1. Register NOW and become part of this fantastic knowledge base forum! This message will go away once you have registered.

Setting levels & calibration

Discussion in 'Pro Audio Equipment' started by dkrausz, Jul 3, 2001.

  1. dkrausz

    dkrausz Guest

    Sorry for these questions (and the long post) but...

    It’s just that I’m entering what could only be described as a sort of “phase two” to my music production ability. I’m finding more and more, very subtle things are effecting my mixes and I’m on the verge of going from “tight and clean” to “open and big” Sometimes it’s there and sometimes it’s not.

    In the past I have always been aware of setting good levels and not to clip things. I try to only peak (but never clip) the level indicators only during the loudest passages of the piece trying to keep some head room available for dynamic expression in between. But in the end, I find that my masters only “translate” well to CD about 50% of the time. ( the conversion dither thing is also driving me nuts as well, but I digress...)

    Anyway, I’m now becoming obsessed with this whole level and calibration thing and it’s driving me CRAZY! It seems that the littlest adjustments to levels (especially on EQ and limiters) seem to make the biggest impacts.

    I produce instrumental music and sound design for TV documentaries and am currently working on a low budget film. It’s “all midi” using a sampler, controller and sound modules as sound sources so I don’t use any live musicians (well I guess I’m alive... sort of)

    This is what happened when I tried to calibrate my system.

    Within PTLE/001, I created a test tone with the tone generator plug-in set at 0db . Of course, when I play the tone on a stereo pair the meter goes right to the top just before the red clip indicator lights up and holds solid right there (no clip) which is exactly what I’d expect.
    OK Duh...

    The master fader on my Mackie 1604VLZ mixer has a +28db red over load indicator ABOVE the +10db mark. (I’m assuming that the area between 0 and +28 is the head room before distortion gets too nasty)
    I do realize there is a big difference between analog over loads and digital clipping.
    So anyway, I patched the test tone out from the Digi 001 main outs (which are rated at +4db line level) into a couple of channels on the mixer panned hard left and right. Trims, master fader and channel faders all set at unity.

    The result was +10 over the 0db mark which is exactly one notch below the +28db red over load indicator on the Mackie.

    So the test tone plug says 0db, the level in PT is pegged but not clipped and my mixer master fader shows +10db. OMG! What's happening??

    First question is;
    Is the level meter in PT showing max output BEFORE clipping AS 0db or is it something higher (like on my analog mixer)? In other words, what is the level indicator in PT actually showing when it’s pegged to the top but not clipped.
    Why IS it pegged if the test tone is only generating a 0bd signal?

    And, where do babies come from? :)

    Last question is about Line Levels.
    Is it a “standard” level that a device such as a sampler or synth should produce when it’s gain is set to maximum? If it is, what is that number?

    The more I learn, the more confused I become. :eek:

    My goal is to go from “knowing” to “really knowing AND understanding”

    Any light shed on this level stuff would be greatly appreciated.

    Thanks,

    dk
     
  2. MPlancke

    MPlancke Member

    Originally posted by butterhead:
    Sorry for these questions (and the long post) but...

    So anyway, I patched the test tone out from the Digi 001 main outs (which are rated at +4db line level) into a couple of channels on the mixer panned hard left and right. Trims, master fader and channel faders all set at unity.

    The result was +10 over the 0db mark which is exactly one notch below the +28db red over load indicator on the Mackie.

    So the test tone plug says 0db, the level in PT is pegged but not clipped and my mixer master fader shows +10db. OMG! What's happening??


    First question, are you using the line inputs on the Mackie or the Mic Inputs. If you're using the Mic Inputs you will be picking up a significant amount of gain from them.


    First question is;
    Is the level meter in PT showing max output BEFORE clipping AS 0db or is it something higher (like on my analog mixer)? In other words, what is the level indicator in PT actually showing when it’s pegged to the top but not clipped.
    Why IS it pegged if the test tone is only generating a 0bd signal?


    If you're generating 0db IE a full scale digital signal in Pro Tools, this is not equavalent to 0dbm in the analog world. 0dbm is (I'm going from memory here) .775 volts into a 600 ohm load. This should be register on your digital meters somewhere around -14 or -12.

