what are the best parameters to use to compress spoken word for de-essing? thresh, ration, knee, attack, release... i could also eq it if necessary. i am planning to use the abstraction that Maelstorm created (by the way, thanks!).

i am running tech for a really low budget theater production that has a somewhat complicated set up. using pd is allowing me to avoid buying/renting a big mixer and running too many cables. its great!

thanks.

i made this mixer strip abstraction that allows for an argument to set the input. i am planning to make a patch with a number of these so that i can record many inputs at once. however, every time i make a patch with even two of these in it - and they try to delete one, pd crashes. even trying to move one around is a hassle. is the abstraction to complex to allow pd to run normally? i dont think that's the case, so what could it be?

http://www.pdpatchrepo.info/hurleur/mixerstrip.pd

Hello all,

I've been doing Pd for a little over a year now. I am about to move to Minneapolis and would love to find a community of Pd-ers or at least DSP-ers. Does anyone know of one? Or can someone tell me where I should be looking? Other website or forum?

Thanks,

Ted

just to be clear, the [int] object doesn't round to an integer, it chops off the fractional part of the number leaving you with the number of whole parts of the value

(i.e., it always rounds down)

is there a [round] object yet?

@lukasz said:

but if change value of y input and x stays at 2, it still matches and wont throw false output until both values change. seems that it's only when Y stays the same that it will continue to match, if x changes it sends a false or 0.

[bondo] works well for getting around that, i find myself using it a lot.

thanks for the help mod! i know this is a common topic, so i've attached the edited patch for anyone who wants to see it.

http://www.pdpatchrepo.info/hurleur/recorder~.pd

it turns out i need to recant my previous statement. my recorder~ seems to be working just fine... i've had issues in the past with having to save one empty throwaway file before i could be positive writesf~ would be reliable (i hadn't used this abstraction in a while because of that). if this comes up again, i'll try to troubleshoot it and report.

just for kicks i attached it, if someone spots something fishy, or if evilsheep wants to take a look at it, it could be helpful.

http://www.pdpatchrepo.info/hurleur/recorder~.pd

@evilsheep23 said:

I'm in Windows 7 also, and [writesf~] works for me, with one caveat: it never works the first time.

Any time I want to use it, I do the open routine (which you've already covered), and then send a "start" and "stop" message, then open the file again, and then it works for me.

i'm in osX and i have the same caveat. does anyone know why this is?

yup that fixed it. i'll just use a symbol atom instead. thanks for your help!

i also tried paring the abstraction down to make it less complex, but that didnt help.

http://www.pdpatchrepo.info/hurleur/mixerstrip2.pd

Could someone point me to (or provide me with) a good description of why [fft] results need to be squared, added, and square-rooted. I understand the concept of what [fft] is doing but that part of the math I don't get. Why doesn't the [fft] object do that automatically? Is there a reason why one would use [fft] without these calculations? Also, why does it need to be renormalized after the [rfft]? Why doesn't [rfft] do that automatically?

I've had fun playing around with filtering white noise with [fft] but I prefer to understand it!

Thank you!

try fiddle~

i am working on the same thing, but my dj hero turntable platter comes to rest at 0 rather than at 0.5 like in the tutorial. anybody have any idea what that's about?