0dBFS - really?

[samplaire]

It’s theoretical but also a practical question,well, a set of questions.

I mix 2 tracks and sometimes the sum (on a master track/fader) shows the signal goes beyond 0dBFS. Ok. So when you bounce the material the resulting file should show 0dBFS (when imported back to PT or any other audio editor or DAW) because there is nothing above 0dBFS - it’s a digital domain. Right?

Ok, to avoid going beyond 0 I put, say, L2 or L3 without any threshold or celing slider movement. And the result seems to be achieved - no red leds on the master meter. Right?

Of course there is this ‚between’ state which can be missed during the sampling process so the peak can go beyond 0. A little bit, right?

Then I put, say, WLM plus which features True Peak limiter. When I enable it right after the L2/L3 it tames the signal sometimes by 3-4dB! How come?

The same is when I apply the WLM To limiter on an imported wav from my first example. How come?

[JFreak]

Exceeding 0 dBFS on a fixed point mixer results in nasty and very audible distortion. Exceeding 0 dBFS on a floating point mixer just starts smearing the sounf´d, which you would first notice on lack of detail on reverb tails.

[samplaire]

I remember the time when recording to DAT and I know the glitches ;-). Thanks for your input but this doesn’t answer my question, however

[JFreak]

You may be talking about intersample peaking, which is why many albums are mastered to -1 dBFS and radio edits to -2 dBFS.

[samplaire]

Yes! I meant it! But I thought it can be something max 1dB. I also master radio commercials to -1. But considering my observations with the 4-5dB it seems this is something else…

[JFreak]

Loudness wars started in 1940's, ever since everyone has tried to fit more than possible into the medium in use.

Bass heavy stuff can have +6dB intersample peaks. Mastering to radio, tv, cd, streaming, etc. has become art in itself.

Just be on the safe side and adjust your limiter output to minus enough.

[BScout]

Intersample peaks can exceed 1dB.
But more likely your issue is metering ballistics.
ProTools’ meters aren’t bit measurements. (Though there are a lot of options in preferences for those meters)

[Darryl Ramm]

Originally Posted by samplaire

Then I put, say, WLM plus which features True Peak limiter. When I enable it right after the L2/L3 it tames the signal sometimes by 3-4dB! How come?

The same is when I apply the WLM To limiter on an imported wav from my first example. How come?

First question would be what do you/the plugin have all the “True Peak Max” etc. set to? Then have you used the trim button to adjust the make up/down gain? Or just manually adjust the gain?

[Sardi]

TP limiting is not allowing any peaks to go over 0dbFS. True peaks can indeed be as much as 3-4dB depending on the source. This is not uncommon.

Also, why are you bouncing out radio promos at -1dbFS? This is going to be turned down immediately to whatever playback level the station is set to, these days probably -23LUFS. To get anywhere near that ballpark, you’d be down near the -10 to -12dbFS peak readings, depending on how squashed the mix is. Have you been asked to deliver this hot?


Sent from my iPhone using Tapatalk

[samplaire]

Thanks for your answers! My questions are rather provocative, they are not moaning, I’m just interested

-23LUFS is for TV here in Poland, radio stations take 0dBFS spots. I’ve been working at a radio station and most of them have a multitool called Orban at the end so everything is pressed to the same level (compression, eq, stereo image) . This way Presley and the newest EDM hits have the same level. I even remember I had to master jingles (sung ones) the way music was 3db lower than vocals not to have the vocals crushed. Orban settings are very aggressive. I once tried to play a song with a fade at the end to the end and it pulled the level up like it was still at its loudest. I hope you get what I mean ;-)