Share some tips and tricks for mixing TV!

[RadioFilter]

I've wondered about this exact thing: sidechaining an EQ gain knob to my dialogue bus. When the VO begins, a little bit of lower mids are taken out of the music. Is this possible? It may not work very well, but I would love to try it out.

Minister, you mentioned the Dynamics III with an EQed sidechain. I don't think that will do it. It's still compressing the entire frequency spectrum of the music, but only responding to a narrower range of frequencies in the VO track. Correct me if I'm wrong.

Is there a way of doing this?

[minister]

Minister, you mentioned the Dynamics III with an EQed sidechain. I don't think that will do it. It's still compressing the entire frequency spectrum of the music, but only responding to a narrower range of frequencies in the VO track. Correct me if I'm wrong.

you have the ability to switch to bell shape curve eq with the sidechain, and selct one or two fequncies. you could also shape it using the HPF and LPF filters...

i personally do not do this. i was just answering the question of what compressor or multiband has a sidchain. WAVEARTS does not. the DIGI 3 DYN does.

[sidereal-studios]

as was mentioned, my other hint is : ride your EQ's.

Care to elaborate on this?

-----

I just completed my first season of my first TV show (a 'reality' show w/ host & guests, non-studio, 2 camera shoots, 2 lav, 1 boom, 2 cam audio = 5 production audio tracks). As this is my first time doing a show on my own, consider it both tips and open for ridicule.

First, my template:

4 mono tracks "dia" (matching EQIII curves on all first inserts)
4 mono tracks "cam" (matching EQIII curves on all first insterts, different from "dia" of course)
2 stereo FX tracks (no insert PIs)
1 stereo "graphics" track (for info graphics that occur in the show; no insert PIs)
1 stereo "transitions" track (whooshes, lower 3rd fx, etc.; no insert PIs)
2 stereo music tracks (no insert PIs)

DIA submix (Comp III, 4:1; Oxford EQ bandpassed from 100 Hz to 12k)
CAM submix (no PIs)
FX submix (no PIs)
MUSIC submix (Oxford EQ scooped gradually to account for DIA)
Master Fader (Oxford EQ, bandpassed 60hz to 15k; EAS Finis limiter, set to broadcaster-spec'd -6db; EAS InspectorXL metering)

Template audio files have all recurring transitions, graphics, lower 3rd sounds, theme music, etc.)

Process, in order:

1. Launch PT Template
1a. Save session copy in > TV Series location ("Ep1")
1b. Close template, open Ep1 PT file (adjust disk allocation)
2. Import session data: Locate and open OMF
3. Import QT movie
4. Confirm sync, duplicate playlists, and lock regions on alternate playlist
(name locked playlists accordingly)
4. Watch show & set markers in real time
(markers are the "to-do" list: anything incl. lower 3rd graphics, transitions, music needs, noise probs, any other issues/considerations)
5. Split out OMF regions into playlists (host, guest, boom, dia4, cam, etc.)
6. Place standard sound transitions, logos, theme music, etc.
7. Edit and crossfade regions
8. Noise reduction (as needed)
9. Sound design needs on FX tracks
10. Track-based volume automation on all production tracks
12. Load music tracks and edit
12. Track-based EQ / compression tweaks
13. Final mix...
13a: Submix EQ and compression needs
13b: Submix fader riding (on Control24)
14. Bounce to disk

[minister]

Care to elaborate on this?

frequency content is not always static. heads move. mic positions move. delivery changes.

use your ears to keep it smooth.

sorry, guess i am feeling stingy today, so don't take offense. ...i figured this out for myself...so, excuse me if i just give hints.

[airon]

We're mostly dialog, so I can only second Ministers recommendation. Ride the EQs.

I use two tracks directly to a aux collector bus and four that pass through an aux compression bus first, with a soft (2:1) but fast (3ms) compressor (RenComp), slight limit(peak control) and then a slight hipass(it's TV, so no rumble). My VO tracks are always recorded in a seperate session, and then bounced out, mainly because my main session is 16 bit, so the 24 bit VO and dub recording sessions give me plenty of headroom and depth before I reign in the VO tracks. That stuff is slightly slammed (5:1) and then soft leveled by a 1:1.5, with pre and post compressor EQs. Additional mixing happens in the final mix naturally.

I also like to ride my EQs, especially for outdoor scenes, but more often than not for indoor stuff. People stand in corners, mumble on to a desk or move up close. You try to make it as real and interesting as you can in the time you have, so I sometimes grab my prearranged-by-frequency EQ gains and pull a little here, cut a lot there, and maybe boost a little over there. I'm always fuzzy inside when I've managed to deal with a lump of grumpy outdoor tracks that now match and flow.

My dialogue if necessary gets a 2:1 compression, which is on the aux compression bus. Precompressed production dialog is a big pain. To any production mixers who does this, please DO NOT do it hard. We do this better in the mix. Soft reigns are fine. We'll do the whipping.

FX and ambiences are pretty basic. Just a few EQs there, which don't get used a lot. Like I said, mostly dialogue. I collect all the stuff on aux bus tracks and hammer out loads of mixes and an 8-track.

Here in Germany we have the -9dB = 0 VU upper limit. Lots of peak limiting. I'm bound to the L1+, which isn't a whole bag of fun for the music, so I compress the louder parts 2:1 as well, which makes the fader work a lot easier and doesn't drive the L1 too much (3-5 dB GR is all it can realy take without getting icky). The station I mix for does no further compression processing, which is lucky for me. I do realise that.

As for fx, most of them are on the enviromental music tracks. Stuff in bars, restaurants and stereos.

In the end I prefer to drive my compressors as little as possible. They're a control tool and a little bit of a soundmaker. I despise overcompressed material and try to avoid smashing my mixes whenever possible. Thankfully, even on our daily 40 minute soap(saved that for the end) I have so far managed to avoid it a lot.

Recently, I have found the Doctor Who mixes interesting. Most of it is very tightly reigned in. Even though the dialog lives at a comfotably low level compared to german TV, it's realy well controlled and even, at least in the shows I watched. Some effects however do pump around a bit. The sword fight in the "Christmas Invasion" episode (Season 2 opener) is an example of less than smooth dynamic processing. Shame realy. Love all the effects they've done in the series.

[Phil Jeffers]

Get it done in the quickest amount of time, for the least amount of money, and cross your fingers that it will pass the Network's QC. This is what executive producers want most of the time. Unfortunately, there rarely is any art involved, or even any sound editing in the strictest sense of the word. Oh, and get used to thinking that ordinary rubbish (i.e. reality TV crap) is really amazing.