Help understanding how max handles audio input signals
I've run into an audio input issue that makes no sense to me:
I've got a little pitch-detection patch using the Sigmund~ object running on a MacPro tower with an RME fireface attached. If I plug a mic into the RME and start playing some notes Sigmund~ tracks them quite accurately. However, it's so sensitive it'll track me playing 25 feet from the mic (an AudioTechnica clip-on with pretty good off-axis rejection.) So I did the obvious and turned down the microphone input gain on the RME... no effect. In fact even when the gain on the interface is set to ZERO sigmund~ continues to track pitches! So I tried putting a gain slider between the ADC object and sigmund~. Even when the signal was attenuated by -65db sigmund~ continued tracking pitches with the same sensitivity as it had when the mic and max-input gains were turned up. I even tried positioning myself all the way across the room with the mic under a towel. Even when I played very softly sigmund~ STILL tracked.
It's almost as if it's getting the signal from somewhere other than the audio interface... but the MacPro I"m using doesn't have an internal mic and nothing else is hooked up to the RME. The audio input under the Max options menu is set to the RME. There are no other audio apps running.
Is this some kind of weird bug, or am I not understanding how the ADC object handles an audio input signal?
max/msp uses floating point numbers which as far as i can tell any reduction in gain just shoves the data further to the right of the decimal place without degrading the original data.
I could be wrong. just a guess.
silence will be reached when you multiply by zero. -65 db/A is quite low but its not gone.
the dynamic range of your audio interface wll be a bit less than computer generated signals, some 22 bits or about -130 db/A is common.
you can implement a simple gating method to ignore signals below a chosen threshold:
there is also a @minpower attribute built into [sigmund~] but I can't make it work properly...
Oy, I feel like such a dweeb! When sigmund's @minpower didn't respond to changes I stupidly assumed I'd get the same results if I gated the amplitude outside of Sigmund~, which of course was not the case. Thanks for nudging me back to consciousness!