William Bowen

I'm experiencing some strange behavior from adsr~. When trigered by a pulse wave, everything's fine. However, if triggered by a cycle~d buffer~, only the first non-zero value in the buffer~ will trigger the adsr~. Below is a simplified patch which illustrates the issue. I've tried this on both intel and ppc installations of 4.6, both seem to exibhit the same behavior. Am I overlooking something really simple?

I'm experiencing some strange behavior from adsr~.  When trigered by a pulse wave, everything's fine.  However, if triggered by a cycle~d buffer~, only the first non-zero value in the buffer~ will trigger the adsr~.  Below is a simplified patch which illustrates the issue.  I've tried this on both intel and ppc installations of 4.6, both seem to exibhit the same behavior.  Am I overlooking something really simple?

adsr-signal-triggering-wierdness

> Am I overlooking something really simple?

The content of your buffer is never going to zero isn't it? Only zero 

William Bowen wrote:
> Am I overlooking something really simple?

The content of your buffer is never going to zero isn't it? Only zero 
will switch off...

-- 
Stefan Tiedje------------x-------
--_____-----------|--------------
--(_|_ ----|-----|-----()-------
-- _|_)----|-----()--------------
----------()--------www.ccmix.com


I just noticed that the preset didn't replicate the patch's settings just as it should have. change some values in the multislider and it will redraw with zeros in between the 'hits'.

I also tried making the spaces in between the 'hits' go to -1. and then using a [clip~ 0. 1.] on playback... still didn't work.

Could this be an artifact of the interpolation of the 512 sample lookup table?

What's more, it will trigger the first value of the buffer~ (if it's non-zero), but none after that... it's like the lookup index must go to zero, not the value triggering the adsr~.

I just noticed that the preset didn't replicate the patch's settings just as it should have.  change some values in the multislider and it will redraw with zeros in between the 'hits'.

> What's more, it will trigger the first value of the buffer~ (if it's

> non-zero), but none after that... it's like the lookup index must go

> to zero, not the value triggering the adsr~.

Yes, obviously, as explained in the help file. (hint: always look for 

William Bowen wrote:
> What's more, it will trigger the first value of the buffer~ (if it's
> non-zero), but none after that... it's like the lookup index must go
> to zero, not the value triggering the adsr~.
>
> does this make sense?

Yes, obviously, as explained in the help file. (hint: always look for 
subpatchers in help files...)

Which subpatcher in which .help file obviously describes that only the peek~d value of indexes 0-31 will be presented on output and, once that output returns to zero (for re-triggering of the adsr~, after having been high for 32 samples), that the remaining 480 samples will not rise more than 1/100 of the value peek~d at its index in the buffer~? (hint: it's not peek~, buffer~, cycle~, or adsr~).

I'm referring to _index_ as the address in the buffer~ (0 - 511), and _value_ as the floating point number (0.0 - 0.999999) located at that index.

hook a [snapshot~ 10] to a print before and after the adsr~ if this isn't clear in the scope~. To make this even more clear, make the multislider all full values which will make the cycle~ output identical to the phase-locked pulse wave selectable from the umenu.

Which subpatcher in which .help file obviously describes that only the peek~d value of indexes 0-31 will be presented on output and, once that output returns to zero (for re-triggering of the adsr~, after having been high for 32 samples), that the remaining 480 samples will not rise more than 1/100 of the value peek~d at its index in the buffer~?  (hint: it's not peek~, buffer~, cycle~, or adsr~).

I'm referring to _index_ as the address in the buffer~ (0 - 511), and _value_ as the floating point number (0.0 - 0.999999) located at that index.  

hook a [snapshot~ 10] to a print before and after the adsr~ if this isn't clear in the scope~.  To make this even more clear, make the multislider all full values which will make the cycle~ output identical to the phase-locked pulse wave selectable from the umenu.

adsr~ signal triggering wierdness