Mixing
Surround
We used to mix in 5.0. I don't really see the point of mixing 5.1. Most decoders don't really know what to do with the ".1" channel of audio, and all the material from the other 5 channels gets folded to the subwoofer channel anyway.
But now we use the ITU-T standard for 5.1. We just make a ".1" channel of empty audio just so that the whining is less when we deliver. "Why don't you make a 5.1 mix?" they say. "What's the difference?" we reply.
It's not worth it to keep arguing about the ".1". So I've given up. We mix in 5.1. We just don't actually route anything to the .1 (LFE) channel.
Well, that's not true anymore. We have gotten mixes kicked back because the meters didn't dance for the QC people at some distribution outlets. So it's stupid, but we send some signal to the LFE track. Usually music, mixed about 20dB lower than the rest of the tracks.
I suspect that the ".1" was originally for optical analog mixes where you actually couldn't lay down the amount of low end you wanted to. But nowadays you can "print" any sound you want without worrying about the optical track.
OK, so anyway, we mix in ITU-T's 5.1 format. That is:
1. Left
2. Right
3. Center
4. LFE (Low Frequency Effects)
5. Left Surround
6. Right Surround
And, like I said, we used to not route anything to the LFE channel. But it's there. And the QC department at iTunes or whatever will be very upset. They'll see that track. Empty. Waiting. Yearning. You need to send some signal to it. Just make the needles dance. -20dB (or more) than the signals going to the other tracks.
Here's what the surround busses look like in Samplitude.
Here are the surround tracks brought into a new project in order to mix to stereo.
Remember that just because there's no LFE track that doesn't mean there's no low end. In fact, the low frequencies are more likely to play back accurately on a home system or even a theatrical system because you aren't using a separate LFE track.
Levels
Our maximum level on any of the 5 output tracks is -12dBfs.
For monitoring I like the 72dB SPL (flat) = -20dBfs calibration. This is a "louder" calibration than the theatrical 85dB SPL (flat) = -20dBfs scale, but I think it works better for TV and video.
Mono
By-and-large you want to keep all your dialog mono and going straight up to the center channel. You don't really want your dialog flopping back-and-forth between speakers. At least I don't.
There's a reason your production dialog is almost all center and up-front. As you change your shot you don't want the dialog to go flying around the room. Keep all that dialog up front.
Will you just look at all those mono sound effects tracks?
Monophonic sound is your friend. There a lot of sound effects which just need to be mixed up center — just like your dialog. If your dialog is all center and some sound effect (a door closing for instance) is suddenly in stereo it "takes you out" of the scene just a bit (especially if you're listening on headphones.)
Is there an exception to this rule? Yes. Distributors hate the idea that you've mixed in mono. So do make sure you have some stereo beds — or even surround beds — in the soundtrack.