I have been reading about the K-System, which I previously thought I had a handle on. You use VU Meters or slow averaging meters and pretend that the 0 level is some amount below the top... align them more or less and use that as a starting place for your mix... mixing with plenty of head room.
 
http://www.aes.org/technical/documentDownloads.cfm?docID=65
 
After reading this article, it seems that setting the volume in the room is an important component to the system. He suggests setting the left and right speaker volumes separately to ensure you are at the right volume in both.
 
I don't normally do this. In fact, I have usually intentionally auditioned tracks at widely varied volumes to see what it will sound like when volume is dialed below conversation level on the highway. Recently, I went back to a mix I thought was finished because I heard it at low volume on the way to lunch with a co-worker and I realized it could be improved for that environment without costing the main listening environment much.
 
Now, I'm wondering if that isn't just another form of the LOUDNESS WAR. Because I was adjusting levels for very low volume playback, was flattening some of the instruments in the mix (not compression in this case-volume automation).
 
I'm curious what you guys have to say.

I try to mix at 83dBSPL at my listening position.
 
It is a standard suggested in an AES paper back in 1973 and it seems like it works well.
 
If you are working on content that is meant to seem loud in it's distribution format then you turn down your playback (after the DtoA conversion) system so it's 83dBSPL when you are mixing.
 
If you are working on content that is meant to seem less loud in it's distribution format then you turn up your playback (after the DtoA conversion) system so it's 83dBSPL when you are mixing.
 
That way you become very accustomed to the sound of your system and room and you can build experience and more easily predict how it will sound for the target audience on the gear that the audience routinely uses for their playback enjoyment.
 
If you make that practice a routine then the occasional check at quieter and louder levels becomes even more informative.
 
If you mix much louder than 83dBSPL you will fatigue very quickly... if you mix below 77dBSPL you will have difficulty making stuff seem loud without blowing out the character... you mix up a characterture of loud (think of the "loudness" button on your home stereo)... and it sounds less than ideal on an actual loud playback system.
 
Years ago film sound people, who have a professional interest in delivering a consistent experience at theaters, suggested that mixing some where between 77dBSPL and 83dBSPL yielded results that translated well to the variety of play back systems installed in theaters.
 
The "K" system is a modern protocol that formalizes those basic and universal ideas and quantifies some of the details.
 
 
best regards,
mike
 
 
 
 

Without going into the K system or SPL's or meters.....
 
Yes.
 
I have my system monitors set to the 50% detente mark on the volume controls and the interface control panel input and output levels set and I don't adjust them very often. (not in years) The levels I work at are comfortable to me setting at the desk and they pretty much remain the same from project to project. 
 
I can, and do change the volume as I work choosing to track and mix at lower levels quite often. I group the sound card master faders and pull them down from 0db for lower volume work.

I set each speaker for 83 dB SPL and aim for 85 dB in the mix position.(Note: do not use a tone to do this. The best signal is band limited pink noise (500Hz to 2.5KHz I think, check the Katz website) The band limited pink noise eliminates errors due to sine waves bouncing around and giving the measurement mic the wrong info. Also the band limited pink noise also eliminates low freq errors as well. ALSO you have to compensate the level of the bandwidth limited pink noise back up to match full range pink noise. But if in doubt use full range pink noise at the correct K ref level. Katz has a pink noise sample on his website at -20 dB FS. Add gain to create the -14 and 12 versions.
 
You also really need a permanent SPL meter set up to keep you honest. As you mix you tend to turn it up slowly over time. The SPL meter keeps you in check. (eg around 85 dB SPL)
 
I tend to mix for longish periods at 85 dB SPL and it seems fine to me. (It IS a lovely level) I also check down very low in volume on a small (mono, important!) speaker with a summed L+R mix. I find this reveals a lot in terms of what I call the critical balance. I find that sometimes things can be a little out of whack on the small speaker at low volume after they seem right on the main speakers. However after making some small adjustments I find the small speaker can be satisfied and when I go back to the main speakers things have changed very little.
 
I also like to listen to 100 to 105 dB on the main speakers for some short bursts. Vital! You will never know how your mix sounds loud until you do. (End of story) I find up there the bass and the reverbs can stand out a little too much and after some fine tuning of those things the 85 dB volume still sounds OK with them too.
 
I find I can satisfy the small speaker at low volume, the mains at 85 dB and 105 dB. The mix is usually perfect at that point and translates perfectly onto anything I play it on.
 
K system is about internal levels within your DAW and keeping rms levels constant everywhere. (tracks, Buses and main stereo mix) It is also about monitoring levels and that is what we have been talking here.

