Is automatic gain compensation for EQ something we want?

[xandde1]

Hi! I’ve been wondering if we should EQ with automatic gain compensation turned ON (especially for mastering).

It seems to me that we trick our ears by moving the entire spectral curve, instead of keeping the spectrum still and changing only a specific region, but I might be wrong.

What do you think?

[DPhil]

I've never used AGC for EQs and I don't see a reason for it. For example adding a highpass makes the higher frequencies still feel the same, there's nothing to compensate imo. Or a simple bell filter acts on a comparably small range. Like you said: why do compensation and moving the unaffected spectrum too?

[plexuss]

Yes, for me ALC would be useful. ALC = automatic loudness compensation - gain beased on LUFS. There are many times in mixing and mastering, especially mastering, where I'd prefer to have ALC. Sometimes I will choose an EQ because it has it over one that doesn't. It just makes my job easier and faster.

[imrae]

Yup, AGC is wonderful. Even imperfect AGC can reduce the extent to which we are fooled by loudness changes.

My go-to EQ is SlickEQ GE, largely because I find the AGC works very well. I recently discovered the free Melda AGC plugin and found it pairs really nicely with the IK Neve 73 plugin; you can easily play with using the input drive to smash off the transients, while adjusting the EQ to rebalance the spectrum.

As far as I can tell Luftikus is straight-up better than the Maag EQ4 because it has auto-gain. On that type of EQ the "high shelf" boosts extend a long way down so is easy to be fooled into over-using it.

[xandde1]

Thank you all for sharing your thoughts!

Do you think that AGC is useful even when cutting? I’m asking that because I recently bought Har-Bal mastering EQ and the loudness compensation is the only thing that really bothers me, because it makes small changes sound bigger than they are. In my head, if I’m cutting some harshness at 2kHz, I should compare the original spectrum to the one with less 2kHz. If I compensate for that cut, I will be comparing the original spectrum to another that has more lows and highs, which will completely trick my ears.

Any thoughts on that?

[shoma]

As I have already mentioned in your other thread.... Don't confuse Har-Bal with another standard EQ. Take a look at the videos where Paavo explains how the eq works and how to use it to get satisfactory results. Small changes can have a big impact, yes. But they should be related to the unchanged sound file in terms of volume, so that you can also make statements about whether you have actually improved something or just made it louder or quieter. That would be misleading. Louder isn't better, but one can be influenced too much by it.

[imrae]

I find AGC most useful for "shaping" when making major changes to the balance of the spectrum. A small cut at 2k seems like a great example of where you might want to turn it off; you're already happy with the general balance and want to focus on removing something unwanted.

[highkoo]

I have never once had the urge for it.

People still talk about Har-bal, wow!
Its all about Gullfoss now!

[Soundplex]

You have a nice vocal, you remove the low rumble, now the AGC makes everything a couple dB louder. How can that be useful?

[xandde1]

As I have already mentioned in your other thread.... Don't confuse Har-Bal with another standard EQ. Take a look at the videos where Paavo explains how the eq works and how to use it to get satisfactory results. Small changes can have a big impact, yes. But they should be related to the unchanged sound file in terms of volume, so that you can also make statements about whether you have actually improved something or just made it louder or quieter. That would be misleading. Louder isn't better, but one can be influenced too much by it.

Thanks for your thoughts, shoma! Har-Bal is just an example, but the same applies to FabFilter Pro-Q3 ou TDR SlickEQ with AGC turned on. Please don’t misunderstand me: I’m really enjoying the results Har-Bal delivers. What I’m questioning is the principle of comparing the original audio source with the EQ’d one at the same loudness level.

If you take away 2kHz and compensate the loudness, you will have less of this frequency, but also more lows, low-mids and highs in relation to the original file. When you do not compensate in this case, you have just the same lows, mid-lows and highs, but less hi-mids, as intended. I’m not saying this is better, but only that it is logical. For drastic EQ changes, I can see how useful AGC can be, but in most cases I’d guess it is misleading.

highkoo, Gullfoss is also great and gets a lot of use here, but they are different beasts! =)

[xandde1]

I find AGC most useful for "shaping" when making major changes to the balance of the spectrum. A small cut at 2k seems like a great example of where you might want to turn it off; you're already happy with the general balance and want to focus on removing something unwanted.

Agreed! That’s exactly how I see it!

[No_Use]

I've asked that too couple of years ago, maybe it gives you further thoughts:
viewtopic.php?f=6&t=419707

[Dasheesh]

i like the agc on eq's, because it takes the signal up the point of added distortion from the eq for a cleaner signal. it's best to use eq's in a subtract rather then an add. boosting eq's are specialized and used to color sound. agc is used in situations when you want to limit the color and added distortion of an eq.

[acousticglue]

It's like being on a diet for nothing

[heavymetalmixer]

I used to think it was needed, but nowadays I don't. Actually, I even think it's not a good idea: If I want to turn down something I don't want the rest to go higher.

We all know that louder is an illusion that makes us think it sounds better, but turning up the fader doesn't sound the same as just turning up a certain frequency range.