How do you calibrate your VU Meter ?

Discussion in 'Mixing and Mastering' started by Blue, Jan 25, 2018.

  1. Fernando Roma

    Fernando Roma Ultrasonic

    Joined:
    Nov 10, 2016
    Messages:
    41
    Likes Received:
    31
    I like to do my gain staging at -18, always worked for me....The Hornet VU Meter have a auto gain function, so i drop it on all my tracks, play the song and at the end, all my tracks are calibrated at -18 rms with picks around -9...
     
  2. Blorg

    Blorg Producer

    Joined:
    Jan 1, 2018
    Messages:
    356
    Likes Received:
    117
    *Any number above -18dBFS, e.g. a pulse train with short duty cycle.
    Conversely, if D&B are too loud @ 0dBFS, turn down your speakers. Or waste bit depth. Irrelevant for 32-bit float; very relevant when rendering to 16 bit int (CD; YouTube, Soundcloud etc.).
    Your track is roughly half as loud as an identical one peaking @0dB, while effectively gaining you ... lower bit depth:dunno:[​IMG]
    That might be ~0dB for peaky stuff like drums, and ~-16dB for tamer stuff. VU meters chop peaks, both because started out as moving coil/needle meters (so inertia, takes time for the needle to swing) & by spec (https://en.wikipedia.org/wiki/VU_meter#Standard_characteristics)
    [​IMG]
    Arcane shit, sorta like a mechanical 300ms loudness integrator, a half-step to today's loudness analyzers. But 90% of real VU meters didn't follow spec, were more of "keep needle below red" type of deals. Have no idea what VST VU meters do.
     
    Last edited: Jan 25, 2018
    • Winner Winner x 1
    • Interesting Interesting x 1
    • List
  3. mild pump milk

    mild pump milk Russian Milk Drunkard

    Joined:
    Mar 29, 2014
    Messages:
    2,683
    Likes Received:
    2,262
    Location:
    Russia
    Set fader to overclip your audio (not clip with plugin limiters/clippers) or gains etc. Export audio as 24 bit and 32 bit float. Both are hot distorted. But ! Load these 2 wav files in daw and gain down. 24 bit file will be quieter, but still hot distorted, unchanged. 32 bit float file will be quieter and clean without distortion and clipping. So, float preserves audio above 0 dBFS. Safer to work with.
    Same happens internally in plugins with their 32/64/80 bit float engines, double/extended precision.
     
  4. Zenarcist

    Zenarcist Audiosexual

    Joined:
    Jan 1, 2012
    Messages:
    3,915
    Likes Received:
    2,500
    Location:
    Planet Earth
    The "In The Box" video I mentioned above recommends tracking internally at 24-bit, while exporting for mastering at 32-bit float.

    I'm not sure what the significance of this is, as the music conservatory professor who is presenting the tutorial does seem to be on a much higher lever than I am with regards to sound engineering :)
     
    Last edited: Jan 25, 2018
  5. Blorg

    Blorg Producer

    Joined:
    Jan 1, 2018
    Messages:
    356
    Likes Received:
    117
    I think all DAWs use floating point math internally, 32 bits or better (if that's what you mean).
     
  6. Zenarcist

    Zenarcist Audiosexual

    Joined:
    Jan 1, 2012
    Messages:
    3,915
    Likes Received:
    2,500
    Location:
    Planet Earth
    @6 min 38 sec

     
  7. No Avenger

    No Avenger Moderator Staff Member

    Joined:
    Jul 19, 2017
    Messages:
    8,912
    Likes Received:
    6,110
    Location:
    Europe
    Right, but profools is for loosers [​IMG]
     
  8. Blorg

    Blorg Producer

    Joined:
    Jan 1, 2018
    Messages:
    356
    Likes Received:
    117
    ProTools uses 32-bit float internal processing (its audio engine) http://avid.force.com/pkb/articles/en_US/troubleshooting/32-bit-reverb (first google hit)
    Don't use current ProTools, but my guess is choosing 24-bit int saves files as 24-bit int files. This doesn't mean that they're processed as 24 bit files internally. They're transcoded to 32-bit float as soon as they load, and all the processing is done in 32-bit float.

