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

VU meters explained

Discussion in 'Recording' started by achase4u, Mar 13, 2015.

  1. achase4u

    achase4u Active Member

    No worries, I have searched the forum for answers but didn't see these questions asked.

    So a couple years ago, I learned about the whole dbfs vs dbu vs VU etc. I also learned about converters being aligned at various places, mostly in the -18 neighborhood, which of course lines up with most plugin interal operating levels etc, because it mostly equates to the age old "line level", 0VU etc.

    The thing is, I have learned you have to think about how you use a VU. I use the Klanghelm plugin on my input FX chain, and when mixing.

    I sort of think of instruments in three categories - peak, rms and mixed instruments. To me, drums are mostly peaks. An electric bass or guitar is rms. Acoustic guitar is mixed. Some vocals are mixed etc.

    So obviously, its pretty damn easy to set up via the VU meter your level for an overdriven amp in a rock song. The whole thing pretty much keeps the meter still and the amp compresses things so much, you don't usually have surprise peaks.

    Drums, I don't really use a VU for - I aim for -7 to -5 on the VU if I do - but usually -6 for max peaks in the DAW.

    Without knowing what my converters are aligned at(Roland doesn't bother listing this as they are budget converters) I set my recording levels to -20 dbfs RMS, to give plenty of headroom, and a clean sound. Sure, if I want to drive my preamp a little harder, I don't mind pushing it a little more than that. Here is my question though. How does one traditionally interpret a VU vs a digital VU? If I am at -20=0VU, do I make sure that the needle only rarely goes over into the red? Or is it supposed to just bounce back and forth from +3 to -3 etc? Because if I try to stay out of the red mostly, the average signal is actually below -20dbfs. Nothing I record except electric guitar and bass really hovers around 0 that constantly.


    Does that depend on your boards headroom and individual design? When it comes to tape(I use plugins with VU's), do you simply do it by ear? Does it depend on the tape type and machine setup?

    I have the feeling that, like everything I care about doing, I am overthinking this. I just like to understand things the best I can. Something tells me that a VU is just a general guide, and as long as its not pinned and flashing, it's fine... and when using tape sims, your ears are more important...
     
  2. DonnyThompson

    DonnyThompson Distinguished Member

    @achase4u @Boswell @audiokid @pcrecord @kmetal @Davedog @dvdhawk @paulears @pan60

    The VU meter was designed as a sort of "RMS" measuring device. In a way, it's kind of measuring a "perceived loudness", and isn't nearly as detailed as modern db measurement devices are, because at the time it was developed, that kind of detailed/incremental accuracey wasn't needed as much as it is today. It's purpose wasn't really intended to measure the signal in detailed increments per se', as much as it was used to measure an "average " level, as well as a "target" level.

    When it was designed, tolerances for audio peaking were quite a bit more "relaxed" than things are today, where you now have a definite, finite db point that you can't go above. Tape was fairly forgiving in that regard, and because of tape's inherent noise level, the idea was to get as much signal to tape as possible to cover up the noise...without distortion/clipping. The development of "levelers" like the LA2a and the Fairchild 670 (which eventually became known as "limiters"), allowed the engineer to increase the RMS with as much level as possible without clipping. Today's digital recording methods are whisper-quiet in comparison to tape, so it's not as necessary to masque any inherent noise.

    That's not to say that you can't use VU meters in modern audio production - you most certainly can, and with the advent of "classic" processor emulations like the LA2, 1176, and various channel strips that implement a VU as the measuring guide, you don't have a choice, you have to use VU metering with those processors, because that's the way they are set up, based on the original design(s).

    Many still use VU metering, but generally not for critical measuring, where things like digital "overs" need to be shown; because in this day and age, if you go above a certain db level, even by the tiniest amount, your project can be rejected.

    There are some broadcast facilities/networks that have their own db level criteria for using audio, and if you send them audio that goes over their specified maximum tolerance, even by the smallest amount, it can be kicked back to you because it violates their tolerance.

    I sent a project out to a video post facility a few years ago - they had a specified max of -9db - and for one lightning-fast moment, the audio I sent them reached a peak level of -8.5, just one time .... and that .5db difference was enough for them to send it back to me for correction.

