below is an example which should explain everything.

thanks for having a look -

	joachim









sr = 44100
ksmps = 32
nchnls = 2
0dbfs = 1

   opcode RndInt, i, ii
iMin, iMax xin
iRnd random iMin, iMax+.999999
xout int(iRnd)
   endop

   opcode RndInt, k, kk
kMin, kMax xin
kRnd random kMin, kMax+.999999
xout int(kRnd)
   endop

instr 1

  kBla init 10
  printk2 kBla
  ;the following line does not affect the
  ;init of kBla as it is only valid at perf time
  kBla random 1, 2
  turnoff

endin
schedule 1, 0, 1

instr 2 ;wrong in my opinion

  kBla init 10
  printk2 kBla
  ;the following line resets the initialization
  ;of kBla as the i-time version of the UDO RndInt
  ;is selected (rather than the k-version)
  kBla RndInt 1, 2
  turnoff

endin
schedule 2, 0, 1

instr 3

  kBla init 10
  printk2 kBla
  ;forcing to select the k-rate version of the
  ;UDO works again
  kBla = RndInt:k(1, 2)
  turnoff

endin
schedule 3, 0, 1








output:
  i1    10.00000
  i2     0.00000

hi dave -

if you look at instr 3 of my example, i think it is clear that a UDO can 
work only at k-rate (as any k-rate build-in opcode).

the point is that a k-rate output variable should force the UDO to 
choose the k-rate version (if there is one in the definition), and this 
is not the case, as far as i see.

best -

	joachim



On 16/08/18 13:49, Dave Seidel wrote:
> This touches on an issue that's been confounding me lately. In the
> Csound manual, in the "opcode" entry, it's stated:
>
>     These opcodes actually run only at i-time. Performance time copying
>     is done by the user opcode call. This means that skipping /xin/ or
>     /xout/ with /kgoto/ has no effect, while skipping with /igoto/
>     affects both init and performance time operation.
>
>
> However, in the FLOSS manual chapter on UDOs, in the second paragraph,
> it is stated: "They run at i-time or at k-time."
>
> Which one is correct? Is the manual out of date?
>
> The issue for me has been that I have a rather complex script
> (Implication Organ). The initial version uses MIDI controls (as in knobs
> & sliders), but I've been rewriting it to use OSC for those functions.
> In the process of rewriting, I had to shift a lot of things from running
> at i-time to running at k-time. When I did this, my UDOs stopped working
> as expected. It took a while before I realized that what was happening
> was that the UDO were only firing at i-time -- even for opcodes where
> all inputs and outputs were defined as k-variables. Is it possible that
> there's a bug? Unless the manual is correct and they only run at i-time.
>
> This seems related to what Joachim is asking above.
>
> I am working with the latest code in the develop branch.
>
> Thanks,