Lufs

sbrady

New member
I hate this thing, my work says make LUFS at -24, Steinberg says -23. I am mixing a 30 minute TV show. If I leave the timeline run, after 10 seconds the meter does not change. What I do is reset the LUFS meter then play the HOST section for 10 sec, reset the meter, them play the story, reset the meter then play the commercial. Adjusting the EQ and level for each section. Seems to get an accurate reading if I stop and start. If I just let it run, after 10 seconds, the meter really does not change.

Can anyone help me master this. It used to me watch the Dorroughs, hit the red chicklet. Thanks.
 
You need to explain a little bit about how you are measuring the LUFS information. If you are stopping and resetting the measurements, you are not making a true integrated LUFS measurement.

There are actually several LUFS settings.

Integrated LUFS measures the cumulative volume to make the calculations. If you have a 30 minutes show, it needs to measure the whole 30 minutes.

Short Term LUFS measures in smaller increments, typically a 3 second window. You can see the ST LUFS rating change over time, especially if there extensive areas of low and high volume.

There is also momentary LUFS, which will constantly change as it moves through the piece, as it only integrates over 0.4 second increments.

TV broadcast standards are typically -24dB LUFS as maximum. Program material that exceeds that can cause a violation for the station, so that is tightly adhered to. You CAN be lower than -24dB without issue. It just won't seem to be as loud as other programs. Other broadcast standards usually specify -23 +/-1 dB, so the difference between -23 and -24dB is irrelevant.

Streaming services vary but many are around =14dB. CDs are typically around -6dB but aren't really bound by any rules except a max of <0dBFS. In all cases you want any true peaks to be at least 1 or 2 below 0dB. If you have a project that measures -24 but has peaks of over 0dB, you will need to use a peak limiter.
 
It's probably showing an integrated LUFS number, so the longer you run it the less it will change (unless there's some extended section that's substantially higher or lower). I use the Youlean LUFS meter that shows a lot of information.
 
You mention measuring commercials, which is really one of the primary reasons LUFS became a standard. You would have program material with a particular RMS measurement with a few peaks approaching full scale, but with a wide dynamic range. Then a commercial would come on that was very highly compressed but still under 0dBFS. The commercial sounded WAY louder than the rest of the program, causing people to complain.

In that case, you should measure the commercials separately from the main program material.

Think of it like traffic flow. Maximum speed is 100mph. !000 cars go by at an average speed of 70mph. A 50 car funeral procession goes by at 35mph. Then a gaggle of 50 hot rodders go flying by at 99mph. If you are trying to study traffic patterns, the average doesn't tell you the whole story. Short Term measurements will tell you potential dangerous situations, like the speed differential between the funeral and the hot rodders.
 
You need to explain a little bit about how you are measuring the LUFS information. If you are stopping and resetting the measurements, you are not making a true integrated LUFS measurement.

There are actually several LUFS settings.

Integrated LUFS measures the cumulative volume to make the calculations. If you have a 30 minutes show, it needs to measure the whole 30 minutes.

Short Term LUFS measures in smaller increments, typically a 3 second window. You can see the ST LUFS rating change over time, especially if there extensive areas of low and high volume.

There is also momentary LUFS, which will constantly change as it moves through the piece, as it only integrates over 0.4 second increments.

TV broadcast standards are typically -24dB LUFS as maximum. Program material that exceeds that can cause a violation for the station, so that is tightly adhered to. You CAN be lower than -24dB without issue. It just won't seem to be as loud as other programs. Other broadcast standards usually specify -23 +/-1 dB, so the difference between -23 and -24dB is irrelevant.

Streaming services vary but many are around =14dB. CDs are typically around -6dB but aren't really bound by any rules except a max of <0dBFS. In all cases you want any true peaks to be at least 1 or 2 below 0dB. If you have a project that measures -24 but has peaks of over 0dB, you will need to use a peak limiter.

I've been lazy and just 'mastering' my mixes to be <0.2dBFS. So I got the Orban meter, and my latest mix hits as high as -10.1 LUFS during the final outro, -12.1 during choruses before that. I suspect my music submitted before to Spotify etc have been similar.
So I should be reducing overall volume on my streaming submissions? Are they compression-crushing my songs, or just turning the volume down?
 
I've been lazy and just 'mastering' my mixes to be <0.2dBFS. So I got the Orban meter, and my latest mix hits as high as -10.1 LUFS during the final outro, -12.1 during choruses before that. I suspect my music submitted before to Spotify etc have been similar.
So I should be reducing overall volume on my streaming submissions? Are they compression-crushing my songs, or just turning the volume down?
If you are submitting a song with integrated LUFS over their target, they will just turn the gain down, so you're giving up dynamic range (potentially) with that kind of CD-level LUFS. If you submit something well below their target (I think they're -14, too), they'll gain/limit it up, is my understanding.
 
I've been lazy and just 'mastering' my mixes to be <0.2dBFS. So I got the Orban meter, and my latest mix hits as high as -10.1 LUFS during the final outro, -12.1 during choruses before that. I suspect my music submitted before to Spotify etc have been similar.
So I should be reducing overall volume on my streaming submissions? Are they compression-crushing my songs, or just turning the volume down?

You might want to read the Mastering and Loudness FAQ page on Spotify.
 
Since they are trying to target a -14 LUFS, if you submit a -20 song, but it has peaks of -1dBfs because it is really dynamic, then they boost the thing by 6dB, those peaks are going to shoot through the roof!

I found this to be interesting:

When a user plays your album, we normalize the loudness level of that album at the same time. The entire album will play back at -14 dB LUFS from start to finish, and the gain compensation applied by Spotify won’t change between tracks. This means the softer tracks will be just as soft as you intend them to be.

However, if the user plays your album in shuffle, or a track from it in between tracks from other albums (such as in a playlist), we can’t apply album normalization so track level adjustments are used instead.



That means a song could sound a lot different when played individually vs as part of an album.
 
Spotify
-1.0 dBTP
-13 to -15 LUFS
>9DR


Youtube
-1.0 dBTP
-13 to -15 LUFS


Soundcloud
-1.0 dBTP
-8 to -13 LUFS
>9DR


Amazon Music
-2.0dBTP
-9 to -13 LUFS
>9DR
16 BIT/44.1KHZ


Leaving up to a full decibel (dB) of peak-headroom on your master WAV files can help prevent clipping and overs when the WAV masters are reduced to a data-compressed format.
 
video production is another set of target values,
and i know nothing john snow about any of that.

i always thought the target was -23 lufs, for maximum headroom for really loud soundtrack effects.
 
Back
Top