Hi All, I'm a long time lurker, first time poster, so thanks for all the great advice I've absorbed from you all already.

I'm designing / mixing some trailers specifically for the internet, and was wondering if anyone had any words of wisdom.

Mainly, I can't decide whether it's best to go down the advert / music route of compressing everything to hell to make sure my trailers stand out against everything else that's out there, or to actually embrace some dynamic range and create a better sounding stand-alone mix (whilst knowing full well if someone doesn't adjust their volume, it's going to sound rubbish compared to the metal they're listening to on youtube.)

I'm guessing there's no point in mixing to -10dBFS or such, but other than that I'm undecided on a lot of things. Personally, I'm in the habit of adjusting my listening level for everything I hear on the net, but does the average person do that too? I've no idea.

Any wise words?

Thanks all.



My room is setup so that I can switch between two listening levels for mixing. I do this through the multiple trim settings available in the 192 I/O, but you can possibly set something up in your monitoring chain if you don't have them.

Basically, it boils down to this:

One setting is -20dBFS pink noise is set 78dBSPL for broadcast level work.

The other is -14dBFS pink noise is set to 78dBSPL for web work.

[With the 192...the A-trim is electrically calibrated as -20dbFS=1.228 volts RMS, and the B trim is calibrated as -14dBFS=1.228 volts RMS. This way, I don't have to touch a monitor controller when I switch reference levels. The speakers receive an appropriate level regardless of which output trim I've toggled to.]

A consistent listening level is important. This way, I don't have to change the way I work (other than to keep in mind that I have less headroom), but I will naturally mix hotter for the web based stuff.

I typically limit at -3dB FS for web, as opposed to the -10dB for broadcast.

  • @Shaun Farley, great detailed answer that helps me greatly. +1. – Jay Jennings Oct 7 '11 at 17:50
  • +1 This sounds similar to what I do... -20 for theatrical, then I bump up the stems by 6dB and babysit it under a brickwall limiter like Maxim or L2007 as needed for the web/DVD version. – Stavrosound Oct 7 '11 at 18:11
  • Thanks @Shaun Farley, that's really interesting. :) So do you approach your mix exactly the same way but print it hotter? For example, if you mixed the same thing for broadcast and then for web, you could reduce your web mix by 7dB and it'd sound pretty much the same as (and have a similar dynamic range to) your broadcast mix? Also, excuse my ignorance here, but why limit to -3dBFS rather than 0dBFS? Is this 3dB headroom to catch any spikes that were too fast for the limiter to respond to? Thanks for your help. – Sam Oct 9 '11 at 22:17
  • @Sam - It would be pretty close. The web version would probably have less dynamic range, because I usually use a little bit of gentle compression across the master bus. I use -3dBFS, because not all processors (limiters included) are designed to account for intersample peaking. From the tests I've done with an LM100, I know that the Waves L1 actually is a true "brick-wall" limiter. I still leave the extra 3 dB just in case, because you never know exactly what is going to happen when it goes through final encoding with the video. – Shaun Farley Oct 9 '11 at 23:53
  • @ Shaun Farley, -3DB is a great rule of thumb to account for possible mono crashdown by the playback devices, since it's bound to happen sometimes. Although I wonder if just shy of 3dB might be better since -3dB summed is exactly zero? Just brainstorming here. – Stavrosound Oct 10 '11 at 5:06

I also set my peak to -3dBFS with the massey L2007 even though most web junk clips 0.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.