ew1 Posted June 25, 2021 Share Posted June 25, 2021 Hello everyone I have 3 questions about a compression scenario. Hopefully someone can explain what’s happening here. The 3 points are demonstrated in the video in the same order as I have written them here. 1. As you can see from the level meters, the peak of the music is around -11 dB. Why is the compressor input meter not matching the other meters? Sometimes it gets to -11 but most of the time it shows a peak of -13. 2. The compressor’s threshold is set at -18 dB. That means the signal’s peak level (-11) is exceeding the threshold by 7 dB. A 7:1 ratio means that for every 7 dB the signal exceeds the threshold by, the compressor only allows the signal to exceed the threshold by 1 dB. That is 6 dB of gain reduction. So with a -18 dB threshold and a 7:1 ratio, a -11 dB peak should result in 6 dB of gain reduction. However, when I set the ratio to 7:1 I only get 2 dB of gain reduction. Why is this? 3. As I increase the ratio from 1:1 to 2:1 the gain reduction increases. At 2:1 I get 2 dB of gain reduction. However, as I increase the ratio above 2:1 the gain reduction doesn’t change. It stays at 2 dB even at 30:1. Why is this? Thanks Quote Link to comment Share on other sites More sharing options...
fuzzfilth Posted June 25, 2021 Share Posted June 25, 2021 Dude, all that zooming makes me seasick. 1. The input meter may be not a Peak Meter ? 2. Knee values other than zero and Attack / Release > 0ms significantly change the response of a compressor. Also, different compressor models react differently to different types of transients and other dynamic signal. Quote Link to comment Share on other sites More sharing options...
ew1 Posted June 25, 2021 Author Share Posted June 25, 2021 1. The input meter may be not a Peak Meter ? 2. Knee values other than zero and Attack / Release > 0ms significantly change the response of a compressor. Also, different compressor models react differently to different types of transients and other dynamic signal. Is this not always true then? : A 7:1 ratio means that for every 7 dB the signal exceeds the threshold by, the compressor only allows the signal to exceed the threshold by 1 dB. That is 6 dB of gain reduction. I have experimented with every knee, attack and release setting and I am still only getting max 2 dB of gain reduction on the -11 dB peak. I still don't understand why this is the case. If the above statement is true then I should be getting 6 dB of gain reduction. Quote Link to comment Share on other sites More sharing options...
fuzzfilth Posted June 25, 2021 Share Posted June 25, 2021 Is this not always true then? Not if the compressor's reaction has been modified by either of the three parameters and also it very much depends on the type and length of transient and the compressor model. Don't overthink it. - Listen and find what bothers you - Grab a control that can change this - Dial it in with eyes closed until it sounds best - Let go - Repeat Quote Link to comment Share on other sites More sharing options...
analogika Posted June 26, 2021 Share Posted June 26, 2021 Isn't the whole point of having those different compressor models (and in fact, ALL the different hardware compressors) that they all react differently? How exactly they react at different volume levels above the threshhold is part of what makes their signature sound. No hardware compressor is going to give *exactly* the stated ratio for ALL values above threshold. Also, if you've got a slow response, the initial attack is going to go through at full volume before the compressor gets a chance clamp down. Do you like the way it sounds? Trust that. Quote Link to comment Share on other sites More sharing options...
Mania Posted June 26, 2021 Share Posted June 26, 2021 Isn't the whole point of having those different compressor models (and in fact, ALL the different hardware compressors) that they all react differently? Oh yes! A FET type of compressor sounds so good on drums for example. Other types just don’t get the same result because they react differently. Very important to use your ears not your eyes. The end result is an audio file, not video. Quote Link to comment Share on other sites More sharing options...
ew1 Posted June 26, 2021 Author Share Posted June 26, 2021 Is this not always true then? Not if the compressor's reaction has been modified by either of the three parameters and also it very much depends on the type and length of transient and the compressor model. Don't overthink it. - Listen and find what bothers you - Grab a control that can change this - Dial it in with eyes closed until it sounds best - Let go - Repeat Thanks Quote Link to comment Share on other sites More sharing options...
ew1 Posted June 26, 2021 Author Share Posted June 26, 2021 Isn't the whole point of having those different compressor models (and in fact, ALL the different hardware compressors) that they all react differently? How exactly they react at different volume levels above the threshhold is part of what makes their signature sound. No hardware compressor is going to give *exactly* the stated ratio for ALL values above threshold. Also, if you've got a slow response, the initial attack is going to go through at full volume before the compressor gets a chance clamp down. Do you like the way it sounds? Trust that. What do you mean by "response"? I assume that's different to attack time. Quote Link to comment Share on other sites More sharing options...
analogika Posted July 12, 2021 Share Posted July 12, 2021 Isn't the whole point of having those different compressor models (and in fact, ALL the different hardware compressors) that they all react differently? How exactly they react at different volume levels above the threshhold is part of what makes their signature sound. No hardware compressor is going to give *exactly* the stated ratio for ALL values above threshold. Also, if you've got a slow response, the initial attack is going to go through at full volume before the compressor gets a chance clamp down. Do you like the way it sounds? Trust that. What do you mean by "response"? I assume that's different to attack time. That's part of it. But different compressors and different compressor types respond differently — some are inherently slower than others, so attack times aren't the same across models. Also, none of them respond perfectly linearly; they will ramp up differently. If you have an attack of, say, 50 ms, until the compressor is at full response, it makes a difference whether ramps up slowly and then fast, or crunches hard immediately but takes a little longer to reach the peak, etc. They also won't respond to all frequencies equally, so that alone will affect how they work on different material. In addition, the gain circuitry works differently, so the signal the compressor itself reacts to may already sound slightly different etc. Again, the point is to find a model that works on the signal in a way that you like and use that. And if you don't like any of them on the signal, to not use one at all. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.