    I think you have a decent understanding of the subject, but it's still good to know what you are doing with your current metering. Your final audio output is really what is most important, and that is where you should be using your most detailed and critical measurement method. There are some very good and accurate software meters out there, and for not much money, either. As long as you stay within a given specified range - which is target dependent, of course - and that you're not going over, (and that everything sounds good on the whole) then you'll be fine. ;)

    FWIW

    d.

    edit: this is a pretty wide topic, there are many factors to consider; I chose to not get into a three page instructional thread on metering, and I was sure others ( cough - Boswell - cough LOL ) would also chime in as well, adding things that I'd maybe overlooked. ;)
     
    kmetal likes this.
  3. paulears

    paulears Well-Known Member

    I love the question "How do you use a VU", and it was always "carefully". Very few people, when they were popular, ever used the 0 point as a limit. You used the red section. How much, depended on how the recorder or next device in the chain reacted. I knew that +3 on the meter was fine on my reel to reel with the BASF tape I recorded to. My system was aligned so that 0db on the mixer meters was 0dB on the recorder meters. That done, I then pushed the levels to +3 happy that it would not be distorted, and occasional red lights on the meters wouldn't really matter. The meter needle then seemed to make pretty good sense. Once I added a DAT machine and an ADAT, everything got tricky because those occasional red light peaks now sounded nasty, so I had to drop the levels. The damn VUs then tickled in the low range, too low to really mean very much, and I had to look at the recorder meters to see the quieter bits were actually getting there. Now we have digital meters that show the wider range. I'd hate to go back to the VU meters I was perfectly happy with back then. They just don't really work that well when we have a big danger warning at 0dB
     
  4. pcrecord

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

    With tools like Insight (comes with Ozone from iZotope) metering comes easy. Altought you still need to interprete what all the tools are saying ;)
    izotope-ozone6-advanced-insight-full.png
     
  5. DonnyThompson

    DonnyThompson Distinguished Member

    Yup. That's kinda what I meant when I mentioned that VU's were meant to be used as a kind of "target" measurement. There were some tape machines - as well as some tapes - that sounded great when you found that "sweet spot" above 0db... depending of course on how the deck was aligned and what kind of tape you were using.

    I still use VU's all the time - for internal processing, mostly because I'm used to that type of measuring in those certain pieces that used it originally; plugs like the 1176, LA2, Focusrite Red, the 670, etc. - and second, because the plug doesn't give me a choice. LOL.

    Then again, who would really want to see an LED meter on an LA2A or 1176 anyway? It'd be kinda like putting satellite radio into a '56 Chevy Bel-Air. Yeah, it would work, but some things simply don't belong together. ;)

    Anyway, as far as the 2-bus goes, I always use a modern, incrementally accurate/detailed digital-type of metering... which also allows me to see RMS measurements, too, if I choose. I'm currently using the IK/T-Racks CS Meter, which gives me Peak, RMS/Perceived, Phase/Correlation, and a Spectrum Analyzer frm 20hz - 20k.

    The bottom line for me, at least for the 2-bus, is that I can get the digital meter to "act" like a VU if I want, but, I can't get a VU meter to act like a detailed modern accurate meter.

    ;)
     
  6. pcrecord

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

    Customers are always impress with the retro look of VU meters. I've put mine all on the same side of my desk on purpose (see pic)
     

    Attached Files:

  7. DonnyThompson

    DonnyThompson Distinguished Member

    There's no doubt that clients are impressed by the visual. I remember when I switched from an analog desk to a digital console around '96 or so... I'm sure that I probably lost a few clients along the way when they saw the CR, and the much smaller size of the digital desk compared to the typical LF analog board. I had a friend/competitor of mine in those days, who also switched to a digital console about the same time that I did, he had been using a Tascam 3500 Series desk... and when he picked up a new digital board, he left the 3500 desk up and powered on, for those clients who were impressed by size.

    I still keep a couple racks around, loaded with gear that I never even use, because clients are impressed by it when they see all that gear. They assume it must mean I'm better equipped, and accordingly, better at what I do. They see all the gear, the little blinking lights, the knobs, the faders, the meters, and it gives them the impression that I'm seriously geared up. And, I am competently equipped... solid computer, good RAM and CPU, good cables, clean power, nice preamp, nice mics, great DAW platform... but with the exception of the mics, those aren't things that they can see...those important things are not as obvious to them as the OB hardware is.

    They have no idea that not one of those pieces in those two racks is even connected to anything, nor do they know that one of the most important pieces of gear I own is a small, unassuming device, about the size of a cigar box, with one gain control knob and 4 little buttons; 48v, Z, Pad and Ø ...

