Today's LUFS Guide ?

Discussion in 'Mastering' started by pcrecord, Oct 27, 2015.

Tags:
  1. pcrecord

    pcrecord Don't you want the best recording like I do ? Well-Known Member

    Hi gang,

    I've been reading a lot about mastering for different distribution method and I'm just not finding a clear answer. What LUFS target should we use for different media ?

    Here's 2 I read :
    -13 LUFS for youtube
    -8 LUFS for CD
    ? Vevo
    ? Websites
    ? Radio
    ...


    Please correct me and let's put up a list unless someone has a reference already !
    Thanks ;)
     
  2. DonnyThompson

    DonnyThompson Distinguished Member

    I think -8 is pretty hot, pal. But, there's no real "standard" for CD's, either (obviously there's not a standard because there are albums out there on CD with DR levels of -3db).
    I think it's really about what sounds good to you. There are commercial Metal, Hip Hop and EDM albums out there that sit at around -10, and Jazz albums that are closer to -16.

    When Tom Bethel mastered the Love CD for me, I think it turned out to be about around -12. ( @Thomas W. Bethel). But, I had specifically asked him to do his best to give us a nice open dynamic range ( of which he had no objections to whatsoever).
    When I sent the original final mixes to him, the LUFS was around -23.

    The last time I checked, YouTube and iTunes were both starting to lower to -16 if anything hotter than that was uploaded.

    The LUFS criteria are really more for broadcast. As far as I know, the current R128 ( EBU) Standard for radio and television broadcast is -23.

    Here's a link to a white paper published by Thomas Lund of TC Electronics that you might find useful:

    http://www.tcelectronic.com/media/2040040/mobile-test-paper-2013.pdf
     
    pcrecord likes this.
  3. Sean G

    Sean G Well-Known Member

    pcrecord likes this.
  4. DonnyThompson

    DonnyThompson Distinguished Member

    Another good source - if you're a fan of preservation of dynamics - is Ian Sheppard. He does do a paid/subscription instructional course, but he also provides a great deal of very useful information for free, too.

    He's a strong proponent for these newer LUFS/RMS standards; and I've personally learned quite a bit from reading his articles.

    http://productionadvice.co.uk/about/

    Here's another link, to an article he wrote specifically about YouTube and what they're doing. The article is dated this past March, ( 2015) so perhaps things may have changed a bit in the course of 7 months, ( I've been reading lately that this has been lowered down to -16 LUFS, but like you, Sean, ( and Marco) I seem to be finding that the data is all over the place). I don't think that this is uncommon, though... whenever new standards are experimented with and implemented, I think that the criteria has a tendency to drift while everyone is trying to figure it all out. What I do like is that the awareness of the past destruction of audio is now becoming much more wide-spread... people are talking about this these days, and it hasn't been shoved aside or disappeared as a crazy notion of some sort. Sheppard's personal mantra is "Dynamics is the New Loud": http://productionadvice.co.uk/youtube-loudness/

    Here's an excerpt from the article:

    "So, for example, at the more dynamic end of the spectrum, Mark Ronson & Bruno Mars’ massive hit “Uptown Funk” measures -12 LUFS (DR 8 on the TT Meter) on CD. Whereas “Love Me Like You Do” by Ellie Goulding is squashed up to -8 LUFS (DR 5) on CD, and later in the playlist, Madonna’s “Living For Love” clocks in at an eye-watering (and heavily distorted) -7 LUFS (DR 4!)... But on YouTube, all of them are being played back at a similar loudness of roughly -13 LUFS."

    His list of artists/clients is pretty decent, too - Keane, Tricky, The Royal Philharmonic Orchestra, Deep Purple, The Orb, Leslie Garret, Culture Club, Porcupine Tree, Andy Weatherall, The Las, Ozric Tentacles, Christine Tobin, New Order and King Crimson, amongst many others. I mean, I guess you don't have to have a resume like that to know how to mix dynamically - but it's obvious that he's been in the trenches, and has applied his knowledge and method; so it doesn't hurt to know that he's not just "some guy" just talking about this stuff. He practices what he preaches.

