Appropriate Levels for Broadcast TV

[Rwave]

Hi all,

I'm curious to know the broadcast standard that most engineers adhere to when mixing for TV.

I'm a younger engineer based at a Chicago ad agency, and I've always been told during my days as an assistant that when mixing for broadcast TV, your levels should not exceed -10dBFS and dialog should be mixed to an average of 0db on your VU meters. Lately I've received mixes and splits from studios on the west coast that peak at about -7dBFS with dialog mixed to about +1-2 on the VU's (many SFX had my meters maxed at a constant +3).

In this particular session I had to record a new AVO, remix, and ship out the new version almost immediately. I choose to knock the levels down to what I usually do (-10dBFS), although previous versions had shipped with mixes that were, obviously, somewhat more hot. For engineers that have been in the industry for quite some time - do you come across this a lot? Was I right to drop the level or should I have mixed to the same level as previous versions?

Thanks in advance for an insight and guidance - it's much appreciated.

[philper]

Hi all,

I'm curious to know the broadcast standard that most engineers adhere to when mixing for TV.

I'm a younger engineer based at a Chicago ad agency, and I've always been told during my days as an assistant that when mixing for broadcast TV, your levels should not exceed -10dBFS and dialog should be mixed to an average of 0db on your VU meters. Lately I've received mixes and splits from studios on the west coast that peak at about -7dBFS with dialog mixed to about +1-2 on the VU's (many SFX had my meters maxed at a constant +3).

In this particular session I had to record a new AVO, remix, and ship out the new version almost immediately. I choose to knock the levels down to what I usually do (-10dBFS), although previous versions had shipped with mixes that were, obviously, somewhat more hot. For engineers that have been in the industry for quite some time - do you come across this a lot? Was I right to drop the level or should I have mixed to the same level as previous versions?

Thanks in advance for an insight and guidance - it's much appreciated.

There have been a lot of discussions on this topic here. I would suggest a search under such words as "dialnorm", "peak levels". "broadcast specs", "Dolby LM-100" etc. and also suggest you look at Georgia's stickies at the top of the Gearslutz Post-Prod Sound forum (http://www.gearslutz.com/board/post-production-forum/) as well as Dolby's site. There is a lot of info to digest on this subject. Almost everyone we deliver mixes to nowadays has level specs of some kind (different from each other), you need to get them and understand them.

Philip Perkins

[spigots]

Are you mixing shows or spots? Your post implies spots. In my experience, it's the wild wild west out there, still, regarding spots.

We are leaning toward what you got, with tight average levels (-8 dBV/+2 dBU) favoring slightly less low end and more compression than I would do for any anything else. It seems to compete better. No science here, just years of listening.

[admix]

He said "Ad Agency", so I'll assume that spigots is right and you're talking about spot work.

Sadly, It is the wild west out there. As long as you send out something that sits happily in that -6/-7/-8 neighborhood you will not have a mix bounced by anyone. High VU levels will not get a mix bounced as long as you sit in that dBFS range.

If my VU meters continued past +3 to form a complete circle, I'm certain that some spot mixes I've listened to would complete the circle and possibly hit +3 again. Ignoring VU may or may not yield a better/louder mix on air. Your mileage may vary.

Further, in ad-mixing-land, there is fairly heavy reliance on stereo compressors/limiters, so the splits you receive from "name brand mix facility" are most probably in need of some limiting assistance. There's constant debate about whether the splits added together should yield the mix, without need of additional compression, but I've seen them arrive both ways, even from different engineers at the same studio. Some argue that the uncompressed splits are easier to deal with while doing revisions. Less problematic when matching things in. The uncompressed splits are fairly easy to spot. Lower your monitors before proceeding. The L1/DBMax and your ears, are your friends here.

I think a general rule of thumb for spot work revisions is that, somebody somewhere, and possibly many people, approved what you're revising. Levels and all. I'd leave it alone unless it was technically wrong and in danger of being rejected by the dub house/network.

Good luck!
-Steve