Muddy sound

Just too much content ?

Any yet another thread …

I’m trying to mix 4-5 tracks (kids-choir, piano, “sfz-drums”, accoustic guitar, a synth line)

Most if the things I’ve recorded (even instruments) are in stereo. I wonder if this is not “overkill” and actually the cause of to “too full sound” (in the area of 700-1500Hz), “too much content”.

What are the common thoughts here :

Is there a difference a “pseudo-stereo” (LEFT/RIGHT have the same signal) or a “real mono” panned right in the middle ?

It is OK to really CUT OFF low & high of a spectrum rather than lowering it a bit ?


Lduo

The EQ primer is my “bible”

…it’s a good place to start. Too much information in the low end can add up to a muddy sounding mix. Most tracks really don’t need much below 100Hz or so… I usually use a low cut shelf on every track except bass and drums.

The are the tracks you recorded actually stereo, or just two-track mono?

Each voice needs to have its niche in the frequency range to appear clearly in the soundfield. The midrangy sounds often clash; vox, top end of the bass, low end of the guitar, keys etc.

This is where parametric EQ is your friend. Some suggest using frequency analysis programs, but I say just listen and you can hear where the mud is blending. it was surprising to me to learn how you can remove portions of an instrument’s spectrum, and by itself it sounds a little off, but blended back in it just “sits down” in the mix nicely. As an example I usually roll off guitars below about 200Hz to not compete with the kick or bass.

One thing many people overlook is arrangement. If the bass, keyboads, and guitars are all occupying the same space in the arrangement, no amount of EQ will fix that. If after a decent amount of EQ you don’t get what you want, look at the arrangement and see if everything is tripping over one another.

Hi ludo. Maybe you know it, but the Limey’s piramid concept really help me when mixing:

http://www.audiominds.com/mixing/pyramid.php

Hmm - thanks you all. It sounds like it is indeed just what it suspected : too much/rich content.

I’ve redone a couple of tracks where e.g. piano is now “simple & one-handed” rather than “double-handed all the way”, and it’s getting better.

Re. spectrum : indeed I see quite some overlap, but it seems such a shame to me to cut down on e.g. a “rich string”. I gather you guys would just do all that : remove remove remove …

Ludo

Don’t get locked into what a track sounds like when it’s soloed. If every track sounds great by itself then there’s a good chance there will be build-up and mud, depending on the content. Normally a great full mix will have many tracks that sound thin out of context with the other instruments. It’s all an attribute of masking…and it’s VERY difficult to learn and hear sometimes. Cutting down on content is on fix, but so it narrowing the frequency bands the individual tracks use.

Give each track its own color. If each one the tracks take up a wide range of colors (great sounding soloed) they are all brown, and brown on brown makes it hard to tell the tracks apart.

what happened to the eq primer?

the website says that it is for "paid subscribers only"

i check this chart all the time while i am mixing. if this is gone, i will have to find a new starting point for my recordings.

if someone has a copy of this, i would greatly appreciate a copy. PM if anyone has one

thanks
-Mike

what happened to the eq primer?

the website says that it is for "paid subscribers only"

That's bizarre. I referred to it only yesterday.

Ho hum

I agree with all the above. They are true regardless of the stereo/mono issue.

Stereo imaging, when used properly, increases the amount of “space” for content to sit in, and increases your mind’s ability to discriminate between parts. So, stereo imaging by itself would not be the cause of creating mush. However, too many stereo miked tracks might be, I’m not really sure about that.

I used to stereo mike acoustic guitar, but I’ve found I get better results with mono miking and using n-Track Reverb to generate a little stereo image. (I click the “expand mono track to stereo” in Track Options, and use the reverb with “Ambience 2” setting for starters, but with damping increased because the early reflections are more important for image than reverb tails.) If you can try this with one of the tracks for your guitar, it’s worth looking into.

I like a stereo miked piano. The choir really is best miked in stereo too. No doubt the synth sound comes stereo out of the synth and sounds best that way. And drums definitely have to be stereo. So, you’re not going to gain on these tracks by going more mono. Of all of them, piano is the least necessary to be stereo miked (and then I’d still use mild stereo reverb).

The advantage to stereo miking is that the image is natural. The disadvantage is that you have little control over the image. You can plug in “MDA Image” and use it in “LR->LR” mode to increase or reduce the stereo “width” for any of these tracks, which is worth a try (along with panning the track, if appropriate).

My guess is that the real problem is the rooms. If they’re too lively, you’ll have too much natural reverb and there’s no way to get rid of it. Reducing the width (with MDA Image) probably won’t mitigate the muddying effect of the reverb – in fact, I’d expect it to worsen it. If you recorded all the parts in the same room, any problems in the room will be amplified. (This is also true of mikes and mike preamps – best to use different ones for different tracks, as much as feasible.)

These are things to think about, but I think they’re mostly second-order compared to the ones mentioned above: arrangement first, EQ-carving second, stereo imaging third. Room and mike issues probably come between arrangement and EQ.

Oh, also: I have a suspicion that, in a bad room, stereo miking is worse than mono miking, for the simple reason that it makes the flaws more clearly audible. If you have room problems, you can try using more damping materials (heavy drapes on walls, gobos, etc.)

Hi - I’ve been experimenting AND learning an awfull lot of all of your comments - thanks.

It really came down to two things in the end :

- too much content (like piano played with lots of notes over the whole keyboard, organ/strings played with lots of notes just the same) so I basically cut out over half of the notes : much better.
- the remainder of the content, I cut more in frequency band pieces.

I feel I sort of “get control” again rather than only trying not to drown.

Thanks you all !

Ludo

You’ve gotten some very good advice here, Ludo. I’ll add this…

Next time you set down to mix, try an experiment. Take a track that you think is muddy. Clone the track. On one copy EQ thin, deep cuts in a few critical frequencies. Then EQ a wide, shallow cut in the other track. Compare the two, and you’ll often see that thin, deep cuts tends to add clarity to the track without drastically changing the tone. I think this is what Phoo was referring to when he mentioned ‘masking’. When 2 instruments share a frequency, it’s best to cut one of them and let the other shine thru. Cut one of them at one frequency, and cut the other at a slighty different frequency.

Also, cut instead of boosting. The human ear is used to hearing frequencies cut in a room (absorbed by furniture and such). But a boost will sound unnatural.

Hope this helps some.


teej