Suggested Output Levels For Mixing?

D

Doctor Varney

Cave dwelling Luddite
The stereo line level inputs of my (physical) mixer each feature a level switch, which toggles between +4 and -10. My computer's PCI L/R out is connected to this.

I output my mix using a VST limiter (or multi-band compressor with the limiter switch on) which lies after EQ and other effects. My speech usually averages -6 and, at the highest points, just slaps the bottom of zero dB on my software's peak meter.

My question is, which of these level options on my mixer gives the truest account of the level coming from the DAW? I use a separate amplifier and passive speakers to monitor on. Thus the volume hitting my ears is dependent on a mixture of sliders on my mixer and the volume knob on my amplifier. Then, to further complicate things, I have this line-level toggle switch on the mixer's input, which lets me choose between +4 and -10.

I normally have the strip's DAW input slider up all the way to zero and my amplifier set to about a quarter, then use the 'main mix' dual sliders to set a comfortable listening volume. Or set the 'main mix' to zero (100%), then use the volume on the amplifer to taste. This seems a bit random. I'm trying to get an idea of what I should be aiming for so that when my material ends up on a person's stereo or ipod or whatever, it will be at a 'standard' sort of volume.

I'd be really grateful if anyone could give me some advice on this.

Thanks.

PS
I like to mix with a compressor/ limiter in place, so I can hear what it will sound like when the mix goes to final render but I do a lot of preliminary mixing with it on bypass. I do not use compression on the way in, but record everything raw, except for the 'Condenser' mic simulation setting on the Fostex MR-8 recorder during voice recording and afterwards a simple mastering preset to beef things up a bit, before sending it to the DAW for editing... which makes the initial recording sound great. All other effects, including final compression, is applied afterwards in the computer.
 
Last edited:
You're mixing all sorts of totally unrelated stuff here, so I'm a little confused due to time-of-day / lack-of-coffee.

1) Your operating level is either -10 or +4. If you're running BALANCED cables from a BALANCED source, it's likely +4. If it's an UNBALANCED source or an UNBALANCED input, -10. There isn't "best" in this -- There's either "correct" or "incorrect" and really nothing in between.

2) ???

3) Profit.

4) Everything seems "random" because (seemingly) you don't know your operating level, which leads me to believe you haven't properly calibrated your monitoring chain (etc., etc., etc.), which means you have no "target" from the start. Everything will continue to seem random until EVERYTHING is somewhat properly calibrated and at the very least, running at the proper I/O levels.
 
Massive Master said:
1)If you're running BALANCED cables from a BALANCED source, it's likely +4. If it's an UNBALANCED source or an UNBALANCED input, -10.
Click to expand...

Can you give me an example of 'balanced' and 'unbalanced' cables and sources? My sources are, as I said, my computer and my Fostex multi-track recorder. My cables are like instrument cables. Like you would use for a guitar. As for my inputs, the labels next to the input sockets on the mixer read: "Balanced or unbalanced".
 
'Cuz most line-level devices that use unbalanced are calibrated for -10...but if you have the headroom to push it to +4, then adjust accordingly.

I use both all the time, but you have to always be aware of where you are at level-wise and how it's affecting things in the signal chain when you mix the two...'cuz not all -10 devices will be happy getting slammed with +4 signals.
 
Doctor Varney said:
Thanks! :thumbs up:

So for what reason shouldn't unbalanced cables use +4? It sounds great, just louder.
Click to expand...

You should match operating levels and connection type as closely as possible between output and input.

If you're using an OEM type sound card the output is likely -10, so match that on your mixer's input.

Balancing helps with long runs, which isn't an issue for you, but if you've got balanced outputs and inputs it's often best to keep them balanced. Going between the two without care can lead to noise (hum & buzz) problems. OEM type sound cards' line outputs aren't balanced. If you don't get hum or buzz with your current wiring you're good to go, but if you do you'll need to change the cabling, maybe custom wire an special cable.
 
I've got no humming or buzzing but I suppose I should use the -10 instead. Thanks for the info. I use XLR cables only on my mic and output from mixer to amplifier but I had have a special pair of connectors made up, using jack sockets, to convert to the phono inputs of the amp. Sounds fine.
 
Last edited:
Massive Master said:
4) Everything seems "random" because (seemingly) you don't know your operating level, which leads me to believe you haven't properly calibrated your monitoring chain (etc., etc., etc.), which means you have no "target" from the start. Everything will continue to seem random until EVERYTHING is somewhat properly calibrated and at the very least, running at the proper I/O levels.
Click to expand...

