Your advice broke my mix!

timvracer

Member
Forgive me for the clickbait title...:-)

Per the advice on the board here about creating loud mixes, I went back to my mixes and went all OCD on the low end part of the mixes. I did not blindly put a HPF on things, but listened to every track, and then added appropriately either a HPF or a shelf to instruments for which the very low end is not needed (I checked below 100hz). I am using the PAZ Analyzer to visualize the frequency peaks across the spectrum, and definitely it verified that I was pretty stacked up on the low end, and my changes had the desired effect of lowering the stack/peaks on that spectrum. The extra headroom also allowed me to better define my kick drum and add some "punch" to it, as well as actually add some very low end to the vocal to give it some deep resonance.

The result to the overall mix, however, was remarkable. It certainly got "louder" (or shall I say, gave me headroom), but now my higher frequencies are too loud. It is as if the higher low end was attenuating the high end, and now it is too loud. Is this how it works? If I then go and reduce those frequencies that seem too loud, am I not just right back to where I started?

Thanks for your thoughts - I am relatively new at this.
 
Last edited:
It is as if the higher low end was attenuating the high end, and now it is too loud. Is this how it works? If I then go and reduce those frequencies that seem too loud, am I not just right back to where I started?

It's not really an "either/or" kind of thing. Sure, if one area is way too loud, it may mask something else, but you CAN have both a fat/punchy low end, nice highs, and loud volumes.

It's about finding the right balance, not so much replacing one with the other.
So..go back to it and fine tune it. Take out some of the highs, etc...find the balance.
 
First things first, what's your mixing environment like?

A suggestion is to not look at the spectrum analyzer, don't look at the fader positions nor the track meters. Try to use your ears to find a balance, bot your eyes.

Post your mix up in the MP3 clinic and people can give you their thoughts on it.
 
First things first, what's your mixing environment like?

A suggestion is to not look at the spectrum analyzer, don't look at the fader positions nor the track meters. Try to use your ears to find a balance, bot your eyes.

Post your mix up in the MP3 clinic and people can give you their thoughts on it.

Thanks, yes.. I use my ear as the final arbitrator, but my ear was not able to know that the LF were causing the problems (thank you to the board for educating me on this!), so the analyzer helps me to see which instruments are putting info there (below 100hz). However, not using it to decide on the mix.

I say "ear" purposefully, as I am 90% deaf in my left ear... which is a topic to itself, I have developed my own techniques on mixing with one ear!

In anycase, I mix on Sonar Platinum, with Focal Pro Alpha 65 monitors in a less than ideal room. I reference my mixes in my car, and on 2 home stereos with B&W towers (on both), one with a sub, one without a sub.

I guess I was just looking to understand the way the mixing process works (digital). I was actually very pleasantly surprised that when I removed a bunch of the stray LF that the rest of the mix "came to life"... Of course I was being provocative with "broke my mix", it was clearly a great improvement (but now I have to balance the higher freq's).

---------- Update ----------

It's not really an "either/or" kind of thing. Sure, if one area is way too loud, it may mask something else, but you CAN have both a fat/punchy low end, nice highs, and loud volumes.

It's about finding the right balance, not so much replacing one with the other.
So..go back to it and fine tune it. Take out some of the highs, etc...find the balance.

Cool... I am on it. Thanks!
 
Hi, you say you "checked below 100hz". Are you saying you started to roll off from 100hz on your low end instruments? I'd suggest around 20hz as a barometer. Also try (and this is ball-park stuff depending on your mix) a bump around 30hz. This will retain the warmth to a mix whilst removing the low lows.

A great way to make sure these moves help your mix is to try the top-down approach. Put these EQ moves on your mixbus at the outset, together with the plugin chain you might ordinarily apply to you mix bus, and mix down from there. Pull up your static mix fist, then apply the mix bus processes, then your bus processes, then your channel processes (if needed). You'll also find you use less plugins on your overall mix, whilst getting the balance between frequencies right because you've already applied a cut and a boost.
 
Forgive me for the clickbait title...:-)

No, it was pretty accurate. The rampant "roll off everything below xxxx" mantra is spouted off way too much and the more it gets repeated, the more it becomes fact to those that don't know any better. You're just the newest victim of the inexperienced groupthink prevalent in here.

Go back to the drawing board, don't EQ out low end that you don't need to, and stop mixing with your eyes.

In the future, track better sounds to begin with so you don't have to go apeshit with EQ just to make the mix work.
 
Forgive me for the clickbait title...:-)

The result to the overall mix, however, was remarkable. It certainly got "louder" (or shall I say, gave me headroom), but now my higher frequencies are too loud. It is as if the higher low end was attenuating the high end, and now it is too loud. Is this how it works? If I then go and reduce those frequencies that seem too loud, am I not just right back to where I started?

Thanks for your thoughts - I am relatively new at this.

If you were applying compression to your mix, the bass was triggering the compressor and pushing the volume of everything down.
Also, treble takes very little energy to get loud.
It could be said that LPF/shelving should also be applied to tracks with offending brightness, and the entire mix should be redone afterwards.
Mixing is a lot like cooking.
When you use less salt, you'll find out that you've been using too much garlic.
 