I have the monitor amp about 1/2 way in my small room.  I usually mix at moderate levels, but have the TC Konnek 48 remote right at hand.  It is easy to pump up the volume to check it, or bring it down to see if the vocals/leads are still clear.  It is really nice to check at different volumes to see how a mix holds up.  And keep my interest, too, after focusing at one level on one part of the song for a good long while.  breaks things up.
 
@

83-85dBSPL is uncomfortably loud for me. I'm normally running around 77dBSPL. I am convinced that monitoring at consistent settings is more important than what those actual settings might be.
 
The biggest danger is that if your chosen level is extremely low you may start out getting overly bass-heavy mixes. However, it's entirely feasible to train yourself to mix at any volume with practice, as long as you keep it consistent. Many professional mixers work at low volume, because a) it helps preserve their precious hearing and b) it allows them to concentrate on the more-important midrange frequencies.

I find the small speaker down at low volume is really emphasising the mid range as Bit is referring to. My little speaker even has a tiny tweeter fitted but I have put a volume control in series with the tweeter so I can even turn it down/off and loose the highs if I need to leaving just the mids behind.
 
I like the idea of my whole mix being funneled into something so small or a bottleneck like that. I spend much more time on it than the mains in reality. I notice that when things are sounding very good and well balanced on that, the mains usually sound killer and there is not much more to do on them.
 
This obviously goes back to the Auratone days when many engineers used them as a reference. You can get the bass right too even on that, it is just a matter of learning what correct bass levels are and how they sound on it. It is also perfect for setting correct vocal levels for a song too. It tends to make you keep the vocals slightly down and in the music more but not buried either. Level wise the small speaker for me would be 70 dB SPL or less.
 
I sit close to it with it directly in front of me. The mono mix thing is also good for checking separation between parts and various things, mono compatibility and if big wide synth sounds are still there in mono! Reverbs are interesting on it too. If you can hear the reverb in the small speaker you have got too much reverb. When the reverb levels are just right they are just nice on the mains and barely audible on the small speaker.

Thanks everyone. This is way helpful!

This will sound sacrilegious to serious HiFi to folks, but until I'm seriously mixing I pretty much use Apple earbuds at moderate to low volume while I'm recording and doing some rough mixing.

As I get further along, I then start introducing reference songs and start using the Focusrite VRM Box while beginning to listen with more volume. I also have better headphones and use those when I feel the need.

Low volume for me works. I'm hoping to still have good ears left when I'm older to hear the birds singing outside, and of course to hear al the music I'm putting all this time into recording.

bitflipper
83-85dBSPL is uncomfortably loud for me. I'm normally running around 77dBSPL. I am convinced that monitoring at consistent settings is more important than what those actual settings might be.
 
The biggest danger is that if your chosen level is extremely low you may start out getting overly bass-heavy mixes. However, it's entirely feasible to train yourself to mix at any volume with practice, as long as you keep it consistent. Many professional mixers work at low volume, because a) it helps preserve their precious hearing and b) it allows them to concentrate on the more-important midrange frequencies.

Glad I'm not the only one that feels this way. I'm anywhere from 69-75dBSPL on my end. I'd never be able to work from 6 pm until 9-10 am the next day if I worked any louder. LOL! Not only that, but it's just uncomfortable for me to be louder than that for more than a few minutes.
 
I do change my volumes and go lower, but always stay around the 70's. I'll blast up for a bit to see how something sounds loud and then I'm back down again. The key thing to remember G....is you have to be at a certain volume to hear the lows and mids correctly. The other side of the coin is, you can actually hear things at a lower volume that you can't hear louder. For example, if you can hear ALL the instruments in a song as the song fades away...and those instruments are still audible, 9 out of 10 times you have a good mix going on.
 
I know about the K-system and have tried it but am uninterested in it. These days I record at -6 dB peak going in, mix out at -3 dB peak coming out. I turn my volume up, I turn my volume down, I check it on 2 other sets of monitors, I shut down a monitor and enable mono interleave and test the mix, and I'm done if it passes that test. Be careful with mono mixes though because anything you spread half decent will be a grey area. There are times where I have thought guitars were too low in mono, so I turned them up and in stereo they were killing loud. Th best thing for me about mixing in mono is how you hear kick, bass and whether or not you have enough lead vocal in the mix. Anything that is double tracked and spread will behave a bit strange, so you'll need to make your final decision on instruments like that while in stereo in *my* opinion.
 
As far as everything else goes though, the less I have to worry about that doesn't improve what I do by leaps and bounds, the better for me because I have less things to mess with and worry about. You can get so wrapped up with this stuff that you wind up not getting as much done as you'd hope. That's a real drag if you work a day job or have a family to tend to and only have limited time to enjoy this stuff.
 
-Danny