Compressor needs better parameter options

I believe that the built-in lookahead “Compressor” effect needs better parameter options.

Most of my points have already been made at the following post:
https://forum.audacityteam.org/t/compressor-in-v2-1-1-throws-error-when-run-as-chain-script/39853/1
(Don’t worry: Problem is described concisely. Also contains a link to proposals organized into a few spreadsheet tables.)

I was asking for help, but it appears that there is no help for v2.1.1. It seems that I would have to create my own help by modifying and recompiling Audacity.

~ jimmy

Audacity comes with another compressor called “SC4” which has plenty of parameters to tweak …
http://manual.audacityteam.org/o/man/sc4.html

But it has no look ahead which is a big minus.

Robert

Yeah, I also really like the SC4 (have used it enough), but it serves other purposes. For example, today I mastered a song and used SC4 to achieve something because I did NOT want a lookahead compressor for the particular manipulation that I needed. For general purpose Spoken Word situations, I really like the regular Compressor, as long as the necessary parameters are available. Without them, it is useless.

Of all things I have ever used in Audacity, I think the standard Compressor is the central feature. This effect needs to absolutely kick butt. If it did kick butt, then I think it would turn a lot of heads and make life easier for a lot of people.

Right now the Compressor is very disappointing. From 2.0.6 to 2.1.1, it has gone backward. It needs its mojo back. Please, please, please give its mojo back in the next release of Audacity.

Please explain.
The only functional change that I’m aware of since August 2009 is that the Attack/Release time now allows smaller values (the maximum value is unchanged).

Hi Steve!
Thank you for replying. Please see my first post and its link for the explanation. I think it would be redundant for me to try to explain everything all over again.

Have done.
I don’t see any problem in providing lower compression ratios in future versions. In fact we could probably have that in Audacity 2.1.3 :open_mouth:
(feel free to remind me after Audacity 2.1.2 is released)

Rather than providing extremely small attack times, my preference would be to include a separate “limiter” setting that is capable of very fast attack. The advantage being that there could then be an option for using the compressor with or without lookahead (or variable lookahead).

(I edited this reply 60 minutes after my original posting of this reply.)
(Edit: I’m not trying to be an arse here. I’m just trying to make points regarding my perception of how things work. Please, if you have examples that show evidence contrary to what I am proposing, then share them. I would rather learn that be stuck in ignorance and not be able to take advantage of more advanced techniques.)

Rather than providing extremely small attack times, my preference would be to include a separate “limiter” setting that is capable of very fast attack. The advantage being that there could then be an option for using the compressor with or without lookahead (or variable lookahead).

I think a separate “limiter” setting would be pointless. There is already the “Fast Lookahead Limiter” effect.
I think an option for no lookahead would also be pointless. There is already the “SC4” compressor effect.

I believe that people need a low-ratio (< 3:1) lookahead compressor with “regular” attack times. The current attack times, which are all more than 100 ms, make absolutely no sense to me.

What is one example of how an attack time of 100 ms or greater could be used? I’m having trouble thinking of one.

Let’s say that you have a recording that has a very large dynamic range, perhaps classical music or jazz, and you want to listen to it in a non-ideal environment, such as a car. Chances are that either the quiet bits will be too quiet to hear over the ambient noise level, or the loud bits will be just too loud. What is needed in this situation is a slow moving compressor to reduce the dynamic range “transparently”. Radio stations do this as a matter of course.

Running compression without lookahead maintains “transients”, but you can’t then use “make-up gain” because sounds with sharp attack (such as snare drums) “slip under” the compressor. The solution is to limit the peaks. This is a very common technique in studio recording and most hardware compressors have this facility.

Steve,
Ah, ok! Thank you for sharing those. I have never needed to do either of those. I am a live sound reinforcement guy dabbling in editing. My main functions have been to take recorded signal from a live situation (mostly spoken word) and attempt to make it nicer.

It appears that there is very good reason to keep longer RMS attack times available. I will make an adjustment to my spreadsheet.

Thank you so much! I really appreciate your insight here. I was definitely lacking.

I do want to point out that (I think?) I have had good reason to use shorter attack times for RMS compression, besides for Spoken Word:

When preparing background music that is to be ducked under spoken word, sometimes (unfortunately) the background music has a crest factor that is too large. About 70% of the time, I decide to perform some compression on it (even if only a little, ≤ 1.5:1), and maybe 25% of the time it needs “significant” compression (about 1.6:1 to 3:1). The most common thing is when one movement in the music is quiet and then the song explodes into something loud. Great music, but the change is too much.

When using Attack Times ≥ 100 ms, I think the compression pump downward is too noticeable and distracting, whether the music is ducked or not at the moment. So, I use the standard lookahead RMS Compressor with a short attack time based upon the lowest frequency in the passband of the the background music. The lookahead sounds so clean and the short attack time makes the ramp-down before the loud part less noticeable to me.

