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

Overly simple question about setting levels and gain staging

Discussion in 'Recording' started by sshack, Apr 27, 2010.

  1. sshack

    sshack Active Member

    I've always been cautious about my track input levels being too hot and one of the things that I've done to guard against this is to automatically pull the faders down for ALL of my instrument tracks in Logic from "0.0" to "-0.6".
    I can't say that I have any basis for doing this other than it seems to give me a good balance of being able to push my outboard preamps to get a good sound and avoid clipping the inputs.

    As of lately, I'm starting to wonder about this logic (no pun intended). With all of these points being contributing parts of the overall gain staging process, can someone speak to a 'best practice' for setting individual track levels in a DAW to provide the optimum balance for quality of sound and reasonable headroom?

    For the most part, I rarely track with any EQ, comp, or limiting, save using my LA-610, but even then it's minimal. My preference is to record as 'organically' as possible (no, I'm not a hippie) and to allow some room for compression, etc. later in the mixing stage.

    I've really searched for guidance on this and just haven't been able to get a solid answer. It seems that a lot of tutorials that I see tend to revolve around software instruments of which I don't use other than the occasional keyboard plugin. That said, all of my recordings are for acoustic/electric guitar, vox, bass and drums.

    Thanks.
     
  2. toddmatthew

    toddmatthew Guest

    Haha... "Overly simple" never describes gain staging. I see it all the time, "engineers" with more experience than myself don't know what they're doing with this.

    But a good rule of thumb I've read, heard from, and experienced myself is to record at much lower levels than you would inherently think of. Generally I limit my peaks to -12db. A professor of mine swears by being able to peak around -30db. The lower you record, the more headroom you have. In the digital age, our noise floor is extremely lower than that of analog. We don't have to worry about tape hiss, and all that jazz. What we do have to worry about is what else is getting in our signal, like mic self noise, anything else going on in the room, noise from your LA-610 (yes it has noise, all amps do), etc. But if you've got a good signal chain and you've controlled your rooms noise floor a bit, you shouldn't have any trouble recording at really low volumes. And then you'll be able to put all the EQ, compression, limiting, delays, reverbs, autotunes (joke! kinda...) that you want!

    There is one downside to this however. Most gear, and even DAW's don't really balance things out for you in the lower volume areas. Meaning you've got exponential control on faders, notice the large distance between 0db to -20 db, and the small distance between all that lower stuff. So it can sometimes be difficult to be really accurate with your monitoring.

    But seriously, "no blood on the tracks" is a good philosophy to live and die by in recording. The louder you record, the more limited your dynamic range will be. I personally still stick to -12, I think it provides plenty of headroom and a good signal to noise ratio.
     
  3. boxcar

    boxcar Active Member

    im a big fan of this plug in.
    i use it to set the recording levels on my tracks.
    i also like my peaks around -12....average -18

    Download Inspector Free by Roger Nichols Digital
     
  4. hueseph

    hueseph Well-Known Member

    I hate to tell you this but there is no point to bringing the faders down in Logic. All of your gain staging has to be done before it even get's into the box.

    Are you still using the Duet? If it is possible to control the input levels to the interface that would be where to turn it down. As it is, I think that you'll find once you turn your faders back up to 0 it should be fine. If it wasn't clipping with the faders down, it shouldn't clip with the faders up. All the faders do is control the output level of each channel/track.
     
  5. soapfloats

    soapfloats Well-Known Member

    This is something I toss around in my head from time to time, as well.
    Not adjusting the DAW input faders, but gain staging related to DAWs in general.

    I have pretty much the same approach as sshack.
    I typically go for peaks between -16 and -12dBfs (sometimes reaching above -10dBfs), while shooting for an RMS level around -22dBfs. (Also typically ends up a little higher, of course.)
    I'll also hit the snare, kick, or bass (2 of the 3, depending on source) w/ my dbx comps to tame the peaks a little. Other than that, I go au naturel.

    Disclaimer: I know there are no "textbook" levels at which to track.
    My main concern is the gain staging related to digital.
    I use Cubase SX3, and have the option of mixing in 24-bit or 32-bit float.

    So here's my question, and I think sshack's for the most part:

    Does your level coming into the DAW (they are usually similar to the final analog level, but each interface/DAW will show different levels)
    vary if you're working w/ 24-bit vs. 16-bit vs. analog (tape, etc).
    Shouldn't the added bits give more headroom, and allow one to track a little hotter (so long as there's no clipping)?

    Can of worms, I know.
    I'd rather get the experiences of the people here, that I KNOW use the stuff and practice it daily, rather than that of the people trying to get me to buy it.
     