    Check out the Rane web site for some technical notes.
    http://www.rane.com/pdf/par.pdf
     
  3. anonymous

    anonymous Guests

    Originally posted by butterhead:
    The master fader on my Mackie 1604VLZ mixer has a +28db red over load indicator ABOVE the +10db mark. (I’m assuming that the area between 0 and +28 is the head room before distortion gets too nasty)
    I do realize there is a big difference between analog over loads and digital clipping.


    OK...let's start here just for shits and giggles...first, there is no Mackie Mixer on this or any other planet that is putting out +28 with out it sounding like it ran through an MXR Distortion +...especially above the +10db mark. Brother, I know there isn't an SSL that can do that, and I'm pretty sure there isn't a Neve that can do that, though the Trident A-Range might get close [it runs on 50 volt rails...Mackie's don't].

    That said...let's move on to the real crux of the buiscuit shall we...


    First question is;
    Is the level meter in PT showing max output BEFORE clipping AS 0db or is it something higher (like on my analog mixer)? In other words, what is the level indicator in PT actually showing when it’s pegged to the top but not clipped.


    It's showing you your absolute '0dbfs' limit. This has little or nothing to do with the "+4" output of your analog desk. You will need to pick a 'headroom point' under '0dbfs' for your reference to "+4 Zero". I usually use -12 on the digital meter scale...but there is no standard. The chicken $*^t motherfuckers at AES [and several other bodies that occassionally come up with standards] are too afeared of pissing off a potential 'revenue stream' manufacturer to name a standard.

    It's basically an 'arbitrary number' you pick on your own. I use -12, some use -14, some of the 'classical guys' like -18 or -20. Seeing as it's a fictional standard...make up your own number, then stick to it. For some reason we seem to make up these numbers as even numbers...though I think I'm tempted to start doing my projects with a "13db<0dbfs" [a.k.a -13 under 'digital zero' which is a.k.a. "0dbfs"] reference level to "+4 Zero". I like the number 13...it's my lucky number.


    Why IS it pegged if the test tone is only generating a 0bd signal?

    The question is which "0db" There are soooooo many to choose from, you're going to have to be way more specific about which Zero is your Zero of the moment.

    And, where do babies come from? All too often some slut trollop you treated like a farm animal one night when you were doing a show in some god forsaken midwestern shithole of a city.

    Last question is about Line Levels.
    Is it a “standard” level that a device such as a sampler or synth should produce when it’s gain is set to maximum? If it is, what is that number?


    Again, a pretty arbitrary number, though far more controlable...

    The more I learn, the more confused I become.

    That just means you're heading in the right direction.

    My goal is to go from “knowing” to “really knowing AND understanding”

    Any light shed on this level stuff would be greatly appreciated.


    Ask David Bock...he knows this $*^t pretty well...he may even be able to explain it in a manner that makes sense...I just know kinda how it works...I really don't know [and frankly don't care] about the "why" part.

    Best of luck!!
     
  4. MMazurek

    MMazurek Guest

    I thought the crux of the biscuit was the apostrophe?
     
  5. MPlancke

    MPlancke Member

    Originally posted by Fletcher:


    I like the number 13...it's my lucky number.



    And all this time I thought 69 was your lucky number.

    Hey, the Little Labs PCP arrived today, all is well in the land of guitars. Thanks again Fletcher and thank Jay for me will ya. Oh, and I meant to ask, who's the new staff member answering the phones these days makes me wanna buy more gear. <g>
     
  6. realdynamix

    realdynamix Well-Known Member

    "And, where do babies come from?"
    ---Butterhead

    "All too often some slut trollop you treated like a farm animal one night when you were doing a show in some god forsaken midwestern shithole of a city."
    ---Fletcher

    Oh My god!!LOL, is there anything left in the cooler?

    Hope this helps also, http://www.sospubs.co.uk/sos/apr98/articles/gainstructure.html
    --Rick
     
  7. dkrausz

    dkrausz Guest

    Gentleman, thank you VERY much for the great responses!

    I get it!! :D

    I know exactly what I have to do now.
    But first, I will make coffee...

    dk
     
  8. anonymous

    anonymous Guests

    Originally posted by MPlancke:


    And all this time I thought 69 was your lucky number.