Hello,

I’d also agree that the compresor could have shorter atack/release times, or at least an option to set up how long they should be. For example, I use Audacity to make some radio promos and by consequence, I need to edit a lot of voice files. When compresing them, I don’t have so many problems, because I try to give them a bigger dinamic range and I am able to do this. However, I miss shorter times to edit the voice and so bring that feeling of broadcast/radio voiceover audios. I have that posibility using Waves compresor, but since they aren’t available for using it in Audacity I must use the build-in compresors. I believe SC4 does not do its job, since it lowers the voice’s volume, but it does not bring that known punch in that kind of material.

Well, those were just my thoughts. Regards.

Why are Waves not available for use in Audacity? Shell VST effects that host multiple plugins should be supported from Audacity 2.1.0 onwards.

Gale

Hmm, could you gibe me some instructions for how to set them up? I tried to load them some time ago into Audition and I could do so, but now I can’t do this for example in RadioBoss, so I’ll be glad if you could bring me some information. Thanks a lot.

Assuming you are using the current Audacity 2.1.1, see How do I install VST plug-ins?.


Gale

Assuming you are using the current Audacity 2.1.1, see
How do I install VST plug-ins?.

Nope, i didn’t like the response of Audacity with Waves; first of all, I couldn’t load or see the presets which the plugins come with, even if the graphical interface is enabled. I don’t know if you can see them in both ways, but at least my screen reader is unable to locate them. And by last but not least, when I move the sliders settings, it seems none of the effects isn’t working when clicking in the preview button or when I aply it after of making the settings required.

I copied the WaveShell files upon the Audacity folder and although the effects are available, that’s what happens with them; I don’t know if anyone has signaled this behavior before.

Anyway thanks a lot for your answer!

Sorry, Miguel, but I did not recall that you were visually impaired. I was only thinking of the fact that recent Audacity could recognise the waveshell files properly.

If your screen reader says there is a Preview button, either you are not using the current Audacity 2.1.1 which has real-time preview of VST effects, or there is a fault in your screen reader.

Really this is not the best place to get help with VI issues. Robert J. H. who contributes much to Audacity Forum is visually impaired, but if you use the Audacity4blind mailing list, everyone there will automatically know you are visually impaired.

Have Waves confirmed whatever compressor you are using is fully accessible?


Gale

Sorry, Miguel, but I did not recall that you were visually impaired. I was only thinking of the fact that recent Audacity could recognise the waveshell
files properly.

Don’t worry, I guess that’s normal because I’m not an often poster, since I try to know some things by asking ere or reading in the existing posts. It’s normal don’t know or recall some things about anyone, although I’m surprised 'cause you know my real name.

I was only thinking of the fact that recent Audacity could recognise the waveshell
files properly.

Well, it can. Even it loads the mono versions of all the plugins; at the beginning I thought it was going to crash but it didn’t happen :slight_smile:. I am able to see the compresors, limiters, the equalicers etc, but with this phenomenon which I have explained above.

If your screen reader says there is a Preview button, either you are not using the current Audacity 2.1.1 which has
real-time preview of VST effects, or there is a fault in your screen reader.

In fact, I have named it like “preview button” because I don’t know how to say it in english, since as you can see, I’m not a native english speaker. I’m using the version 2.1.1 because it also has the vocal remove/extract effect. And yes, when you click in the start playback button (sorry if I have translated it wrong) it’s possible to change the settings in the current effect, such as the volume parameters in the limiter or the OmnyVerb controls. But regarding to Waves plugins, I am unable to use them with the graphical interface enabled or without it. When I move the sliders, the incoming sound is procesed with the default options; I tested it using the Enigma effect. I can’t know what happens with the graphical interface, so maybe they could work with it.

Really this is not the best place to get help with VI issues. Robert J. H. who contributes much to Audacity Forum is visually impaired, but if you use the Audacity4blind mailing list, everyone there will automatically know you are visually impaired.

Anyway I only meant my screen reader in case of doubt, since I’m a blind person, and sometimes I can’t see literally what is on the screen. So I thought maybe you could use the effects or see something. I’ll check that list, and although I am not used to employ them, I’d be interesting to take a look.

Have Waves confirmed whatever compressor you are using is fully accessible?

No, they haven’t. However, I have used allmost all the waves effects in different sound editors with my screen reader, so it’s a bit strange that this is not like that in Audacity.

Ah, another thing: I alwais come here for help, because basically this team is who develops Audacity, and so it’s good to get answers from the people who knows the manner of how it works.

Best regards, and thanks!

Exit Audacity and delete the Users<your user name>\AppData\Roaming\Audacity\pluginsettings.cfg file (I assume you use Windows Vista or later).

Then restart Audacity, create some audio and open a Waves plugin. Press the Play button in the effect then move a slider that you know should affect the sound. Does the sound change? And does the sound change in the same way when you Apply the effect?

When we changed to Real-time preview, factory presets were moved into the menu accessed by clicking the “Manage” button. Have you checked in that menu?

Note that 2.1.1 has an issue that some VST effects do not preview the chosen factory preset or user preset unless you choose it while playing audio. We have tried to fix that for 2.1.2.

David Bailes (an accessibility expert) and Steve who works on this Forum usually look at the audacity4blind list. David uses Windows and Steve uses Linux, so for those platoforms there are people there who know in depth how Audacity works.


Gale