  6. rmsaudio

    rmsaudio Active Member

    I have some thoughts - I hope these are helpful. I don't tend to track at different levels between 16 and 24-bit but in theory I think you could track at a lower level in 24-bit and still have the same or better dynamic range that you would in 16-bit. So in otherwords, I don't think it's as much about headroom as it is the range of 0 to loud. 24-bit allows you the freedom to sample audio at a quieter volume, allowing you to more accurately record the original dynamics of the performance.

    With that thinking in mind, I will typically set my levels fairly conservatively going into my pres and to the converters (particularly for percussive instruments like piano). Once the signal is in the digital domain I manage the signal at each processing stage to make sure it's not clipping anywhere within the sofware.

    I'm not sure if this is helping.. Here's another way to think about it. It's not as much about getting a "hot" signal recorded. That was the way of thinking with analog as analog tape gives you some margin of error if you start to distort (headroom). With digital, it's about getting a "clean" signal recorded. So gainstaging before conversion is important to keep noise floors and distortion levels low. If you get clipping at the conversion stage with digital, it is unforgiving,

    Once you're in the digital domain, think of 0dBFS as your absolute limit (I actually prefer -0.5dBFS and only for the loudest peaks). So if your signal was recorded at -20dBFS, you have some room if you EQ, maybe compress to increase and even out the volume of the track. You'll want to constantly monitor/manage to make sure that any of the processing being done in the software isn't creating "overs" at any stage though.

    The way to get a "hot" sounding mix in the digital domain has more to do with managing the overall spectral content and dynamics of the mix as opposed to gainstaging before A/D conversion or recording levels during tacking.

    Cheers :)!
     
  7. RemyRAD

    RemyRAD Guest

    There can be advantages to recording at 24-bit but some of these advantages cannot really be produced. 16-bit nets you 96 DB of working level. 24-bit provides you with 140 DB of working level. But here is the catch in the get along. THE ELECTRONICS THEMSELVES GENERALLY CANNOT PRODUCE MORE THAN 110 DB OF USABLE RANGE. So what about those other 30 DB? In a sense it's like vaporware. This is what our software is going to do when we can figure it out, maybe, etc. It allows for some better processing at levels that would naturally be below the noise floor of 96 DB. But if it's still that low, you won't hear it in the mix, not until you raise the level and then still have to deal with whatever extra low-level noise you will introduce. So 24-bit can actually provide for false confidence. I frequently record hot digitally and I am not grossly upset when short duration transients may clip. This can actually be advantageous to certain kinds of recorded material. Most monitor systems provide between 5% to 10% speaker distortion on average. So you are not going to hear a 1 or 2% of harmonic distortion when your monitors will only present you with a minimum of 10 to 20 DB of harmonic distortion.

    Not all distortion is bad or inappropriate
    Mx. Remy Ann David
     
  8. sshack

    sshack Active Member


    Don't hate, I'm here for the learning. :)

    I actually have an RME FF800 now, which is capable of doing WAY more than I need or know how to use. In fact, I would venture to say that I'm probably not even using it optimally yet as so much of it is over my head and I'm hard pressed to find real world examples or tutorials that I can apply to my scenarios/situations.
    And quite honestly, I've never even given the 'bit' aspect of this conversation much thought. I'll admit, I don't fully know the difference either. No pride here. I'd love to be educated in these areas of recording and enginnering, but I'm not and unfortunately I don't learn as well by just reading. At the end of the day, the majority of my learning has come from experimenting....just turning knobs, moving mics, trying stuff out.

    It's been so long since I've left my faders at "0", I'll just have to go back and try it. Again, it seems in so doing that I would have to scale back on driving my preamps, particularly if I want my peaks around -12.

    I appreciate the conversation and input from everyone though.
     
  9. TheJackAttack

    TheJackAttack Distinguished Member

    If you don't want to scale the preamps back then buy yourself a set of XLR pads. They come in 10dB, 20dB and 30dB varieties. Then you can keep your preamps cranked and still have the levels low. But in fact Hueseph is correct. The digital gain controls in a DAW don't affect the signal coming in at all.

    As to learning any interface, the best learning proceedure is to have it all set up. Then read for a bit followed by experimenting with that aspect of what you just read. Lather, rinse, repeat.
     
  10. sshack

    sshack Active Member

    No worries...I'll just go back and experiment some more. The thing that probably moved me to post this is that I just bought a drum kit and am now venturing into the realm of trying to get a good recorded sound. I'm in the middle of trying just about every mic and pre combo that I have, including padding said mics (when applicable) and/or preamps. It's certainly a whole different world from my norm of recording guitars and vox, but I'm loving it.

    Hopefully I can post a sound clip soon for critique.
     
  11. hueseph

    hueseph Well-Known Member

    The faders you are pulling back: are they in your DAW or are they in the Fireface's mixer software. You should be able to pull those back. That would be the right place to do so.
     
