Forums > MaxMSP

poly~

December 6, 2013 | 7:08 am

I’m building a drum sequencer. I had 5 voices for the drum sounds, now 10. Each is a sub patch with various function boxes for adsr, number boxes for frequency, etc. all being synthesized from cycle~ and noise~. My problem is that i’ve noticed the patch has become quite slow since adding an additional 5 voices (now 10 altogether). My cpu usage is only at around 10%.

My understanding of poly~ isn’t absolutely clear right now but would I be able create a poly~ object of these 10 sub patches in order to make this patch a little bit more responsive? Maybe I could create one set of function and number boxes and send those attributes to each voice of poly maybe? Or what would maybe be a better solution?



dtr
December 6, 2013 | 9:01 am

Post your patch. Could be a lot of things causing inefficiencies.


December 14, 2013 | 7:56 am

i think i figured it out, i moved all my pan objects into the sub patcher and temporarily got rid of all the meter~ objects and for some reason it seems to be running fine.

i just have another issue i’ve run into that i would appreciate some insight on..

It seems like i need to create some sort of connection between the sequencer’s tempo and the domain of each function box in order to avoid clicking because when i have the tempo fast enough and the domain short enough, i’m getting a POP at the beginning of the sound being trigger even though my adsr function boxes are set to fade in .

Also, is it a good rule of thumb to scale every amplitude .25 to avoid clipping in a situation like this? i’ve read something like that somewhere and its been in the back of my mind for some time now… thanks for any help in advance


December 14, 2013 | 8:10 am

regarding the popping, i guess what i’m asking is there a way to interrupt those waveforms without causing a pop or without adjusting the domain to be the same as the tempo of the sequencer?


December 14, 2013 | 9:34 am

Hi
you should really post an example of what you’re describing, a small excerpt from your main patch. As for amplitude scaling, divide total amplitude, 1., by the number of combined signals, then scale each audio signal by that value. For example, 8 voices should each be scaled by 0.125, 2 voices by 0.5.

<code>

– Pasted Max Patch, click to expand. –

</code>

Brendan


Viewing 5 posts - 1 through 5 (of 5 total)