Should the ideal rendered mix peak closer to 0.0db or less?

  • I have been rendering (finalizing) my mixes so that the master level is in between -2.0 to -0.5. But should I push them closer to 0.0db? Does is make that much difference? The increased volume usually sounds good to me even if it is a very slight increase.

    Larry Mar @ Lonegun Studios. Neither one famous yet.

  • Louder always sounds better. It is really a creative choice. I think it also depends on what you are planning on doing with the songs. If you are putting them on a streaming service they seem to volume level things before it goes across the stream so there may not be any real discernable difference to the listener. Professional producers are always trying to get as close to 0 as possible it seems. Try bumping it up just a hair and see what you think. As long as you don't push it into clipping you are good.

  • Since digital became a thing every producer in the world has been trying to make their songs the loudest possible. I have looked at pro recordings where the bass drum hits are completely clipped off to the point of being square waves. Sounds great.


    Most rock songs the signal is full on every moment of the song. I have no idea how they get that done without large amounts of distortion.


    Sadly many albums now have to be remastered since they were made too crunchy :D Another casualty of the Loudness War.


    So, I don't think you can make them too loud. The last song I did, the first mix I made was massively compressed. To the point I thought it was a little crunchy. I opted to use a later, less compressed, version but the hot version sounds a lot stronger and way more rocky. My wife pointed out, when she heard the later mix, that it was not as good as the first.


    I work entirely in Goldwave when "mastering" a mix. Their are a few different compression options. The best results I get are when I compress the track multiple times at differing levels and amounts after normalizing the volume.


    Compressor settings:

    - Attack 0 mS

    - Release 0 mS

    - Smoother OFF.


    EXAMPLE for a typical rock song (varies with content):

    - EQ the song.

    - Normalize to 0dB. Apply 2dB compression at threshold -12 to 16 dB.

    - Normalize to 0dB. Apply 2dB compression at threshold -6 to 9 dB.

    - Normalize to 0dB. Apply 2 - 3 dB compression at threshold -3 to 6 dB.

    - Normalize to 0dB. Apply 12dB compression at threshold -1 to 2 dB.

    - If needed - Normalize to 0dB. Apply 6 - 12 dB compression at threshold -1 dB. To clean of any straggling peaks.


    I am not saying this is a good technique. Because it is not. Just giving an example I use as a complete studio noob to try and get the loudest song I can. I would imagine using multiband compression in your DAW would get you a much better sounding result. Goldwave just lets me EASILY SEE the song down to the sample level. And I have developed a feel for where and when to apply compression by looking at the waveform peaks.


    One thing the Loudness War has taught people is "nothing is better than having well recorded material". Record everything as bright and clear as possible. Any dark or muddy parts will greatly limit how loud your final mix can sound.


    From an electrical engineering standpoint I cant imagine any gear worth its weight would have an issue playing full volume (normalize, peaked) audio. You know ahead of time the loudest possible voltage your DAC can possibly make. But I am noob. Curious what lightbox as talking about.

  • I'm using multiband compression on the Bass and sometimes drums. I like to keep the guitar tracks as raw as possible so I only use compression on relatively clean tones and that's where I use a stomp. I don't know if this is the right way to go about compression in the studio but it sounds decent to me. I have experimented with some sidechain compression and have done some parallel compression but that's the limit of my knowledge thus far.


    I think lightbox is referring to the metering of db when you use digital to analog. So, a 0.0db on a digital meter (in DAW) may read +0.5db (and clipping) when it gets converted back to analog... something like that.

    Larry Mar @ Lonegun Studios. Neither one famous yet.

  • I'm using multiband compression on the Bass and sometimes drums. I like to keep the guitar tracks as raw as possible so I only use compression on relatively clean tones and that's where I use a stomp.

    I was going to bring this up (but I usually ramble on too long so I did not). I do not compress individual tracks ever. This would probably get even louder recordings. My bass, drums, guitar, and vocals are usually pretty compressed at record time anyway.


    I am too ole school and use my DAW like a tape deck. One shot charlie style. No raw signals recorded. Maybe someday I will try to reamp with the Kemper :/

  • Most rock songs the signal is full on every moment of the song ... an example I use as a complete studio noob to try and get the loudest song I can

    No matter how "loud" you mix, streaming platforms will take measures and turn it down.

