In this episode of the show I'm going to address a question from Mark Fordham, a Dynamic Range Compression course member.  We will be addressing the issue of proper recording levels and how they affect the overall quality of our projects. This is one critical area you will need to understand in order to produce a professional mix.


    35 replies to "Ep 30 | How To Get The Cleanest Audio Possible"

    • Derek

      Great video!!! I will be doing numerous tests on my interface when I get home to see how to optimize my recordings.

      Question: I record in Logic, and will sometimes use the “Gain” plugin to add (or reduce) some volume if needed. Does introducing this type of plugin “add” more noise, or will it just be “amplifying” any existing noise. Do you use Gain type plugins or is that something that you’ve already accounted for during the tracking part of your project?

      Thanks all the great info!!

    • Max

      Hi, David!
      Thanks for the information. It’s very useful.
      I’ve noticed that during the recording of electric guitar its pick-ups catch the radio station and all of that noise is being recorded in the mix. I’ve tested the ground, it’s grounded. I use shielded cables, even stereo one. The gear is high quality: Gibson Les Paul standard and Marshall amp. Still the problem is not going anywhere. Do you have any suggestions what may help?

      • David

        Actually try and break the ground to see if this issue goes away, this could be a simple as a three to 2 prong adapter. Another way to isolate this is by using a direct box that can also break the ground.

        • Beau

          HI David, awesome post- very informative. However, when I conducted the basic noise floor test I am not getting a clean wave like you are at the different input levels. Throughout the recording there seems to be an almost regular spike in the wave and I have no clue where it is coming from. I am using a TASCAM UR22 interface. Also, it seems like there is only two gain stages on this particular interface. The first stage goes all the way to 90 percent of the dial and then the second stage blasts in. I am thinking my interface is broken- it is 3 years old- lots of use. Also, when I did the microphone test speaking into it at 3 inches away using the shure 58 I couldn’t get consistent levels around %50 without turning the dial up to nearly max. Finally, what scale are you using in your DAW to measure your input- K-20, K-16, K-14, Digital??? Thanks, Beau

    • Ron


      I was under the impression that the input gain should be set so that the input hovers around 0 on the mixing console. Are you saying that othe input should actually hover between -12 & -6 on the console PLF meter?

      • David

        I’m speaking of the input signal on your audio interface. If you using a digital mixer to record that has an internal audio interface than yes, you should set this input signal not to peak as close to 0db without clipping but to peak at a -12 to -6db max. This for the reasons I’ve shown in this video.

        • Ron

          I’m using a sound craft signature 22 MTK that does multi track via the USB in my church. I send the main outs to auditorium. The inputsi am referring to are on the actual mixing board and I am referring to what in inputs should be for the live sound for the auditorium… Should they also read around -12 to -6 on their respective PLF meters?

    • Kolai

      David i have a focusrite isa one preamp with an ADAT lightpipe out connected to motu 8pre interface. with the 8pre gain pots at zero i assume im bypassing the preamps on the 8pre by using the ADAT cable using only the gain on the isa one? but my real question is am i adding the noise associated with the 8pre in this setup?

      • David

        Well on a technical note, anytime you add something in the chain you’re adding noise. In the situation of a pre-amp, this isn’t necessarily a bad thing. Now on your exact setup, I am not 100% certain since I’m not using your specific interface on how this routing works. I can say that the cleanest way to connect things are as you’ve described. The only real way to test is to use the pre’s without and record just a noise floor with no input, then do the same test with the ISA One and compare the noise floors. Even if you hear more that doesn’t mean this is an issue as these preamps do add a specific character that is often desired.

    • Roobsta

      Great video.

      I use Cubase. Do you know if Cubase has anything comparable to the tool you use to (1) identify the floor noise and (2) subtract the noise? If not, any third party tools or plugins that you would recommend for eliminating the floor noise?

      • David

        Actually, if you’re running a PC, you can download and install just the Reaper plug-ins into Cubase. This will give you the ReaFir I’m using here.

    • Greg

      I tried this with my entry-level Edirol interface without any cables attached.

      7-1 o’clock -80db
      1-3o’clock -mid 70s db
      3-5 o’clock was terrible. -65db

      15db 3x louder!
      It was hard to hear the difference at normal listening levels, but the hiss is there.

      Normalizing the track was the eye-opener.

      Great exercise!

      • David

        Thanks for the feedback and comment Greg!

    • jeffrey

      I haven’t started using the things you teach in the videos yet. haven’t
      decided yet on whether to purchase a laptop or upgrade my desktop for audio
      editing capabilities. if I do upgrade what are the soundcard, memory and processor recommendations?

    • Pete

      I am new to recording on my own and this video was GREAT! I have heard about gain staging but this was explained so well. I am using the focusrite itrack with auria pro and I can’t wait to record my next track. I also enjoyed the mixing vocal series I recently purchased. Thank you. Pete

      • David

        Glad to be of help Pete, thanks for the comment!

    • Steven

      Great info Dave. I’ve wonder about this while recording. I typically set my interface gain to about 5 on the dial. Dial goes from 1 to 10, and then adjust my source(guitar amp)volume until I’m around-12db on my DAW input signal. Would I be better off lowering my interface gain down to say 2-3 on the dial and turning up my source until I get -12bd on the DAW input? Thanks again, excellent course!


      • David

        Good question Steven, so the only reason I would change anything from what your doing is if your preamps seem to be adding a good amount of noise at that halfway point. I suspect their not, in which case your fine. If they are, you may possibly get a cleaner signal by bumping the pre gain down a touch and upping the output gain from your guitar. However, I think your mostly fine with how your doing things. 50% gain on a preamp is in the typically area of most clean signals.