  12. sshack

    sshack Active Member

    Yeah, see...I'm not sure about the faders in Total Mix (FF interface) either. The faders that I'm referring to pulling down are in Logic, so yes, the DAW. I've tried messing around with the levels for the FF800 and honestly, it only confused me. Are you familiar with the RME?

    This is where I feel like I'm doing something wrong. When I record, I get good tracks, but I wonder if I knew what the hell I was doing with the levels, FF800 included, that I could get an even better recording.

    I'm all ears......
     
  13. hueseph

    hueseph Well-Known Member

    Most of the mixer software for the interfaces are based on the same concept. Turning down the fader in RME's Total Mix should lower the level going into your DAW. That would be the right place to adjust the input level. You still have to watch for clipping but this will allow you to push your preamps for sure. Thanks for bringing this up by the way. It's something that I've never tried with my DAW.
     
  14. TheJackAttack

    TheJackAttack Distinguished Member

    The Totalmix controls CAN affect the levels depending upon which controls you are adjusting. To really help out your understanding go to the matrix and click a few things. Play back or record through those channels whichever and then go to the mixer view and watch what is blinking. Pull those faders up and down and check it out. You can also adjust pan settings too. I have mixed down sessions via the FF800 rather than a DAW for instance and I know Cucco has utilized the Totalmix to provide three or four different headphone/2 track mixes before. The trick of course is to experiment extensively and write questions and notes to answer those questions prior to using a new configuration at a session or live gig.
     
  15. sshack

    sshack Active Member

    I recorded some scratch tracks yesterday to experiment for that very reason and really didn't see that adjusting the faders in Total Mix made any difference to the levels in Logic. I went back and watched some RME tutorials as well and I think Total Mix is really more applicable to providing mixes for overdubs, etc. There are a few more things that I want to try tonight though just to flush out some ideas.
    Another aspect is how the FF800 handles I/Os from my outboard verb and compressor (which, technically I won't get until Wed)...but I have messed around with the reverb (LXP-1). It really just boils down to me getting my head around how things get routed into and from the FF, into and from Logic, back out and so on.
    Once I 'get it', I'll have it.

    Maybe Cucco will chime in at some point and set me straight.

    You guys are helping me think through it though, which I appreciate.
     
  16. TheJackAttack

    TheJackAttack Distinguished Member

    When you're recording through the FF and if you have selected a mono track, then obviously there is no pan setting etc. When looking at the Totalmix mixer, the top row is the gozintas-what's coming into the FF. The bottom row is the gozoutas-what the FF is sending out to the hardware/speakers/headphones/etc. The middle row are returns from your DAW-these are useful for sending individual channels back out for submixes or headphone mixes or out into a summing box for analog mixdown. The first thing I would do is practicing routing your recorded tracks back into the FF as individual sticks and play with them that way, sending them to various outputs or singly to the headphones. Play with the pan and faders etc.
     
  17. boxcar

    boxcar Active Member

    i had a look at the FF800 manual and it's a lot like the my onyx 1200f.
    total mix is only for setting up sub groups and heaphone feeds as far as i can see.
    levels are set on the hardware box through the trim pots.
    ajusting the sliders in logic only determine the level of the master bus output.

    the inspector plug i sugested works great, just insert it on the track your recording in logic and bring the hardware trim pot up for that channel till you have the level you want on the inspector plug.
    that will be your true imput level for that track.
    i have a template with the inspector plug on every track. the nice thing is you can use it to set your imput levels and then on playback(when you add compression and e.q. and such) you can see how much gain your adding to that track so you don't clip .
    then when i mix down, i put it on the master bus and i can see what all my tracks and master bus plugs are adding up to. i don't like to go past - 3 to -5 there, unless im doing a final mix for myself and using waves L2 or something like that, i'll go -1.

    just a sugestion.
     
  18. hueseph

    hueseph Well-Known Member

    There's some handy info.
     
  19. boxcar

    boxcar Active Member

    levels are set on the hardware box through the trim pots and the plug in.lol
    i know that in nuendo,i can look at vst imput levels but those meters are not very accurate.
    i would think you could do the same in logic but im not familiar with that app.

    EDIT: actually i had the vst imputs confused with the vst outputs in nuendo so there is no way in nuendo that i can see how i would accurately tell what the true DAWS imput levels are except by using a track plug in.
    unless someone can point out another option.
     
  20. sshack

    sshack Active Member

    There really aren't any trim pots on the FF800, unless you're referring to the gain pots on the front four (ok, five) channels...and those are only applicable when using the preamps in the FF. That's obviously very straight forward, however that doesn't apply for my other preamps.

    JackAttack - I'm familiar with the format and again, understand it conceptually, it's just making the practical connections (again, no pun intended). However your comment about routing recorded tracks back through the FF is a great idea, something that I never thought of. I appreciate the tip...maybe I can make some progress today.

    You guys rock.
     

Share This Page