I'm recording some one shot drum sounds for sampler fodder and would like to know the most appropriate rms integration times for percussion. Let's say I have a bass drum that sounds for around 500 ms, do I want the integration time to be below this?
I ask because I've gotten conflicting measurements with the bx_meter and the rms meter in ozone. Ozone allows to change the integration time....
RMS integration time for one shot drum sound
Very confused...-
- KVRAF
- 12615 posts since 7 Dec, 2004
There are numerous ways to window the RMS including rectangular, none of them are the correct way because no such thing exists.
You might find methods in various industries which have been agreed upon and are in common use and in very specific cases you may not.
As far as I'm aware broadcast levels metering methods were only very recently standardized in North America! (Or have they not yet been agreed upon? https://en.wikipedia.org/wiki/Loudness_monitoring https://en.wikipedia.org/wiki/LKFS https://en.wikipedia.org/wiki/Loudness )
Someone with some level of expertise in sampling and knowledge of whether standards exist for windowing the RMS measurement of a one-shot sample containing an exponential decay such as that of a drum might be able to give you some hints, but you won't find a single solid "this is the way it's done" answer.
You might find methods in various industries which have been agreed upon and are in common use and in very specific cases you may not.
As far as I'm aware broadcast levels metering methods were only very recently standardized in North America! (Or have they not yet been agreed upon? https://en.wikipedia.org/wiki/Loudness_monitoring https://en.wikipedia.org/wiki/LKFS https://en.wikipedia.org/wiki/Loudness )
Someone with some level of expertise in sampling and knowledge of whether standards exist for windowing the RMS measurement of a one-shot sample containing an exponential decay such as that of a drum might be able to give you some hints, but you won't find a single solid "this is the way it's done" answer.
Free plug-ins for Windows, MacOS and Linux. Xhip Synthesizer v8.0 and Xhip Effects Bundle v6.7.
The coder's credo: We believe our work is neither clever nor difficult; it is done because we thought it would be easy.
Work less; get more done.
The coder's credo: We believe our work is neither clever nor difficult; it is done because we thought it would be easy.
Work less; get more done.
-
- KVRAF
- 7579 posts since 17 Feb, 2005
Well, consider there's no long term RMS to look at. And also the usage of the newly normalized sample, it may or may not need to be weighted.aciddose wrote:There are numerous ways to window the RMS including rectangular, none of them are the correct way because no such thing exists.
You might find methods in various industries which have been agreed upon and are in common use and in very specific cases you may not.
As far as I'm aware broadcast levels metering methods were only very recently standardized in North America! (Or have they not yet been agreed upon? https://en.wikipedia.org/wiki/Loudness_monitoring https://en.wikipedia.org/wiki/LKFS https://en.wikipedia.org/wiki/Loudness )
Someone with some level of expertise in sampling and knowledge of whether standards exist for windowing the RMS measurement of a one-shot sample containing an exponential decay such as that of a drum might be able to give you some hints, but you won't find a single solid "this is the way it's done" answer.
-
- KVRAF
- 12615 posts since 7 Dec, 2004
If you didn't need to consider the "long term" which is in this case the exponential decay the best measurement would be a rectangular RMS.
http://mathworld.wolfram.com/Root-Mean-Square.html
The weighting applied in a running-sum RMS meter is actually the inverse. It weights the most recent input and older inputs decay exponentially.
In addition to that the RMS value is passed through an envelope which userb420 refers to as "integration time". (Extremely varied implementations. This may be any sort of window.)
These sort of metering systems are actually the opposite required to properly measure "power".
In addition you have frequency weighting which has nothing to do with time ("long term" vs. "short term") at all and in this case frequency weighting may be absolutely the most important factor if the goal is to measure "loudness" rather than "power".
Typically peak levels are used (normalization) and RMS meters don't enter into the equation until the sounds are in place in a mix.
http://mathworld.wolfram.com/Root-Mean-Square.html
The weighting applied in a running-sum RMS meter is actually the inverse. It weights the most recent input and older inputs decay exponentially.
In addition to that the RMS value is passed through an envelope which userb420 refers to as "integration time". (Extremely varied implementations. This may be any sort of window.)
These sort of metering systems are actually the opposite required to properly measure "power".
In addition you have frequency weighting which has nothing to do with time ("long term" vs. "short term") at all and in this case frequency weighting may be absolutely the most important factor if the goal is to measure "loudness" rather than "power".
Typically peak levels are used (normalization) and RMS meters don't enter into the equation until the sounds are in place in a mix.
Free plug-ins for Windows, MacOS and Linux. Xhip Synthesizer v8.0 and Xhip Effects Bundle v6.7.
The coder's credo: We believe our work is neither clever nor difficult; it is done because we thought it would be easy.
Work less; get more done.
The coder's credo: We believe our work is neither clever nor difficult; it is done because we thought it would be easy.
Work less; get more done.
-
- KVRer
- Topic Starter
- 25 posts since 14 Nov, 2012
Yeah the only reason I investigated recording at a certain RMS level was if I relied purely on peak normalization with in the sampler, the sounds with less attack sound louder and I wanted a more uniform loudness within the library in order to mitigate the chance of choosing the louder samples. I elected to record at -18 rms with a 300ms(VU) which corresponded to -18 rms on ableton's meters as it makes things more practical and it seems to be working so long as I compare apples to apples and not oranges or bass drums to bass drums and not cymbals.
Are there any plugins with frequency weighted metering that I should give a try? Don't think I'll be needing them for this exact purpose but I am interested none the less.
Thanks everyone for your contribution to the discussion!
Are there any plugins with frequency weighted metering that I should give a try? Don't think I'll be needing them for this exact purpose but I am interested none the less.
Thanks everyone for your contribution to the discussion!
-
- KVRAF
- 2657 posts since 13 Mar, 2004
EBU R128 which is based on ITU BS.1770 employs K-weighting afaik.userb420 wrote:
Are there any plugins with frequency weighted metering that I should give a try? Don't think I'll be needing them for this exact purpose but I am interested none the less.
So e.g. this one by Audiocation
https://www.audiocation.de/en/plugin#fragment-4
But then again, R128 is intended for program material using a fixed RMS time, not sure how good it would suit for drum one shots.