Kithairon's YouTube thread

[MogwaiBoy]

I think the "don't boost, only cut" rule/guide thing is a bit overstated. I know in general you will run into less digital-ness but people will have their own preference.

I am waiting for an EQ that can convert boosts to inverted cuts, though

[Guenon]

Just a heads-up, Kithairon, your video has some conflicting info.

On the other hand, what you describe as the most important tip (and imo it's a really good tip indeed!) : "Do not EQ if it's not necessary. Listen to the sound. Think to yourself, is it bad? Should I EQ it?"

But then again, earlier in the video: "Cut the sub on most elements."

As cutting the lows is obviously EQing , and you actually don't need to do this on most elements, it goes against your most important tip. In other words, in great many cases that is exactly "EQing when it's not necessary."

See here, for example:



Yes, that's a good watch despite the "provocative" name

[Stamped Records]

The problem with that, is how do we know when and when not to eq if we don't play aroubd and get it wrong a few thousand times.

[BlackOctopus_Slade]

Cool input and video. I want to chip in here too, as there is so many times I hear lifeless music because of overly EQing elements, and much of that comes from the assumption you have to EQ everything. Even stems I may get for mixing, the producer had over done the EQ on his source sound. This is very genre-specific too, of course. Some great tips in there.

But as mentioned above, some conflicting views on EQ's approach in mix or productions such as low-cutting most all elements for sub. This can do more harm than good.

The don't boost part is not always true in most cases to be honest. Cut first then boost later is a better rule of thumb in some channel chains. For instance, with many analog variety of EQ you can get amazing sparkle or lift from boosting. But you may need to tame first before doing so. In many elements to gain attack you wont want to cut all around where you can notch and boost instead. Adding air is a great way to boost too. Also running post-compressor is a great way to boost where the compressor may have dampened (i.e. parallel this is important).

Not always such thing as "bad frequencies" either. Clashing? sure, it happens. Figure out which one needs to take the lead. Bad resonant frequencies? Definitely tame these. Dynamic EQ sometimes is best bet for this.

The More analog a sound is the less EQ, this is a good input too. But also, it is to be stated that MOST analog signals and recorded instruments will also need much more EQ treatment in the long run. Maybe not as "drastic" depending on signal but definitely way more focus than a drum sample already processed. So this isn't fully accurate.

My main input I give is that any and ALL things that are already highly processed, i.e. in most sample based productions/electronic music productions and so forth, wont need near the attention of EQ at the start as lets say, a rock recording or recorded material when it comes to EQ (or even compression). Overdoing the EQ stage will end up sounding super lifeless, digital and not very natural at the end. It's good though to get the idea of WHAT EQ is doing.

Bottom line, don't overthink, overdo, or overuse in most cases with Mixing and processing such as EQ. Let your ears be the boss. Does it sound good? Is the source sound problematic? Does it fit together due to great production design and giving you better 'puzzle pieces' to fill the spectrum? These are the things that matter most. EQing then can be used to fix or adjust to be sure they do.

[Stamped Records]

The question is, what does "very little EQ" actually mean? This is where the conflict arises in the mind of people who don't understand EQ. Unless you're going to be taught by somebody, in a studio, on a live project, all these words really mean very little indeed.

For example, a little EQ could refer to the amount of bands used, or it could refer to the amount of cut/boost - and there in, lays the endless circle of experimentation that will be your true teacher.

Personally, I disagree with the whole ethos of not boosting, although, generally, I think a good rule for boosting is one boost per track - that doesn't mean you have to boost once per track, but if you do boost, one is enough, because in theory, cutting in the middle of where you would put the two boosts has the same effect, and is much better practise.

Another thing that can lead to good EQ results when looking for cut points, is not to stop at the first point that sounds good. I've found, while learning, that EQ can be a trickster and that if I go past the first point that sounds 'right', I often find that a little further in whichever direction I am going, sounds more right. Basically, cutting on either side of the frequency you want to keep, will sound good, but one will sound better and accent the frequency you want to keep, more.

Also, over the years, I must have seen a thousand times, someone saying, cut narrow, boost wide, but not a single individual has ever specified approximate q values that represent the terms, narrow and wide. It's all a bit abstract.

