Mixing 101: trouble with levels in/out of channels.

[TheElement]

Just watched David Guetta masterclass. He was getting +6 on the master channel. so for electroc music seems you can really push things more. Heard Oliver Helms say he doesnt use compression. His mixes are already high in volume. so I guess forget dynamics and use up all the headroom is the key..however way you get there.

[Len911]

Headroom is generally associated with magnetics or old VU meters, or analog magnetic tape, and distortion with vacuum tubes and big transistors, not really microprocessing transistors of the chip type and digital.

The simulated programmed digital variety were never designed to be "pushed".

It's like playing headroom with a Van de Graff static generator making your hair stand on end, which can be fun, the simulated variety though is like sticking your finger in a light socket, similar, but with worse results.

[nolosolo]

When I saw mixing 101 in the title, I thought I found what I was looking for. As a hack I can usually get the job done with what I have to work with but just ran into a case where no matter what I do, the final mix is weak and the more I mess with the mix, the worse it gets. Here's the situation and BTW, please excuse my total lack of proper terminology in describing my problem.

The case of the lost masters. First I'm only using a Boss digital 8 track recording direct stereo tracks from all the instruments (E-guitar, Vdrums, keyboards) and now editing with Audacity. I found an old instrumental I wanted to add vocals too. Unfortunately the original tracks for it are lost and the final instrumental mix seems very compressed to get a good output which in itself is fine.

The problem is simply trying to mix in the vocal track. No matter what I try to do to it, I need to lower the main mix to keep it from not be buried in a muddy mess or clipping all over the place. (you can see it poking out in the pic) I've played with every logical effect on the menu with no clean results. (normalization, leveler, more compression, dynamic range, etc. etc. etc.)

Now I know what everyone's thinking "this guy needs to go back to school" but this thread is "mixing 101", right?

ANY help will be greatly appreciated! I already submitted the song and it would be nice to exchange it with a better mix before it's evaluated.

Thanks in advance

[Len911]

It "looks" like you need to compress the vocal to fit to the instrumental track, and not the instrumental to the vocal track, in this case, because you don't have the original tracks.

[nolosolo]

It "looks" like you need to compress the vocal to fit to the instrumental track, and not the instrumental to the vocal track, in this case, because you don't have the original tracks.

Hi Len, tried that but the vocal gets totally lost in the mix.... Here's the link to the track is anyone is interested is hearing it as shown:

http://www.taximusic.com/hosting/home.p ... st=kentone#

Title is: "(my)Same Old Song"

(sorry, don't know how to direct link to it)

[lesmac]

I think Len's on the right track with compressing/limiting the vocal to squash down some transients first.
I would have some fun cleaning up the instrumental track then mixing the tracks together later through a processing chain with minimal settings to glue them together.
2 cents from me.

[nolosolo]

I would have some fun cleaning up the instrumental track then mixing the tracks together later through a processing chain with minimal settings to glue them together.
2 cents from me.

Thanks Les, could you elaborate on the "cleaning up" functions? I was thinking about getting more contrast between vocal and background via EQ or something else to make it pop out without the volume difference maybe?

[lesmac]

Hi Nolo,

Thanks Les, could you elaborate on the "cleaning up" functions? I was thinking about getting more contrast between vocal and background via EQ or something else to make it pop out without the volume difference maybe?

Listening on laptop and consumer headphones the bass was boomy and it seemed a bit muddy up the middle, so yeah I was thinking eq adjustments.: a mid side processor so you can eq up the middle and maybe drop the level a bit as that is where you want your vocal to go as well. Don't know if the cymbals were a bit harsh at times, maybe low pass filter a bit. I'm no pro so these are just humble suggestions.
All the best.

[nolosolo]

Ah, think I'm getting the idea. Sounds like a plan to me!

Thanks again!

Ken

[mojobone]

Been doing this for a few months now, still sometimes having problems getting proper levels in my mixer (Logic). I'm trying to understand the signal path, perhaps there is a reference book or article I could read to get the main idea.

I have some patches that send the signal hot to the mixer, so I'm clipping even if the fader is all the way down. So, I go and lower the output signal from the patch to keep from clipping, but often I'm not able to get enough volume from the patch with the slider all the way up. I've added limiters to the channel strip to compensate, but I think I'm doing something wrong/stupid. In general I try to keep my sliders at or below 0 dB. (is the space above zero the "headroom" I often read about?

Maybe I'm mixing at too high levels, maybe I should start everything lower and just turn up the main output or amp on my speakers instead? Does anyone know what I'm talking about? Does anyone know what time it is? Does anyone...

I recommend the Yamaha Sound Reinforcement Handbook; it's a bit of a slog, but it covers ALL of the basic concepts that you'll need to know before delving into Mike Senior's Mixing Secrets For The Small Studio and Bob Katz' Mastering Audio; The Art & The Science.

It sounds to me like your problem could be dealt with at the patch level in your synth, but the way you should address it is kinda dependent on whether it's modeled, analog or a ROMpler. For example, the Korg M1 family has single patches and Combis of up to 16 parts or layers, so you can imagine that there might be significant level differences among/between them. This is one of those times that you can actually use normalization. (with caution; any level increase of this kind will also bring up the noise floor) Generally speaking, when you compress or limit you lose not only headroom but also transient detail and punch. If you're interfacing hardware with a mixer you have to be aware of the differences between line level, instrument level and mic level and whether the mixer operates at pro +4dBv or consumer level -10dBv. For analog modelers that build everything out of sine, saw and square waves the sounds can sometimes be so bland that you may need to squish 'em 'til they grow hair, just to make them sound interesting.