Hi, how can i calculate zero crossing in pd? is there any librery or something?
thanks
R.
On Sun, 2011-01-16 at 06:15 -0800, ronni montoya wrote:
Hi, how can i calculate zero crossing in pd? is there any librery or something?
http://www.google.com/#q=site%3Alists.puredata.info+%22zero+crossing%22
Roman
2011/1/16 Roman Haefeli reduzent@gmail.com
On Sun, 2011-01-16 at 06:15 -0800, ronni montoya wrote:
Hi, how can i calculate zero crossing in pd? is there any librery or
something?
http://www.google.com/#q=site%3Alists.puredata.info+%22zero+crossing%22
This link feels a bit like "reed the FAQ" (I was also asking the samehttp://lists.puredata.info/pipermail/pd-list/2010-12/085571.htmlin principle some weeks ago) ok, but: There is no answer.
AFAIK it's impossible to get sample correct bangs from (anaylysed) audio.
On Sun, 2011-01-16 at 22:01 +0100, Dietrich Pank wrote:
2011/1/16 Roman Haefeli reduzent@gmail.com On Sun, 2011-01-16 at 06:15 -0800, ronni montoya wrote: > Hi, how can i calculate zero crossing in pd? is there any librery or something?
http://www.google.com/#q=site%3Alists.puredata.info+%22zero +crossing%22This link feels a bit like "reed the FAQ" (I was also asking the same in principle some weeks ago) ok, but: There is no answer.
AFAIK it's impossible to get sample correct bangs from (anaylysed) audio.
I didn't mean to say 'RTFM' or similar, but it's an ever arising topic and probably the answer is already covered.
I do think it is possible to get sample correct bangs from analyzed audio: http://lists.puredata.info/pipermail/pd-list/2010-12/085681.html
Roman
Thank you Roman,
I do think it is possible to get sample correct bangs from analyzed
audio: http://lists.puredata.info/pipermail/pd-list/2010-12/085681.html
Ok! After reading this tip again I understood it something better. But..well, one block later may be accourate theoretically but not real(time). Anyway this may cover some needs e.g. writing tables for scoping audio...
regards Dietrich
But..well, one block later may be accourate theoretically but not real(time). Anyway this may cover some needs e.g. writing tables for scoping audio...
That's why I created [zeroxpos~] - so that it gives you the position of the 1st, 2nd, 3rd etc zero crossing within the block. Then you can use a [bang~] connected to a counter like [cup] within the block object to calculate the time from when the counter was last initialized, and work out when the zero-crossing happened.
It only works when audio is buffered into a table though - it's not true real-time but one block away.
Ed
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 2011-01-17 02:01, Dietrich Pank wrote:
But..well, one block later may be accourate theoretically but not real(time).
you should re-think your definition of "realtime".
fmaer IOhannes
On Mon, 2011-01-17 at 02:01 +0100, Dietrich Pank wrote:
Thank you Roman,
I do think it is possible to get sample correct bangs from analyzed audio: http://lists.puredata.info/pipermail/pd-list/2010-12/085681.htmlOk! After reading this tip again I understood it something better. But..well, one block later may be accourate theoretically but not real(time).
True. But this is a logical problem that also an external wouldn't be able to solve. The results of some analysis of data cannot be delivered before the data was delivered.
With a block size of 1 this latency could be improved, but it would be still one sample late, so still not real-time in that sense.
Roman
Seems to me like nothing is ever "realtime" in DSP anyway...
Pierre
2011/1/17 Roman Haefeli reduzent@gmail.com
On Mon, 2011-01-17 at 02:01 +0100, Dietrich Pank wrote:
Thank you Roman,
I do think it is possible to get sample correct bangs from analyzed audio: http://lists.puredata.info/pipermail/pd-list/2010-12/085681.htmlOk! After reading this tip again I understood it something better. But..well, one block later may be accourate theoretically but not real(time).
True. But this is a logical problem that also an external wouldn't be able to solve. The results of some analysis of data cannot be delivered before the data was delivered.
With a block size of 1 this latency could be improved, but it would be still one sample late, so still not real-time in that sense.
Roman
Pd-list@iem.at mailing list UNSUBSCRIBE and account-management -> http://lists.puredata.info/listinfo/pd-list
Thank you for explanation, guys.
I'm still stuck in thinking like Reaktor (esp. "A to E" trigger) and have to learn to go other ways. The block system makes me crazy. I'm still don't have any idea how to make e.g. granular-like playbacks (fast retriggered granulats) syncronized to any other audio. Of course: "realtime" is the wrong word I used.
regards Dietrich
On Mon, Jan 17, 2011 at 10:20:51AM +0000, Jamie Bullock wrote:
On 16 Jan 2011, at 14:15, ronni montoya wrote:
Hi, how can i calculate zero crossing in pd? is there any librery or something?
Do you want the zero-crossing locations, or the zero-crossing rate?
zero crossing rate is attached as an abstraction - it needs some kind of [sgn~] object, which you can get from library like zexy or purepd or make on your own.
Ciao
Frank Barknecht Do You RjDj.me? _ ______footils.org__
-
Dietrich Pank -
Ed Kelly -
Frank Barknecht -
IOhannes m zmoelnig -
Jamie Bullock -
Pierre Massat -
Roman Haefeli -
ronni montoya