Q: I would like to try a summing mixer. Will my old Tascam do?
Punk Production - adding excitement to professionalism
Achieve multiband compression WITHOUT a multiband compressor
(Almost) everything you need to know to earn a living by recording music
How to edit out pops in speech or singing
Drop-In Stratocaster Pickguards Give You 35 Pickup Tones To Find Your Signature Sound
The Waves CLA-76 compressor plug-in on snare drum, with video
Q: What key should I sing in?
Q: "Why is the signal from my microphone low in level and noisy?"
"We can fix it in the mix" is an oft-quoted saying in recording. Basically if this is said at any point during the recording session it means that a problem is being stored for later. And it might not be possible to fix it. The correct response to "We can fix it in the mix" is always, "No, let's fix it now!"
The plain fact however is that the process of mixing starts with repairing faults in the original recording. Perhaps the producer and engineer in the recording session were sticklers for detail and there isn't one tiny little detail that could possibly be called a fault. That sometimes happens, and so do blue moons.
Suppose for example that in the original session, all of the tracks were recorded from beginning to end with no editing. That would now be considered a very traditional way of doing things, but it can happen. What the mix engineer will receive is all of the music, plus the singer getting his note before each verse and chorus, the background vocalists getting their notes, the hum and buzz of the guitar amplifiers, and the squeaky drum throne that makes its presence felt during the a cappella section of the song. Then there will be the count in, and the fidget noises at the end as the last chord dies away. All of this needs to be cleaned up by the mix engineer. In modern times with powerful DAW software this can be done during the recording process, but you can't expect it to be done, or perhaps not all of it. And indeed the mix engineer would want to do the trimming of the ending.
At the other end of the spectrum the producer and engineer in the original session might have done a lot of editing. Any imprecision can result in clicks at the beginnings and ends of edits, which the mix engineer would have to clean up.
There might be other issues of a more musical nature, such as consonants at the ends of words not lining up in the background vocals. Sometimes a little slackness can be tolerated, but it's a rare background vocal track where there is no opportunity for improvement. This applies even more so to double tracked lead vocals.
Where the original recording is of a band playing conventional instruments, then there is an obvious target for the mix engineer to aim for - the sound of a band. This implies that the drums must sound like a drum kit, with all the individual drums and cymbals in the kind of proportion you would expect in real life, and the other instruments and vocals blended in a similar way to a thousand other mixes of similar bands.
Working with the cleaned-up multitrack, the mix engineer will consider the individual instruments of the band and decide whether any are inadequately recorded. Suppose for instance that the bass has been recorded via DI (direct injection). This can often result in a bass that is rather thin-sounding, with significant differences in level from note to note. The mix engineer would EQ and compress the bass so that it sounded fuller, without so much unwanted variation. Similar improvements can be made to other instruments that are lacking in any way. The instruments that already sound good should not normally be altered individually but considered in the context of the whole mix. The blend among the instruments will be achieved using fader levels and EQ.
Where the multitrack recording is made from synthesized sounds, or heavily-processed conventional instruments, then there isn't any obvious target for the mix engineer to aim for, other than recordings that have already been commercially successful in a similar style.
The mix engineer will treat the vocals differently according to whether the recording is of a solo artist or of a band. In general the vocal of a solo artist should stand out front. The vocals of a rock band would be placed at a lower level among the instruments. For a solo artist, it is normally expected that all of the lyrics should be easily intelligible. For a band, then it doesn't matter so much if some of the lyrics are somewhat obscured, and indeed it gives fans of the band an opportunity for discussion as to what they are and what they mean.
The lead vocal is an area where the mix engineer can be quite creative. Listeners don't expect a lead vocal to sound like a real person singing purely acoustically. No, they want a bigger and fuller sound, with EQ, compression, delay and reverb. And often they will like a little extra warmth too, produced by a vacuum-tube microphone or preamplifier in the original session, or harmonic generation done digitally in the mix.
As well as getting a pleasing sound for the lead vocal, the mix engineer needs to get the vocal to blend with the instruments. It should sound like an intrinsic part of the overall mix, not just something that sits on top of it.
A professional mix engineer may spend an entire day mixing just one song, and then return to it with fresh ears the next morning. Someone new to mixing simply does not have the precision listening skills that are necessary to warrant spending so much time. After a certain point a mix doesn't get better, and in fact it might get worse. The more experienced you get at mixing, the more time you will want to spend, although it is important not to fall into the trap of never considering a mix finished. This is easy to do with DAW software where you can save a mix and return to it at any later time.
One way to approach mixing is to identify the section of the song that is the 'main texture'. It isn't the most heavily orchestrated section, which is probably near the end. Nor is it the vocal and acoustic guitar introduction. Somewhere close to the start of the song however will be the eight or sixteen bars that the mix engineer must get right, or nothing will really work. Looping this section so that it can be refined and honed, honed and refined, until it can't get any better, is a good way to work.
When the main texture section sounds as good as it possibly can, then it will probably be found that the balance achieved works pretty well for the rest of the song too, with fairly simple adjustments for earlier and later sections where necessary.
There's a lot more to mixing of course, but it is an active part of the production process. The mix should bring out the best of what's there in the original multitrack recording, respecting the work of the original producer and engineer in the session, and of course the musicians and singers.