    LOL ;)
     
    pcrecord likes this.
  8. audiokid

    audiokid Staff

  9. pcrecord

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

    Thing is, it's not that VU can't be accurate. It's just that they are slow.. Once you get the hang of them they are handy and look great !
     
  10. achase4u

    achase4u Active Member

    Growing up seeing VU's sort of imprinted something in my mind - the VU may be slow and not very accurate for faster transients - but I feel like its as good as any modern RMS meter, right? I like using VU plugins mostly in the tracking phase. If I have guitars, bass or vocals coming in, I watch the meter. I have mine calibrated so that 0 VU = -20 dbfs - which is theoretically the same as say, an analog consoles headroom. So to me, the use equates to the same thing, unless you are accounting for transients being dampened in the analog world, i.e. compression/less worry of an over. So I suppose my use of the VU plugin has been "correct", in that I try to stay around 0 - if it bumps into the red some, its ok. I've never had an over working this way. Drums I simply use the TPM of my daw.

    I think I am confusing two issues here that can't really be equated to one another - tracking levels in dbfs and in all analog system. That is, the rms of a waveform into a DBFS scale as it relates to converter alignment(i.e. hitting the sweet spot of A/D) and the rms of a waveform as it relates to tape(which can be aligned and biased etc etc very differently)

    Trying to nail -18bfs rms going in usually always gets peaks far too close to 0dbfs. So even when aiming for 0 vu on my -20 calibration, I am really averaging -25 or so usually if I look at a digital RMS meter.

    The reason I ask all this, is because I like to try new things and commit to sounds(Alan Parsons, anyone?) I like getting things up front the way I want them to sound. I love mixing, but I hate mixing. Meaning, I like mixing when its icing on the cake. This also means I have lately been tracking through Slate virtual tape and UAD studer tape. I feel like my favorite analog albums really don't have blatantly audible tape distortion. Getting anywhere near 0 vu on 456 15 ips on the Studer is too gritty. I digress.

    I wish I could be more concise - but I am trying to figure out how to fit metering into my workflow. The VU seems to make sense because its simple. I just want a needle to look at when I am setting a level.
     
  11. audiokid

    audiokid Staff

    If I'm reading into this right, don't look at the VU level for helping in getting a capture translating to the mix upfront. VU meters are there to tell you something is happening.
    Brains aside, up front mixes comes from, excellent performance and micing, high end tracking gear, well produced music and samples that gel with the organic counter parts. The moment I stopped using VU's to gauge how a wave was going to sit in a mix was the day I stepped up my game.
    The best advice I can offer in regards to meters: Don't come close to the red with digital and listen to what analog sounds like rather than ever trusting a VU meter. Analog gear has a mind of its own.

    I'm thinking about getting a rack of them for my Neos. Its like adding more lights and chrome to my ride.
     
  12. kmetal

    kmetal Kyle P. Gushue Well-Known Member

    +1 on burnt , or half burnt bulbs.

    a personal area of interest of mine as of the past few weeks is the physcoacoustic effect of lights. The lower the lights the more dimension I hear. I think it has to do with our sight requiring a a lot of our brains resources. I belive its a brain intensive sense, relative to the others. A lot of musicians aren't even looking at what they're doing.

    I'm slowly losing my affection for physical vu,s like one of my pres and tape machines have. They are kinda cool cuz they physically move, but they take up a lot of space in the rack if you have a lot, and with all the other options available, this is one place where I'm future focused, and have to sentiment towards 'classic'. Although one pair would be cool to have for fun.
     
  13. DonnyThompson

    DonnyThompson Distinguished Member

    About ten years ago, the VU lights on my Studer 1/2 track (B77) died (not at the same time, but that was even worse, one on and one off...LOL) and when I went to replace them, I didn't have any white lights, just a pair of red ones.

    So I put them in, and they worked fine, but the first time I had to look at them in the dimmed lights of my control room, it was like the house in the Amityville Horror. LOL... two evil, red "eyes" staring out at me from the dark...

    I'd never realized that my 2 track had a "personality" before that... but I swear, it actually had an expression on its "face".

    I'm positive I heard it whisper... "We're gonna get along just fine, partner... just as long as you keep me degaussed.... Oh, and don't even start thinking about connecting up any noise reduction to me, either...." :confused:
     
    pcrecord and kmetal like this.
  14. kmetal

    kmetal Kyle P. Gushue Well-Known Member

    Artists....
     

Share This Page