How'd you do that then?
 
Doctor Varney said:
The stereo line level inputs of my (physical) mixer each feature a level switch, which toggles between +4 and -10.
..snip

My question is, which of these level options on my mixer gives the truest account of the level coming from the DAW?
Click to expand...
..snip

I would guess that you should get this info (and much more!) from the manual for your mixer, and also the sound card.
Barring that, set the mixer's pair of faders, for your inputs and 'trim/gain if it has it (and master fader while we're at it) all to 'zero/'unity -no gain added , no eq etc, set your DAW output to 'zero/unity.
Play something, note the level on the DAW master meter, now compare that to you mixer track (and/or master) meter in both modes ('-10' or '+4'? Which aligns better with what's coming out of the DAW?

Thus the volume hitting my ears is dependent on a mixture of sliders on my mixer and the volume knob on my amplifier.
Click to expand...
..snip
This -your listening level- is a completely different thing than your final mix level.

I normally have the strip's DAW input slider up all the way to zero and my amplifier set to about a quarter, then use the 'main mix' dual sliders to set a comfortable listening volume. Or set the 'main mix' to zero (100%), then use the volume on the amplifer to taste. This seems a bit random. I'm trying to get an idea of what I should be aiming for so that when my material ends up on a person's stereo or ipod or whatever, it will be at a 'standard' sort of volume.
Click to expand...

Repeat and rinse ..
Your listening level- is a completely different thing than the final mix levels ;)

- If you don't have a spare output control for playback use your amp volume control is one way (or the mixer master- just not if/when you're capturing a mix!

- Put some ref tracks into you DAW -on their own clean dedicated track, also at 'unity to compare can help. Note since they're already 'mastered'- a) they'll likely be loud(er) already,
b) for comparisons like this you don't want ref' playback playing through any effects/volumes etc that you might have on for your mix.

Rather than worrying about 'clearing and zeroing my DAW's master for this every time
-and having to put it all back for the mix, I made a dedicated 'mix/ref playback bus'.
It goes to the master out too, but is always 'clean and zero'd, and only gets my internally bounced finished mixes or ref tracks. (it also becomes the source for my final mix exports FWIW. If there's mutiple version, mutiple songs in a project they all get a one-stop clean clear equal path.
 
First off, I'm a bit confused by something earlier. If you plug an unbalanced (-10) source into a destination set to expect balanced (+4) it should sound quieter, not louder. Am I missing or misunderstanding something about what you're saying?

Second, a bit more detail about balanced vs. unbalanced. The difference is nothing to do with audio quality. Instead it's all about the ability to resist outside interference. Unbalanced is a simple electrical circuit that relies on the cable screen (or shield in America) to protect it from outside electrical interference (or not in the case of simple RCA cables that use unscreened bell wire). This protection is only so-so in the face of any strong interference.

Balanced is more scientific. As has been said, it's a 3 wire system. The audio signal is fed twice, once with the audio waves starting out in a positive going direction and the other starting in a negative-going direction. (By the way, I personally dislike the commonly used description of hot and cold since both wires are effectively "hot", just 180 degrees out of phase with each other.) If you think about your physics, if you combine two identical signals that are 180 degrees out of phase, they'll cancel each other out leaving no sound at all. The wires carrying these two signals are twisted together in the audio cable so that any outside interference appears equally--and in phase--in both legs of the circuit. Then comes the clever part. At the destination end, one leg is phase inverted to match the other. This means the audio is in phase again and just gets a bit louder but any noise is now out of phase and cancels itself out.

This is why, where there's no interference either system is fine--but as soon as you have a source for hum or clicks, balanced is hugely better. The longer your runs, the more at risk you are--and things like thermostats or dimmers can suddenly start causing interference after years of clean audio.

Finally, just to emphasise something others have said: your mix level and your monitor level are two separate things and need to be treated as such.

Your mix needs to be something far enough under clipping so as to not let transients go to high (and, if you apply any mastering later, to leave space to do this). We always say "go by the sound not the numbers" but, in terms of level, you have to be aware of what's happening on your meters and keep something like 6dB of space between your peaks and zero level during the mix process. You can bring it up higher once the mix is finished before you burn a CD or whatever.

