Hi,
I can't seem to understand what the second argument of [env~] is for. The first argument is the window lenght for RMS analysis, but the second argument is supposed to set "period in samples per analysis (window/2 default)". It seems I am getting two analysis values per window size with the defaults of [env~ 1024 512] How would that be different from [env~ 512 512] ?
Thanks for any help here! P
On 2/7/20 11:04 AM, Peter P. wrote:
Hi,
I can't seem to understand what the second argument of [env~] is for. The first argument is the window lenght for RMS analysis, but the second argument is supposed to set "period in samples per analysis (window/2 default)". It seems I am getting two analysis values per window size with the defaults of [env~ 1024 512]
this means that you will get a new value every 512 samples (aka 8 blocks, or 11.6ms@44.1kHz). the value you get will take the last 1024 samples into account to calculate the RMS.
How would that be different from [env~ 512 512]
this means that you will get a new value every 512 samples (aka 8 blocks, or 11.6ms@44.1kHz). the value you get will take the last 512 samples into account to calculate the RMS.
so you get the same number of values, but they use different time-frames for doing the averaging.
gdsamr IOhannes
- IOhannes m zmölnig zmoelnig@iem.at [2020-02-07 12:38]:
On 2/7/20 11:04 AM, Peter P. wrote:
Hi,
I can't seem to understand what the second argument of [env~] is for. The first argument is the window lenght for RMS analysis, but the second argument is supposed to set "period in samples per analysis (window/2 default)". It seems I am getting two analysis values per window size with the defaults of [env~ 1024 512]
this means that you will get a new value every 512 samples (aka 8 blocks, or 11.6ms@44.1kHz). the value you get will take the last 1024 samples into account to calculate the RMS.
How would that be different from [env~ 512 512]
this means that you will get a new value every 512 samples (aka 8 blocks, or 11.6ms@44.1kHz). the value you get will take the last 512 samples into account to calculate the RMS.
so you get the same number of values, but they use different time-frames for doing the averaging.
Wow, crazy! Thanks for this explanation! Would be interesting to know for what different applications this (eg a long window with short period) can be useful. Anyway, it's very good to know.
cheerz, P
so you get the same number of values, but they use different time-frames for doing the averaging.
Wow, crazy! Thanks for this explanation! Would be interesting to know for what different applications this (eg a long window with short period) can be useful. Anyway, it's very good to know.
For continuous measurements, you can also calculate the RMS in the audio domain by squaring a signal, low-passing it with a 1-pole low-pass, and taking the square root. The period of the filter, roughly, is the analysis window. See Zölzer's "DASP" from 2008, page 229, for more information.
D
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
-
Dario Sanfilippo -
IOhannes m zmölnig -
Peter P.