    Check the line with red text to see how much they turn these stupid maximized tracks down ;)

    You can check that on your own by right-click in a Youtube video and select "Stats for Nerds". ;)

    Also ... if you want loud, turn up your amp, monitors, headphones. You get a LOT more punch out of a dynamic track compared to a track that's been compressed to death.


    From an electrical engineering standpoint I cant imagine any gear worth its weight would have an issue playing full volume (normalize, peaked) audio. You know ahead of time the loudest possible voltage your DAC can possibly make. But I am noob. Curious what lightbox as talking about.

    The problem is that checking the individual samples' level isn't enough. Between samples the level can go beyond what one thinks. If you normalize a sample (or multiple samples) to 0dB, there's no more room for the proper "curves" that can go beyond the highest sample. People always think in these kind of staircases when it comes to samples. But once you convert the digital audio back to analog, the "analog" curve is recreated ... unless it's clipped.

    Check the image below to see peaks between individual samples (inter-sample peaks):

  • I'm using multiband compression on the Bass and sometimes drums. I like to keep the guitar tracks as raw as possible so I only use compression on relatively clean tones and that's where I use a stomp. I don't know if this is the right way to go about compression in the studio but it sounds decent to me. I have experimented with some sidechain compression and have done some parallel compression but that's the limit of my knowledge thus far.


    I think lightbox is referring to the metering of db when you use digital to analog. So, a 0.0db on a digital meter (in DAW) may read +0.5db (and clipping) when it gets converted back to analog... something like that.

    I would recommend not compressing anything when you are tracking it. Do it in the DAW. I never apply effects when tracking that my DAW has in it. That gives you the ability to make any changes you want after you have recorded the track. This is one thing that I can't understand about the thought of putting a bunch of effects on your guitar tone in your profiler or modeler when you record a track. The only thing I will put on the input signal is overdrive. I record either a clean or overdriven track and apply the rest of what I want in the DAW, unless it doesn't have it then I will put that on the input signal, such as a wah. Once you record it with the effects on it, you have to record it again if you decide later that you don't like it. I find this approach better than reamping.

    I was going to bring this up (but I usually ramble on too long so I did not). I do not compress individual tracks ever. This would probably get even louder recordings. My bass, drums, guitar, and vocals are usually pretty compressed at record time anyway.


    I am too ole school and use my DAW like a tape deck. One shot charlie style. No raw signals recorded. Maybe someday I will try to reamp with the Kemper :/

    I always apply compressors to the individual channels. That is where I want to control each track's dynamics. I will also sometimes put more than one compressor on a track. I set them both to very light compression in succession of one another. Then you get an end result that sounds pleasing and is in a dynamic range that I want but doesn't pickup any negative artifacts from compressing it a lot more one time.

  • No matter how "loud" you mix, streaming platforms will take measures and turn it down.


    The problem is that checking the individual samples' level isn't enough. Between samples the level can go beyond what one thinks.

    The whole problem stems from people playing songs back to back on a device like an MP3 player or stereo. All of the songs need to be at the same volume. On a streaming platform I can see they are smart enough to adjust each track so they appear to be at the same volume. The only way to make them all similar across platforms is massive compression and peaking them as high as possible. The physical limitations of the medium become the volume limiter.


    The example you gave of the waveform being rebuilt is very cool. But I would imagine the unit would be designed knowing in advance this is the case. And should be able to recreate/oversample/etc within safe limits. And we also have to remember people love MP3s smashed down to 128 kbs that sound like a robot singing in the shower 8o


    But I love this talk guys. This is how/when learning becomes fun. Hanging out with smart people talking shop on the stuff you love. Thanks for the great info.


    EDIT: I was going to reference that Metallica album as being one that got remastered since it was too crunchy. Rush Vapor Trails is another I know off the top of my head.

  • I never apply effects when tracking that my DAW has in it.

    I apply the effects during the recording process because I need that feedback, but these are DAW plugins. I assume you mean the same. I will use Kemper stomps/effects when I do a DI track for reamping.