If you were applying compression to your mix, the bass was triggering the compressor and pushing the volume of everything down.
Also, treble takes very little energy to get loud.
It could be said that LPF/shelving should also be applied to tracks with offending brightness, and the entire mix should be redone afterwards.
.. Option / compromise # 99 - One of the reasons we like compressors with side chain eq. Now you get to choose how much various frequency bands control the compressor. Very potent stuff there :D
Mixing is a lot like cooking.
When you use less salt, you'll find out that you've been using too much garlic.

Good one! :)
 
Last edited:
As somebody else said, sounds like the low end was hogging compressor real estate. So when you lowered the bass the floor rose for everything else. Don't re-eq. Remix. The eq did not get brighter, it got louder. Remix.
 
Thanks everyone for the insights, super helpful. Just the notion of how LF's impact the mix was really good to learn. As you can imagine, I over-reacted to the information and pulled out too much LF (rolled off at 100hz) on too many tracks, so I've gone back in as has been suggested and started balancing more sanely. The mixes are coming out much better and with more clarity, and indeed louder.

Unfortunately, these tracks were done in a basement, I don't control the source, so I am having to work with what I got.
 
.. Option / compromise # 99 - One of the reasons we like compressors with side chain eq. Now you get to choose how much various frequency bands control the compressor. Very potent stuff there :D


Good one! :)

I have a question on this... does the side chain cause the entire track to be compressed? I am using a multi band compressor to compress just the narrow band frequency with an offensive ear piercing treble (elec guitar), and it seems to work well. I have not used side chain, but I am curious now to know if it is the same as what I am doing with the multi-band?
 
I have a question on this... does the side chain cause the entire track to be compressed? I am using a multi band compressor to compress just the narrow band frequency with an offensive ear piercing treble (elec guitar), and it seems to work well. I have not used side chain, but I am curious now to know if it is the same as what I am doing with the multi-band?
Yes, but two different things.
The MB splits it into specific freq bands- so you can zero in on just that trouble spot.

The side chain is simply a term referring to an access point to a compressor's detector.
In this context, with an eq to alter how a compressor 'hears the freq balance of the signal.
They're always seeing (reacting) to level, but getting in there with eq allows you to make it more or less sensitive to various frequencies.

When you compress mixed material you can give some thought as to what might be 'driving the compressor (before you even get changing the the side chain.
What's loudest (sticking out above the average'
How fast (how short or long) are the loud things lasting - compared to the attack time. Very relevant as the compressor ignores level (to some extent) as it ramps up to the attack time.
And last, after those, it reacts to the tone balance of the signal, vs the detector's eq.
 
I have a question on this... does the side chain cause the entire track to be compressed? I am using a multi band compressor to compress just the narrow band frequency with an offensive ear piercing treble (elec guitar), and it seems to work well. I have not used side chain, but I am curious now to know if it is the same as what I am doing with the multi-band?
Sidechain and multiband are different things.

Sidechain involves one compressor. The whole signal is adjusted based on what the sidechain "hears".

Multiband really is multiple compressors in parallel, each reacting to and adjusting it's part of the frequency spectrum.

I don't use either. What I like to do is "preemphasis/deemphasis". Basically, wrap complimentary EQ around the compressor. For example, I very often will shelf down the low end a little before the comp and the shelf it back up a bit on the other side. The effect is similar to what you can get from sidechain EQ, but different, and just more intuitive to me.

One thing I think many people don't realize is that a compressor is pretty much always working off of an average of some number of samples that have happened in the past. Whether it's the attack/release process or an rms window (or both), it's always kind of smearing its response. This has an effect a bit like a lowpass filter in the detector circuit even without explicit sidechain EQ. It's not exactly the same, but it works out that a compressor responds to low frequency content more because those low freqs tend to stay louder for longer.
 
Sidechain and multiband are different things.

Sidechain involves one compressor. The whole signal is adjusted based on what the sidechain "hears".

Multiband really is multiple compressors in parallel, each reacting to and adjusting it's part of the frequency spectrum.

I don't use either. What I like to do is "preemphasis/deemphasis". Basically, wrap complimentary EQ around the compressor. For example, I very often will shelf down the low end a little before the comp and the shelf it back up a bit on the other side. The effect is similar to what you can get from sidechain EQ, but different, and just more intuitive to me.

One thing I think many people don't realize is that a compressor is pretty much always working off of an average of some number of samples that have happened in the past. Whether it's the attack/release process or an rms window (or both), it's always kind of smearing its response. This has an effect a bit like a lowpass filter in the detector circuit even without explicit sidechain EQ. It's not exactly the same, but it works out that a compressor responds to low frequency content more because those low freqs tend to stay louder for longer.

Got it, excellent and thanks. I think MB is what I needed on the guitar, although I found some use for the side-chain on a kick drum (again, inconsistent drummer) where I needed to trigger off the "click" (HF) to bring it down on the hot kicks. It worked quite well. I did not want to just EQ it because the smack/click was just right except in those few times the kick was too hot. I guess there are lots of ways to skin the cat, but this was a great chance for me to try and learn the sidechain thing.

Thanks!
 
Last edited:
Back
Top