[ramseysounds]

1. Don't boost the signal, only cut.

One of the biggest EQ myths known to man.
Just get the low end sounding great and then EQ everything else.
The longer mis-information like this is spread, the more kids just starting out will learn bad habits and wrong ways of doing stuff.

[DJ Warmonger]

1. Launch iZotope Tonal Balance Control in fine mode.
2. EQ until your spectrum fits exactly in the middle of a graph.
3. Profit. (No, seriously, just signed another contract yesterday)

[Stamped Records]

3. Profit. (No, seriously, just signed another contract yesterday)

So only 3 months to wait for that 100 dollar check Kidding man, congratulations.

[Stamped Records]

1. Don't boost the signal, only cut.

One of the biggest EQ myths known to man.
Just get the low end sounding great and then EQ everything else.
The longer mis-information like this is spread, the more kids just starting out will learn bad habits and wrong ways of doing stuff.

I also concur with this opinion. Now that I've come to grips with EQ, I feel extremely misguided by a lot of information regarding EQ.

[harveywright1995]

cool video

[Stamped Records]

I think knowing what you are doing, is the only secret there is, although some of the things that have been 'big' for me while experimenting, seem too big not to have seen mentioned by anyone all that much. Like, if I had a time machine, I'd tell the beginner me that there is usually one big cut if its necessary at all, to get the eq process going, if a track needs it, I will 9 times of 10, find a nice place to take a massive chunk out of the frequency range, while doing no harm whatsover to my percieved balance.

Also, I think it's key to do most of your work with all tracks in and good attention has been paid to the balance. A good balance in the beginning is key and too much muting and soloing will cause the mix to continuously lose shape when you bring things back in after workin on other things. Obvious stuff.

Hard for a beginner to reset a preset too, but thats what I have to do a lot, it's the bet thing for it.

Really the hardest thing is being objective about your own material, and trying to see it as a whole instead of the dozens of parts and sections it contains.

[phoenixdenim45]

There's some interesting perspective on eq techniques in the book Mixing With Your Mind https://www.mixingwithyourmind.com/

He advises not sweeping the EQ to find your frequencies as your ears are very good at adjusting to these changes so you quickly lose perspective. He suggests "guessing" what you think the settings should be with the eq bypassed, then unbypassing it to see how close you came. If you were wrong, bypass it again and guess again. It's a subtle difference to the targeted sweep techniques mentioned above, but gave me a new approach to play with.

Also one other cool tip is while mixing when you're eqing a track, listen to the other tracks around it as opposed to the track itself. He uses the example of vocals and piano where if you are trying to make the vocal track stand out, make sure you are listening to the vocal instead of the piano as you try to find some helpful eq positions.

[Eclectrophonic]

Just a heads-up, Kithairon, your video has some conflicting info.

On the other hand, what you describe as the most important tip (and imo it's a really good tip indeed!) : "Do not EQ if it's not necessary. Listen to the sound. Think to yourself, is it bad? Should I EQ it?"

But then again, earlier in the video: "Cut the sub on most elements."

As cutting the lows is obviously EQing , and you actually don't need to do this on most elements, it goes against your most important tip. In other words, in great many cases that is exactly "EQing when it's not necessary."

See here, for example:



Yes, that's a good watch despite the "provocative" name

Why would the peak level alter on that piano at the beginning from -1.6dB to -1.3dB when looped?

[Guenon]

Why would the peak level alter on that piano at the beginning from -1.6dB to -1.3dB when looped?

As it's a looping MIDI clip triggering an instrument instead of an audio loop, when ever it loops from the beginning there is also the instrument's tail of the last pass still sounding. The first hit also happens to be the loudest one (it sets the loudest peak, -1.6 dB, right away on the first pass), so when the phrase loops, there's 0.3 dB extra from the tail, adding up to -1.3 when looping over and playing that loudest first hit again.

[Burillo]

1. Don't boost the signal, only cut.
1.a) Or boost the whole signal and then cut.

this is a long-standing myth that just isn't true. there is no difference between boosting and cutting. there was a Dan Worrall (?) video about this somewhere, with phase-cancelling tests and everything, proving conclusively that this is just not true at all.