    FWIW

    -d.
     
    pcrecord likes this.
  5. bouldersound

    bouldersound Real guitars are for old people. Well-Known Member

    To nobody in particular: When you quote RMS figures do you know if you're using a sine wave reference or a square wave reference? Do you know the difference?
     
  6. Sean G

    Sean G Well-Known Member

    This is all unknown territory for me @bouldersound ... maybe you could share your knowledge on this
     
  7. DonnyThompson

    DonnyThompson Distinguished Member

    Great point, Boulder. Generally, for music, the newer LUFS standards/measurements are based upon sinusoidal waves peaking at 0dbfs. ( This is the current criteria stipulated by the AES, see link attached below).

    http://www.aes.org/publications/standards/search.cfm?docID=21

    http://www.aes.org/tmpFiles/aessc/20151029/aes17-2015-i.pdf


    You will need to check your DAW's meters ( or your mastering program's metering) to insure that they are calibrated based on this standard, and it's important to know that not all of them are by default.

    As to the difference... if my memory from physics class serves me correct - LOL - sine waves are naturally occurring and can be reproduced, whereas square waves are "unnatural" and are actually the result of overlapping sine waves.

    -d
     
  8. bouldersound

    bouldersound Real guitars are for old people. Well-Known Member

    The 0dB RMS reference is either a sine wave or square wave peaking at 0dB. Of the two the sine wave has less power, is less different from some given audio and gives lower numbers which look "louder". I believe the TT DR Meter uses sine wave reference while my ancient Sound Forge uses square wave reference.
     
  9. DonnyThompson

    DonnyThompson Distinguished Member

    As far as I know, most of the modern stand-alone LUFS meter plugs I've seen ( I looked at a lot of them when I first began mixing to LUFS standards) are based on a sine wave at 0db(fs).

    I do recall that older versions of Sound Forge, Bias Peak and Wavelab all used square waves as the reference, ( my experience with these was back in the 90's and into the early 2000's) I know that my last version of SF did ( v 8?) although I'm almost positive that Sequoia has always used a sine reference. I have no explanation as to why those older versions used a square wave reference, unless it had something to do with their audio engines, or their internal processing - or maybe basing it on the sound cards that were available at the time ?... I'm not sure, though... do you know? I'm now curious. LOL

    I do know that there are still a few meter plug-ins that don't use the current sine wave reference... for a while, IK Multimedia's CS Meter was off by 3 db, as was Logic 9 and Slate FGX ( which is a mastering VST); and Waves PAZ was off by 4.5 ...
    I believe that T-Racks ( IK Multimedia) has since "fixed" their CS meter, but I can't say if the others have since been "corrected" to adhere to the currently accepted LUFS/Sine standard or not.
     
  10. bouldersound

    bouldersound Real guitars are for old people. Well-Known Member

    I think if you're comparing LUFS to LUFS it's a no-brainer.
     
  11. DonnyThompson

    DonnyThompson Distinguished Member

    LOL... no, I'm questioning - long before the LUFS standard was put into place - if there was a particular reason as to why some programs chose to use a square wave reference while others used a sine?

    Did it have anything to do with the audio cards/converters that were available at the time? Or the coding of the program? Or maybe the OS ? etc.

    Why would Sequoia have used a sine reference, while SoundForge and Bias Peak used a square?

    I'm asking because I don't know....
     
  12. bouldersound

    bouldersound Real guitars are for old people. Well-Known Member

    Yeah, I don't know either. Most likely there wasn't a real standard so they just flipped a coin. I think as long as you know which reference you're using it doesn't matter. Why is pin 2 hot?
     
  13. Thomas W. Bethel

    Thomas W. Bethel Well-Known Member

    One of the better LUFS meters was this http://www.tcelectronic.com/lm2-plug-in/ it has unfortunately been discontinued.

    It has been replaced by the LM2n plug in more info here http://www.tcelectronic.com/lmn/

    Of course if you own Ozone it is built into the advanced version. There are some freebie plugins but the best LUFS meters have to be paid for.

    As to what is the proper LUFS for your project...most companies publish specs of what they want on line and you can see them by searching the WWW for technical specifications of the company you need to know about.

    Best of luck!
     
    pcrecord likes this.

Share This Page