ReaComp release curve option

[JamesPeters]

Please excuse me if this request has already been made. I searched and couldn't find it.

In one of Dan Worrall's latest videos about ReaComp, it seems he was struggling to get the sounds he wanted from it (even after the "weird knee" fix was done):

https://forum.cockos.com/showthread....77#post2121577

It was after watching that video that I realized: the fact I can get ReaComp to work so well for me is probably because I have been dialing in ReaComp more "by ear" (and "in a vacuum") and less by what someone else would do based on their use of other compressors (including in various tutorials about compressors). For me to get the kind of response I want, I usually end up shooting for 6dB of gain reduction (or more, then compensating by raising the wet output higher and blending some dry in), plus I have to set the attack a bit differently (faster usually, then I adjust the knee and release sometimes).

It turns out the release curve of ReaComp is anti-log, which is unusual compared to most compressors. This means to get ReaComp to sound the way you might expect based on other compressors (closer anyway), you have to deal with that response curve "throwing things off" and use different settings.

Here's the first thread I found with this being mentioned (it has image attachments so you can see the release shape compared to other compressors):

https://forum.cockos.com/showthread.php?t=32583

Later it was brought up here (from this post onward):

https://forum.cockos.com/showthread....72#post2087572


I humbly request that a second response curve be added to ReaComp, if feasible, such that it can behave more as people might generally expect from using other compressors.

Thanks for reading.

[ashcat_lt]

I don’t really do a lot of compression that involves either attack or release, but I think this is a good idea. People (with better ears than mine) can hear that and even if they can’t tell you why, they know they don’t like it or it’s not what they expect. It can be cool sometimes to be unique, but I’m a big ReaComp fan, and think it should be everybody’s go to compressor. This makes it a little tough to recommend for some things.

[Ivannn Bennnettt]

+1 and for attack as well

[JamesPeters]

I figure I'll give this a one-year "bump" just in case Justin remembers this post from 2009 about it:

Originally Posted by Justin

There may eventually be an option to change it...

Yes I'm a jerk.

[ReaNico]

Originally Posted by JamesPeters

I figure I'll give this a one-year "bump" just in case Justin remembers this post from 2009 about it:


Yes I'm a jerk.

11 years and nothing?

[JamesPeters]

Honestly I get great results with ReaComp anyway. It'd just be easier to dial in for some sounds if this feature were added. It took me a while to work around this aspect, but it's workable once you "get it". I can't quite dial it in like a lot of other compressors.

The previous fix of the "weird knee" was the main thing holding me back from using ReaComp; it was way too finnicky before for most of what I was trying to do. (Although later I realized that "weird knee" is quite useful for keeping transient attack in a somewhat different/"automatic" way instead of using later attack times, and I still use "weird knee" sometimes today.)

[mawi]

+4 I am also in favor of an option to switch from a linear to an antilogarithmic release progression, like the mpressor from elysia has. The linear progression is good for noticeable effect compression, e.g. on drums. For a more inconspicuous compression the antilogarithmic progression is simply better suited.

[mawi]

I just read that Oxford Dynamics also has a Linear and Exponential Timing option, which changes the release and attack behavior. Such an option I would also find interesting for ReaComp.

[ashcat_lt]

IDK for sure, but I suspect that what we see is actually linear interpolation on the dB scale, which ends up looking exponential on the voltage scale because that’s how dB works. Analog compressors don’t actually work that way, nor would a compressor that tried at all to act like one.

[mawi]

add an anti-log release curve and a better auto release, please.

[enroe]

Originally Posted by JamesPeters

I humbly request that a second response curve be added to ReaComp, if feasible, ...

Yes, just a logarithmic release-curve would improve ReaComp
(again) considerably (if doable without too much CPU-cost)!

[Bouroki]

They won't do this simple fix unless Dan Worrall complains about it...

[enroe]

Originally Posted by Bouroki

They won't do this simple fix unless Dan Worrall complains about it...

Haha hey, he did already complain about it:
https://www.youtube.com/watch?v=7Yit769SN64

Look at 17:30 min and the following parts ...

[Bouroki]

Originally Posted by enroe

Haha hey, he did already complain about it:
https://www.youtube.com/watch?v=7Yit769SN64

Look at 17:30 min and the following parts ...

Ha, thanks! I need to revisit the whole video as he packs so much information into the 20-30 minutes.

If that didn't get them to fix it then we are doomed