    Larry Mar @ Lonegun Studios. Neither one famous yet.

  • lightbox I am talking about the optimal loudness of the mix so that what I hear translates to what the audience hears. Like even if they turn down the volume they will still be able to hear those nuances I hear when rendering at the optimum output level. Does this makes sense? Not sure I am saying this correctly.

    Larry Mar @ Lonegun Studios. Neither one famous yet.

  • I apply the effects during the recording process because I need that feedback, but these are DAW plugins. I assume you mean the same. I will use Kemper stomps/effects when I do a DI track for reamping.

    That is correct, you are adding it in the DAW. One thing really cool about the interface that lightbox recommended to me that I purchased is that it has a DSP in it and you can put effects on just the monitors. It doesn't become part of the recorded track but gives you what you need to get into the groove for the recording process.


    If you are putting a DI track in for reamping, I would keep that completely dry for the same reasons I mentioned before. That way you have the ability to add and remove whatever you want to down the line. You have a lot more options with a dry track for reamping.

  • But I love this talk guys. This is how/when learning becomes fun. Hanging out with smart people talking shop on the stuff you love. Thanks for the great info.

    I agree. I have been looking for a good forum for these types of discussions with people that take it seriously and not just throwing jabs at each other. I was thinking we need a good recording section here for those discussions instead of cluttering up the music section with our techie threads.

  • I am finding his input to be invaluable. I am very thankful that he is taking the time to post on these threads.

    Of course. Pun on lightbox was intended.


    I have two new mixes almost finished. Check them out when I post in Link Your Music section and we can critique and discuss to make them better. I'd look forward to it.

    Larry Mar @ Lonegun Studios. Neither one famous yet.

  • Of course. Pun on lightbox was intended.


    I have two new mixes almost finished. Check them out when I post in Link Your Music section and we can critique and discuss to make them better. I'd look forward to it.

    I know that feedback is valuable when learning these things. I am always very hesitant to give my opinion on these things. Every single time I do it seems to draw negative comments back toward me, even though the person asked for input. The last time all I did was say that I thought the drums were a little too far forward in the mix on the speakers I was listening to it on at that moment but I would need to get it in my monitors to really get a good feel for anything else that may stand out to me. That caused a comment to come back to me that they thought they sounded fine and I must not have much experience with mixing. What that should have told the person is that my speakers are different than theirs and on this particular set of speakers it didn't sound as good as it did on his and he may want to revisit it and see if he could pull the drums back a little without losing the sound he was going for. I don't mind giving honest input but it is just one man's opinion. There is no reason to be rude to someone that took the time to listen and give the feedback. He could have just said thank you and moved on. For some reason that has been the normal response type that I have received.


    I have been doing live mixes for many years. I always feel like there is something new I can learn to make what I do better. I tend to get a lot of good comments about my mixes and I am a bit in demand in my area even though I have told people that I am not real interested in doing that type of work right now. I am not trying to brag and I hope it isn't coming across that way. The reason that I mention this is that I am not new at this stuff. I have mixed a lot of bands, some national touring bands. I am never giving feedback in a malicious manner. I even try to qualify it with a softening statement of something like "this is just my opinion" or on the speakers I am listening to it on this is what I am hearing" but that has not helped at all. I have just found it better to keep my opinion to myself sometimes.


    Sorry for the rant / babble.

  • The whole problem stems from people playing songs back to back on a device like an MP3 player or stereo. All of the songs need to be at the same volume. On a streaming platform I can see they are smart enough to adjust each track so they appear to be at the same volume. The only way to make them all similar across platforms is massive compression and peaking them as high as possible. The physical limitations of the medium become the volume limiter.


    The example you gave of the waveform being rebuilt is very cool. But I would imagine the unit would be designed knowing in advance this is the case. And should be able to recreate/oversample/etc within safe limits. And we also have to remember people love MP3s smashed down to 128 kbs that sound like a robot singing in the shower 8o


    EQ, EQ, EQ... Never under estimate the EQ. What you are really talking about and looking for is "perceived loudness." You can affect that with different frequencies. It doesn't have to be a super loud master with the right EQ. Combine the loud master with the right EQ and you have a song that will penetrate the skulls of everyone that gets near it.