Your monitoring is simply what you are listening to and should be controlled by the volume knob on your amp (and/or any monitoring control you happen to have on your audio interface). Outside your DAW you can set this however you like without affecting the mix going on inside your DAW. There are actually suggested numbers to use for SPL (Sound Pressure Level--i.e. volume) when mixing to avoid listener fatigue and hearing damage. However, without the gear to measure this, giving numbers is a bit pointless. Suffice to say, keep things at a comfortable, not overly loud level.
 
Bobbsy said:
... There are actually suggested numbers to use for SPL (Sound Pressure Level--i.e. volume) when mixing to avoid listener fatigue and hearing damage. However, without the gear to measure this, giving numbers is a bit pointless. Suffice to say, keep things at a comfortable, not overly loud level.
Click to expand...
And to encourage listening/mixing at rather very low levels at times as well
 
Yup. I guess one thing to mention is that an effect of listener fatigue is that levels that start of sounding just right will start to be perceived as progressively quieter as your ears are de-sensitised. When you start to notice that, best to take a coffee break in a quiet spot rather than cranking the volume up.

As an aside, when I design the sound for a stage musical it's awfully easy for the show to suffer "fader creep" as the run goes on. Mixing by ear alone, what was perceived as "just right" to start with starts to sound quiet and boring as you get used to it over a period of days or weeks. I always make detailed notes of volume settings on the first night and use these to check myself rather than deafening the audience who comes in two weeks into the run.
 
I love the cool cure for mix creep' someone quoted instead of 'can we have more such and such?
'What's too loud?
 
One of the hardest things to learn is that sometimes if one singer is too quiet the best solution is to turn everyone else down rather than turning that one singer up.

OT aside, this discussion reminds me of one of my earliest days working in TV. At the first station where I worked most technicians (me included) did a bit of everything and one of the jobs was setting up the early colour cameras before each studio use. To adjust the colour there were separate gain controls for the video output of each of the three camera tubes, red, blue and green. Just like audio above, it was all to easy just to turn everything up to match but, after a few days of this the gain controls would be all set near maximum.

The crust old maintenance engineer taught me that often it's best to turn two colours down rather than one colour up. This advice has stood me in good stead on matters video and audio for about 40 years now!
 
mixsit said:
And to encourage listening/mixing at rather very low levels at times as well
Click to expand...

I think 83db at your listening position is ideal according to the fletcher munson curve, but I cannot mix at that level constantly, I would maybe spend 5% of my mixtime at that level just to check frequency balance now and again, 95% of the time I mix at a lower volume than 83db.
 
A lot depends on the music how loud it can actually be, and still feel comfortable. This is part of what all the "loudness wars" stuff is about. When the music is slamming at flat-line levels....you ain't gonna be able to listen to that even at 80 dB SPL for very long. If the music has better dynamics, the ears are not subjected to a steady SPL without any rest, so you can go a little higher for the punch, and not feel the fatigue as quickly.

I kinda like to *mix* at 85 dB SPL...but when just tracking/editing, I keep it around 75, and then, maybe turn it up occasionally if needed. Too low for mixing has issues same as too loud...IMO. The balance of the elements and their frequencies changes if you go too low from the target level of the particular music you are mixing.
I just can't mix a punchy, up-tempo Rock tune at very low levels....it sounds weird... :D ...but I like to turn the mix down way-low occasionally too, just to see how that balance is affected.
 
miroslav said:
A lot depends on the music how loud it can actually be, and still feel comfortable. This is part of what all the "loudness wars" stuff is about. When the music is slamming at flat-line levels....you ain't gonna be able to listen to that even at 80 dB SPL for very long. If the music has better dynamics, the ears are not subjected to a steady SPL without any rest, so you can go a little higher for the punch, and not feel the fatigue as quickly.

I kinda like to *mix* at 85 dB SPL...but when just tracking/editing, I keep it around 75, and then, maybe turn it up occasionally if needed. Too low for mixing has issues same as too loud...IMO. The balance of the elements and their frequencies changes if you go too low from the target level of the particular music you are mixing.
I just can't mix a punchy, up-tempo Rock tune at very low levels....it sounds weird... :D ...but I like to turn the mix down way-low occasionally too, just to see how that balance is affected.
Click to expand...
I agree, as I mentioned somewhere in the mid 80's gives the flattest frequency response according to the Fletcher Munson curve but I find that too loud. The real lesson here is that you don't mix at one constant level, you adjust your level throughout the mixing process.
 
Oh absolutely....I start the mix at around 75 as I rough it out....but then when I get to the final few mix passes, it's at about 85.
The thing is to not spend 8 hours on one mix all at 85 db SPL. :D
 
You must log in or register to reply here.
Back
Top