    No, that's my favorite number [activity...whatever]...13 is my 'lucky number'...met the wife on a Friday the 13th, the youngest kid was born 13 minutes past the 13th hour on October the 13th...now, is that some 'Willie Dixon' $*^t or what?

    Hey, the Little Labs PCP arrived today, all is well in the land of guitars. Thanks again Fletcher and thank Jay for me will ya. Oh, and I meant to ask, who's the new staff member answering the phones these days makes me wanna buy more gear. <g>

    You're probably talking about 'Samara'. She's great, she knows DAW's inside and out. The running joke around the office is that Samara is to Digital what I am to Analog...it's probably true...but I'm trying to learn more about the digital $*^t, and I have a feeling she's going to learn more about the analog stuff purely by osmosis hanging around the shop.
     
  9. WSpeckmann

    WSpeckmann Guest

    ...well, well, hmmm...

    things are a little more complicated once you start to exchange audio material with others - and you certainly have to - sometimes.
    So we do in fact have standards for headroom levels etc. (at least here in europe). There is this fabulous EBU recommendation R68. According to this your 0dbFS level is 9 db above your nominal full output (which is +6dbu for radio stations here, i.e. 0dbFS = +15dbu)
    Very easy - unless you start to work with it. First: 9db can be v e r y tight if you're recording real music - gotta have good limiters. (Digi stuff is set to 14db headroom)

    There you are - building up a wonderful, new, a l l d i g i t a l radio station. No level problems at all, 'cause you're connected via aes/ebu or s/pdif. Everything is computer controlled, even most of the music is played from harddisk. There are some additional CD & MD players, a DAT deck and a audio CD Recorder. Just check the levels - no chance to do this on the fly - you have to calibrate everything in these wonderful touch screens on your new studer digital desk. Pop in your favourite CD for a while - ohoh: too loud ! There's no such thing as a -9db/R68 CD ! They're all a s l o u d a s p o s s i b l e ! Okay, 9db pad on the CD inputs. What's the level/headroom of the mp3's from harddisk ? Just ask the computer guys. Big blue children's eyes. Can you give me an example of full level ? Yes ? O.K., it's somehow minus elevenish (dbFS), so you need to add 2 db. Everything else should be fine. It's all digital ! You're recording the first on air session for some guests - the CD isn't really loud - it's -9dbFS. If you play it on your CD player it comes with -18dbFS - rember the 9db pad ? Everybody's complaining. Same with your recorded DAT and MD. Then a young band brings the DAT copy of their new recordings - B a n g - it's nearly smashing the desk: a full 0dbFS is sent to the desk and further on to the transmitter (as it's analog, it's of course +15dbu).
    This can (and will) be v e r y expensive for your radio station - the level situation is strictly controlled by Deutsche Telekom - the guys who provide the lines to your transmitter.
    The problem is that it's not just you choosing your individual headroom - every product (CD/MD/DAT) is produced ass l o u d a s p o s s i b l e. (Ever checked a Phil Collins CD with a good peakmeter?)

    I hope this adds to the level confusion
    Kind regards - nevertheless
    Walter :eek:
     
  10. dkrausz

    dkrausz Guest

    umm... thank you, and yes.
    :) :roll: :eek:
    :confused:
     
  11. Tymish

    Tymish Guest

    Um...there are faders to control your output levels to air right? ;)
     
  12. WSpeckmann

    WSpeckmann Guest

    Yes Yuri,
    you are right, there are faders, but you cannot add 9 (or 12 or 14 db or even more) if the level is too low - and if you are using faderstart for DAT or MD (which i s usually done that way) most people are n o t fast enough to avoid this high level 'Blow out' to the transmitter.
    The average radio moderator isn't even expecting this to happen. Bring up the fader to zero, then have a look & listen, rise or lower it by 2 db, that's the way it goes. (Usually a radio desk is perfectly calibrated, you could use switches instead of the faders)
    'Pick any headroom you like' - it's just not that simple. If you use daw's to edit, play normal cd's, copy to dat, burn to cd things turn out to be much more complicated for the normal user.

    Kind regards
    Walter :p
     

Share This Page