Darren Gibbs

After experimenting with several off-the-shelf phrase loopers that 

never do exactly what I want, I decided to just build one that does. 

I thought it would be fairly trivial but have run into problems. I 

started with tapin~/tapout~ (out wrapped back into in with gain 

control for regeneration) which works fine for a very limited set of 

features. But I also want to be able to do things that groove~ can 

do, like pitch shift, play in reverse, change the delay time on the 

fly etc. I setup a buffer and a record~/groove~ pair to implement 

the delay with regeneration, and then I have a separate groove object 

pointed at the same buffer for the audio output which I can run at 

various speeds and reverse without tweaking out the regen cycle. 

This works conceptually, but within a few seconds the loop in the 

buffer becomes filled with clicks. This didn't happen with tapin~/ 

tapout~. Have I made a conceptual error? What I want I suppose is 

something that does what tapin~/tapout~ does in terms of smooth loop 

recording, but with groove~style access to the buffer.

After experimenting with several off-the-shelf phrase loopers that  
never do exactly what I want, I decided to just build one that does.   
I thought it would be fairly trivial but have run into problems.  I  
started with tapin~/tapout~ (out wrapped back into in with gain  
control for regeneration) which works fine for a very limited set of  
features.  But I also want to be able to do things that groove~ can  
do, like pitch shift, play in reverse, change the delay time on the  
fly etc.  I setup a buffer and a record~/groove~ pair to implement  
the delay with regeneration, and then I have a separate groove object  
pointed at the same buffer for the audio output which I can run at  
various speeds and reverse without tweaking out the regen cycle.   
This works conceptually, but within a few seconds the loop in the  
buffer becomes filled with clicks.  This didn't happen with tapin~/ 
tapout~.   Have I made a conceptual error?  What I want I suppose is  
something that does what tapin~/tapout~ does in terms of smooth loop  
recording, but with groove~style access to the buffer.

delay-concepts

This might have something to do with the fact, that the record~ object dosen't stream audio, but records in a loop..... This will generate clicks when the record and playback point are the same.....

I think the stutter~ object does exatly what u want... it keeps a history of the incomming audio stream, and is able to copy snippets of it to a playback buffer~ .... the buffer can then be phased thru...

On 1 mars 06, at 08:33, Darren Gibbs wrote:

> I setup a buffer and a record~/groove~ pair to implement the delay 

> with regeneration, and then I have a separate groove object pointed at 

> the same buffer for the audio output which I can run at various speeds 

> and reverse without tweaking out the regen cycle. This works 

> conceptually, but within a few seconds the loop in the buffer becomes 

You can't record and play a same buffer at the same time, i.e position 

in the buffer. If the read and write position are crossing, you may get 

a click recorded in the buffer. You could use 2 buffers instaead and 

swap, or do some logic to avoid these conflicts (which can become quite 

tricky if the read speed is changing during time).

Beside of that, when you record new material in a buffer, you erase the 

previous content, so you'll also get a clik. There are a couple of 

objects that may be usefull for you: xrecord~ and co, by Thomas Grill. 

Centre de Recherches et de Formation Musicales de Wallonie asbl

> I setup a buffer and a record~/groove~ pair to implement the delay 
> with regeneration, and then I have a separate groove object pointed at 
> the same buffer for the audio output which I can run at various speeds 
> and reverse without tweaking out the regen cycle.  This works 
> conceptually, but within a few seconds the loop in the buffer becomes 
> filled with clicks.

You can't record and play a same buffer at the same time, i.e position 
in the buffer. If the read and write position are crossing, you may get 
a click recorded in the buffer. You could use 2 buffers instaead and 
swap, or do some logic to avoid these conflicts (which can become quite 
tricky if the read speed is changing during time).

Beside of that, when you record new material in a buffer, you erase the 
previous content, so you'll also get a clik. There are a couple of 
objects that may be usefull for you: xrecord~ and co, by Thomas Grill. 
(see maxobjects.com for a pointer)

p
_____________________________
Patrick Delges

Centre de Recherches et de Formation Musicales de Wallonie asbl
http://users.skynet.be/crfmw/max


> This works conceptually, but within a few seconds the loop in the

> buffer becomes filled with clicks. This didn't happen with tapin~/

> tapout~. Have I made a conceptual error?

Probably, without seeing a stripped down example which exhibits the 

By the way stripping down such a thing usually points you directly to 

the problem and you might find out yourself.

If you find the culprit do not hesitate to post the solution as well 

(others might have similar problems and will be happy to go the same route)

Darren Gibbs wrote:
> This works conceptually, but within a few seconds the loop in the
> buffer becomes filled with clicks.  This didn't happen with tapin~/
> tapout~. Have I made a conceptual error?

Probably, without seeing a stripped down example which exhibits the 
behaviour its hard to tell.

By the way stripping down such a thing usually points you directly to 
the problem and you might find out yourself.
If you find the culprit do not hesitate to post the solution as well 
(others might have similar problems and will be happy to go the same route)

  [][]  [][][]  [][]  [][][]
[][][][][][][][][][][][][][][]

         Stefan Tiedje
         Klanggestalter
     Electronic Composition
               &
         Improvisation

            /~~~~~
     \   /|() ()|
     ))))) )|  |  |( \
     ///     _/)/ )))))
             ___/   ///

-------------------------x----
--_____-----------|-----------
--(_|_ ----|-----|-----()----
-- _|_)----|-----()-----------
----------()------------x-----

14, Av. Pr. Franklin Roosevelt,
94320 Thiais, France
Phone at CCMIX +33-1-49 77 51 72


delay concepts