    • Ephraim

      Regarding hissing sounds. I recently bought my equipment (MXL 990s mic, Focusrite Scarlett 2i4) and began regarding. I, like many beginners, turned the gain way up and backed down. I guess I just wanted to see pretty, color changing lights first and then feel I was doing the right thing by backing out a bit. Not enough. There was a distinct constant static/hiss. I checked everything and then finally figured it out by trying a different XLR cable. It was a cheap cable. So, while I will follow Dave Maxey’s very helpful advice on input gain, I will make sure the cables used aren’t crap to begin with. Thanks, Dave. Love the video podcasts.

      • David

        Great point on good cables, thanks for sharing.

    • Luc

      Hey Dave! Thanks for the info. I would love to get your views about using effects at the source level when recording. I like to use delay while recording vocals. Should I always record a dry signal and add plug-ins afterward? What about recording guitar tracks while using pedals etc?

      Luc Gilbert

      • David

        Hey Luc, I personally don’t add FX or EQ to my recording chain. Reason being, you don’t want to lock your tracks into something that may not work later one. The only processing I’ve every done at the source is with electric guitar as you’ve mentioned (I do slightly compression and gate the my voice for the podcast too). The reason I do this for electric is because I’m confident enough in my guitar tone to know what I need and can still make eq adjustments later in the post production phase.

        On vocals, yes I’d always record a dry signal with little or no EQ and I would not record any vocal track personally with delay or any other FX’s for that matter. My general rule here is if I can do it in post than I do. This just gives the greatest flexibility.

    • Frank

      Thank for the tips. The Reafir tip is great. I’ve spent a ton of time automating volume trying to do some of the same principles.

      • David

        Your welcome Frank, thanks for note.

    • Kimble Arbuthnot

      Hi Dave

      Curious. I have a Mackie Onyx Blackjack and a Rode NT1A microphone, which I have done the same testing. I found that when there was no mic plugged in I got the same noise characteristics a on your 2i2, possibly a touch worse. I found this strange as I have always known this interface to be very quiet.

      So I connected up the mic and did the same testing in a quiet room, and most the spurious hiss was gone, only a small amount being present with the full 60db gain. It appears that (with the Blackjack) most of that noise was due to the circuitry being turned on high with no load connected, obviously not high on the list of design constraints.

      Is it not possible that this could be the case on the 2i2? I was quite surprised at the amount of noise that you got and really thought that the Scarlett interfaces were quieter?

      As a matter of comparison I performed the testing primarily with the hardware monitoring on the interface with Sennheiser HD580 headphones, but also confirmed the ‘findings’ on Harrison Consoles’ Mixbus and the iPad app Auria.

      • David

        Like I stated in my video, this is the reason I don’t every push my input signal so hot to get into the final stage of gain on the pre-amp. The reason its typically never an issue is because proper setting of levels with most decent mics give more than enough input source signal. Keep in mind I also adjust the gain of the test so you could hear the noise floor. The noise floor would typically never be that loud even at full gain. With a mic plugged in, you shouldn’t even need that much gain as this was my point to keep a clean signal. Thanks for the question.

    • Mayo Pardo

      Excellent video/podcast David. I too record in a less than perfect environment (don’t we all?) and removing background noise is crucial.

      I use a Dynamic Mic (Heil PR 40) and the Scarlet 2i4, but I find that I have to put the gain on the Scarlet almost at max before the LED Halo meter even lights up. The main reason I got the Dynamic Mic was to help eliminate picking up background noise. But maybe it’s counter productive if I have to put the gain up that high? I’m not using any other pre-amp because reviews on the Scarlet frequently say how good they are in the 2i4 and 2i2.

      I was previously using a Mackie 402 VLZ-3 but my results were not as good.
      Being a newbie at this, a lot of my problems may be “user error”.

      So assuming you were creating a voice over, (like Mark was doing) if you get relatively clean sounding audio at the levels you indicated in this podcast, would you send them to a client at that level or would you boost the levels further?

      • David

        If the client were going to edit the wav files, then I would send them adjusted to -3db peaks. Otherwise, I would “master” the audio before sending the tracks for a final project. The track for my voice I recorded in the show was recorded at the same levels as I demoed the SM58 at. The difference between the 2 mics was the mastering process. I left the SM58 untouched for the demo but mastered my main audio as I always do for the show.

        In that, my mastering chain for a voice-over is this, 1st –>Set peaks to -3db–>Noise Reduction with ReaFir–>EQ to cut unneeded freq–>Multband Compression to level freq–>EQ to boost anything needed–>Full band Compression–>De-essing–>Mastering Limiter to get max db at -.01 db.

    • Adam Parmenter

      I am still stuck in an analogue mindset. “Turn up till you flash the red light then dial back.”. This will make definite changes in my tecnigue.

      • David

        I hear ya Adam, too me a while back in the day to retrain myself as well. You just want to keep that noise floor down as long as you got a good source signal. Thanks for the comment.

    • les malcolm

      Very interesting David!, i will certainly work on this noise reduction situation.I did some recording and heard back ground noise, when music and track came along the noise was cut out to a degree but was still there,with the tips you gave me on the noise issue, i will be making comparison Thanks again Les

      • David

        Your very welcome Les, thanks for letting me know your thoughts.

    • Dan Updegraff

      Good tip Dave, thanks! I am amazed that my M-Audio Profire 610 reacts pretty much the same as your interface. I don’t think I ever recorded anything with the preamp knob near 100%, but I will be sure to stay well below it now.

      • David

        Ya, I never have either, ;) Thanks for the update Dan, keep in touch.

Leave a Reply

Your email address will not be published.