    Why anyone would want to choose 24-bit over 32 float is a mystery to me, disk space costing what it does... I mean, for an average project, might amount to a fraction of a cent:dunno:

    Curiously (or not), Windows (since Vista) does the same thing with all bit depths. An audio driver may be running @ 44.1/16, but Windows audio stack (mixer and stuff) transcodes it to 32-bit float & does all the math in 32-bit float :)
     
    • Agree Agree x 1
    • Useful Useful x 1
    • List
  9. Zenarcist

    Zenarcist Audiosexual

    Joined:
    Jan 1, 2012
    Messages:
    3,915
    Likes Received:
    2,500
    Location:
    Planet Earth
    I'm using Reaper @ 24 bit :) I mostly record analogue inputs such as guitars, vocals and real synths, so 32 bit float isn't such a great help to me, plus I get to gainstage my plugins correctly. But I am always open to new ways of working, and it's interesting to discuss these kinds of things :wink:

    PS thanks to this thread, I have only just discovered that Reaper has finally implemented (since v5.32) pre-fader metering! I've been waiting ages for this :rofl:
     
    Last edited: Jan 25, 2018
  10. No Avenger

    No Avenger Moderator Staff Member

    Joined:
    Jul 19, 2017
    Messages:
    8,912
    Likes Received:
    6,110
    Location:
    Europe
    Why are you doing this? It's nearly impossible to clip a 32bit floating point channel...
     
  11. Zenarcist

    Zenarcist Audiosexual

    Joined:
    Jan 1, 2012
    Messages:
    3,915
    Likes Received:
    2,500
    Location:
    Planet Earth
    Except on analogue inputs. I aIso got into the habit of gainstaging my plugins correctly, so I will never distort or clip my audio signals. There should also be smaller file sizes and less CPU overhead?
     
    Last edited: Jan 25, 2018
  12. Zenarcist

    Zenarcist Audiosexual

    Joined:
    Jan 1, 2012
    Messages:
    3,915
    Likes Received:
    2,500
    Location:
    Planet Earth
    24 and 32 bit Recording Formats in REAPER

     
  13. Blorg

    Blorg Producer

    Joined:
    Jan 1, 2018
    Messages:
    356
    Likes Received:
    117
    *your audio interface/driver is set to 24-bit
    Reaper, like most modern DAWs (and Windows itself), uses float internally. That's why you never have to worry about your DAW clipping (or "running out of headroom") internally.
    What does that mean, and how does changing bit depth break it?
    I, too, can't plug a meter into the first effects slot, or route insert A to insert B. Computers is hard :(
     
  14. Zenarcist

    Zenarcist Audiosexual

    Joined:
    Jan 1, 2012
    Messages:
    3,915
    Likes Received:
    2,500
    Location:
    Planet Earth
    All I can say is when I changed to my new workflow it sounded better IMO, possibly due to the plugins running at their optimal settings? Who knows, but my VST amp sims & various FX sound great. Digital VST instruments account for less than 10% of what I do, so gainstaging just happens to be where the action is for me right now.
     
    Last edited: Jan 26, 2018
  15. No Avenger

    No Avenger Moderator Staff Member

    Joined:
    Jul 19, 2017
    Messages:
    8,912
    Likes Received:
    6,110
    Location:
    Europe
    You're right. :like: I'm always very careful with analog input levels (keeping them at -6dBFS max to be on the safe side), I also thought we're talking about in DAW level, so I just forgot about that. And of course your inputs don't work with 32bit FP.
     
    • Like Like x 2
    • Agree Agree x 1
    • List
  16. Blue

    Blue Audiosexual

    Joined:
    Sep 6, 2015
    Messages:
    1,807
    Likes Received:
    954
    OK.Some of you obviously use LUFS Meters instead of VU meters.
    Does somedy know which target is fine to set in the LUFS meter when mixing ( in EDM ) ?
     
  17. No Avenger

    No Avenger Moderator Staff Member

    Joined:
    Jul 19, 2017
    Messages:
    8,912
    Likes Received:
    6,110
    Location:
    Europe
    The EBU recommendation is -23 LUFSi (integrated = whole program/song) which is, to me, by far not loud enough for any kind of modern music (maybe even not for classical music).
    As different platforms have different LUFS targets, there is no real law, or rule.
    I've seen and heard songs with less than -10 LUFS which is, again to me, incredible loud and has a significnt lack of dynamics, but opinions and perceptions differ.
    For EDM I would set a target of -14 to -12 LUFS.
    True peak (TP by LUFS measuring, not by DAW) shouldn't exceed -1dBFS.
     
    Last edited: Jan 26, 2018
  18. Blorg

    Blorg Producer

    Joined:
    Jan 1, 2018
    Messages:
    356
    Likes Received:
    117
    Is that in Whitworth VU RMS, or metric?
     
  19. No Avenger

    No Avenger Moderator Staff Member

    Joined:
    Jul 19, 2017
    Messages:
    8,912
    Likes Received:
    6,110
    Location:
    Europe
    Kelvin, smart-arse. [​IMG]
     
  20. Von_Steyr

    Von_Steyr Guest

    I once took a lufs meter on my phone to a date and measured how loud her average yada-bull-crap is. After i measured it was way above anything i could work with and even peaking severely into my left side of the brain, i paid and danced off in slow motion moonwalk style.
     
Loading...
Loading...