Cycling 74 » Topic: Cant mute output quick enough when recording http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/feed Wed, 19 Jun 2013 06:54:19 +0000 http://bbpress.org/?v=2.2.4 http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-50958 <![CDATA[Cant mute output quick enough when recording]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-50958 Fri, 18 Jun 2010 16:18:04 +0000 gavspav My patch records into buffers using groove~ and poke~

I want to kill the output when recording to avoid feedback.

I’m having trouble – seem to get a tiny glitch where the recording starts before the output has been killed.

Is this cos my record switch is happening at control as opposed to signal level?

Tried writing a tiny patch to demonstrate this but it worked fine so didnt demonstrate my issue!

Using print messages I can see that the mute message occurs before the recording stars. But something I don’t understand is occurring.

Anyone know what I’m on about?

]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182762 <![CDATA[Re: Cant mute output quick enough when recording]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182762 Fri, 18 Jun 2010 16:27:02 +0000 Tim Lloyd It’s a little hard to tell without seeing the part of the patch that is causing the problem. Depending on how you’re doing the muting there could be a couple things causing it.

But a little feedback isn’t likely to be causing a glitch imo, that’s more likely to be due to the loop-point glitch that’s pretty inevitable when looping a recording. Try using the grooveduck abstraction instead of groove~ and see if the problem persists. If not, then you’ve done nothing wrong, it’s just an inevitable artifact of looping in that situation.

Is this feedback you’re trying to avoid acoustic (ie you’re recording through a mic and monitoring on speakers) or have you built it into the patch in some way for another reason?

Also, double-check how you’re using poke~ as you might be causing a 1 or 2 sample error somehow as you start recording.

]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182763 <![CDATA[Re: Cant mute output quick enough when recording]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182763 Fri, 18 Jun 2010 16:34:52 +0000 gavspav Yeah trying to avoid acoustic feedback.
Think the glitch is caused by recording starting slightly before the sound has stopped.

I can delay the recording and get around it but obviously this creates latency and I’ve had to delay it by up to 350ms

I’ll try to get a proper example together.

]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182764 <![CDATA[Re: Cant mute output quick enough when recording]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182764 Fri, 18 Jun 2010 16:45:19 +0000 gavspav Thanks for looking – let us know if there are any obvious errors:

– Pasted Max Patch, click to expand. –
]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182765 <![CDATA[Re: Cant mute output quick enough when recording]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182765 Fri, 18 Jun 2010 16:58:33 +0000 Tim Lloyd There’s two different kinds of glitches happening in that patch, the obvious one where recording starts suddenly, and a more “aliased” sound that I think is happening because of how you’re using poke~.

Give me a few minutes and I’ll show you a neater way of doing it, with no glitches.

]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182766 <![CDATA[Re: Cant mute output quick enough when recording]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182766 Fri, 18 Jun 2010 17:44:16 +0000 Tim Lloyd Sorry for the wait, I ended up making it possibly more complex than you’ll need. In this patch, there is a constant recording signal, and to enable recording it is just turned on and off into poke~.

To avoid the glitches that happen when you start and stop recording, the buffer playback is mixed with the audio input, and crossfaded in and out over 1 millisecond before the audio input to poke~. This means that when you hit record, poke~ immediately starts recording, then the audio input to poke~ is faded up (and the playback down) over 1 millisecond. Then when you stop recording, the audio input and playback are faded down again over 1 millisecond, and after their amplitudes are at 0, poke~ stops recording.

This completely smooths over any record clicks so that each recorded segment of the buffer is glitch-free. That takes care of one type of glitch, the other is fixed by sending the sample index of poke~ a -1 whenever no recording is wanted. Simply sending it a 0 will constantly be re-writing the first sample of the buffer, which might cause a glitch.

Hopefully that makes sense and the patch does what you need. I’ve used index~ for playback, so you won’t be able to do any vari-speed with it unless it’s modified.

– Pasted Max Patch, click to expand. –
]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182767 <![CDATA[Re: Cant mute output quick enough when recording]]> http://cycling74.com/forums/topic/cant-mute-output-quick-enough-when-recording/#post-182767 Fri, 18 Jun 2010 18:19:49 +0000 gavspav Thanks for this – I’ll have a look at through and try and bend it to my will!

It is part of a large and sprawling patch which allows you to scratch live audio with a pair of dj hero controllers.

I have tried index/play and groove for playback and finally settled on groove.

Being able to link the record head to the output of groove was a bonus.

Thanks again.

]]>