dev builds
themes manual
device status forums
mailing lists
IRC bugs
dev guide

Rockbox mail archive

Subject: Re: Dynamic Range Compressor revisited

Re: Dynamic Range Compressor revisited

From: Björn Stenberg <>
Date: Tue, 1 Oct 2013 09:11:32 +0200


Your attached files were too big for the mail list handler to accept. I uploaded them here instead:

Here is the text from Ryan's mail:

Here are some demo files. The artist is Iona, from the album
"Heaven's Bright Sun". It is a live recording.

The first set at ratio 4:1 and -9dB Threshold are more realistic
settings (less extreme). On these you can observe how my
modifications to the compressor preserve more of the crispness in the
dynamic onsets. The overall feel is less "dark", while the other
sample is a bit more muddy. Hearing them side by side, I can see how
the average person may find the difference rather subtle.

The second set use the "lmt" ratio setting, and -24dB Threshold show
the difference between the more extreme settings. You can see that
even though it sounds bad, the setting showcases the increased
perception in the amount of dynamics by using an attack setting (all
samples 15 ms attack time).

With attack time set to 0 (no sample recorded with this setting), the
"new" compressor sounds much more like the "old" compressor, except
with the look-ahead action, the onset of compression does not come
with as much correlated distortion. The less rapid gain change is the
advantage here.

Of course, I would prefer if some people try some listen tests on
their own devices. If there are problems with the bit format kludge,
I will be happy to fix that if it's necessary for somebody to do a
fair evaluation.

Thanks to anybody who gives this consideration.
Attached are the sound samples. Look to the file names to get an idea
of the settings used on each sample.

Best Regards,

Received on 2013-10-01

Page template was last modified "Tue Sep 7 00:00:02 2021" The Rockbox Crew -- Privacy Policy