Jump to content

Low Bounce Output - confused


Recommended Posts

I've created a tune that I'm actually pretty proud of (my first I'd actually proudly play for someone other than myself).  I want to mix it into one my DJ sets for my mixcloud radio show.  The trouble is when I compare the levels to commercially released tunes in the same genre, mine is almost exactly 3dB lower than.  I'm not looking to be the LOUDEST out there, but I don't want to have to raise the volume of my track (or lower the rest) by 3dB to make the transition not incredibly obvious volume-wise. 

 

Here's my current finalization workflow:

 

  1. I've mixed the tracks to what I believe is good level balance, cutting unused frequencies where they exist, tamed as many level spikes as I can find.  When playing the tune start to finish, the level meters on my master output peak around -7dB  and RMS is around -19dB.  
  2. I then bounce the project to a Wav file with nothing on the master bus, and normalization set to OFF.
  3. I then open a new "mastering" session in logic pro x, add my newly bounced waveform 
  4. I add some final EQ tweaks, compression, limiter (also tried the adaptive Limiter) etc. on the master out, using pretty tame settings, nothing crazy in terms of gains or ratios etc.  My limiter is set to -0.2 dB output, gain set to keep the actual limiting amount to peak at around a 1.5dB reduction max
  5. I play the tune in Logic, I see a pretty solid max output of -0.2 dB
  6. I bounce the master output into what we can call my MASTERED copy for the sake of this question.  I've bounced both with and without normalization on, normalization seems to help bring things up a bit  
  7. I play the tun in my beat matching software, it show up on the meters there about 2-3dB lower than all the other tunes I use....  :cry:

For the life of me, I cannot figure out why it still is so low in comparison.  I've attempted to reduce any flared up peaks via EQ, volume automation, etc, I've monitored and remixed the entire track at least 4 times.  

 

I'm willing to admit that I need lots of practice in mixing, but can anyone see something simple I'm forgetting like some checkbox hidden somewhere that affects the master output by 3dB etc.?  For the record, all my mastering plug-ins are logic's built-in plug-ins. 

 

Thanks in advance :)

Link to comment
Share on other sites

1. Bounce of your mixdown seems to be fine. 

2. It reads like, after you bounce the mastered version, you still have the peak way below 0dBFS

 

So my first guess is that you somewhere have a plugin output gain at -3dB, or just the master fader is set to -3dB. Or maybe on the Logic Limiter you have the output ceiling at -3dB.  

 

Have you checked the levels of the reference song with your mastered song in Logic? 

 

PS: Normalization will bring the tune just by the 0.2dB in your case. So this cannot be the cause. 

Link to comment
Share on other sites

@triplets -  I'm not really sure how to answer that. (sorry).  Since I'm not outsourcing mastering, I'm essentially just trying to emulate loudness with comparable songs.  All I can say is it seems my track is seems low.  My limited understanding of this all (trying to learn) tells me that my RMS levelsis too low by the time the peaks are just shy of clipping.  Is there a value I should be shooting for? 

 

@stardustmedia I have checked, there are no plug-ins that have a -3db and my final limiter's output ceiling is set to -0.2.  I will have to check the reference song in logic.  Rookie question: The typical channel meters next to the track faders are measuring in dBFS correct, or is there some adjustment I need to make to read that?

 

Additional thoughts:

I think I need to compress the overall master signal to get the difference between transient peaks and RMS signals reduced.  Not so much to smash the dynamic range completely, but more to control problem areas.  Is this the right idea?  

 

Also, yesterday I tried putting an Adaptive Limiter on near the end of the master chain with a marginal gain increase, and then put a Limiter (does this count as a "brick wall" type limiter?) on immediately after it, and it seemed to help boost the signal.  It seemed to do the trick.  Is there a problem with this "Dual limiter" approach?  

 

I greatly appreciate the schooling :)

Link to comment
Share on other sites

First, why not set your ceiling at 0? Are the files you're comparing your mix with peaking at -0.2 dB FS as well? It's only 0.2 dB but still, might as well compare files that have the same peak value. 

 

Next, loudness is a vast topic that has to do with more than just levels and dBs — so in general it's expected that your self-produced track won't compete in loudness with tracks that were professionally produced and mastered. Here's a good article on loudness by Lagerfledt: https://www.gearslutz.com/board/music-computers/468170-loudness-when-producing-mixing-tips.html

Link to comment
Share on other sites

@triplets -  I'm not really sure how to answer that. (sorry).  Since I'm not outsourcing mastering, I'm essentially just trying to emulate loudness with comparable songs.  All I can say is it seems my track is seems low.  My limited understanding of this all (trying to learn) tells me that my RMS levelsis too low by the time the peaks are just shy of clipping.  Is there a value I should be shooting for? 

Logic has a multimeter that can tell you the RMS levels with yellow bars and yellow numbers. RMS is the average loudness measurement.

Import one your references into Logic and look at the multimeter to see their RMS levels.

Link to comment
Share on other sites

So what I've found is that I have are few (almost random) transient spikes.  I can see on the waveform where they occur (usually in two spots), but I'm not exactly sure what causes them because admittedly I can't hear them standing out.  

 

When I use logic's multimeter to check the frequency spectrum across the main out, nothing really stands out as "too high" to me, or a sudden jump, however I can see the little peak number sit there at -6.9, -7 or so, and then jump to -3.2 suddenly 

This to me says its either just a cumulative effect of all tracks mixing together poorly at this one point, or some frequency that is causing a peak that I'm not hearing, too high, too low perhaps.  

 

Any tips on identifying the source of and taming the peaks of these little pesky rogue peaks?

Link to comment
Share on other sites

Those two (random) peaks cannot be the cause for low loudness. The only thing they may cause is the limiter to reduce more dB when they occur. 

 

Just give it a try and set the limiter with some stronger settings. The gain increase should be obvious. If the loudness then still is very low, I guess the issues derives from somewhere else. BTW: I don't suggest that this "poor man's mastering" will sound any good, but it MUST increase loudness. 

 

To find those peaks: You have to go thru channel by channel, then plugin by plugin to find the culprit. 

Link to comment
Share on other sites

If normalization is bringing your signal up, then your peak values are somewhere below the 0dB level. 

 

Those pesky peaks you are talking about seem to be the culprit. The sample editor can find the highest peak and you may be able to do some gentle limiting to help bring up your overall value (RMS level).

 

Your final RMS value should end up somewhere between -12 to -8dB.

Link to comment
Share on other sites

Something else to think about is this: "the screen of your television is actually gray," as you can very-plainly see when you turn the thing off.  Yet, a black object in the movie that you're watching does appear to be black.  Why?  Contrast and context!  Your brain doesn't work in absolutes: everything that you perceive is in context with everything else that you are perceiving from the same source at the same time, and in contrast with them.

 

  • The web-site page that's in front of you right now isn't "white."  If your monitor is decent, it's the color of paper.  The font that was used for the words that you are right-now reading also isn't "black."  But it will be an evenly-darkened variation of the "paper" color, so that the result is "monochrome," as the visual designers intended.  In some fonts, punctuation-marks are very slightly darker, especially big ("navigational") things like "bullet-point bullets," as used here.  Take a look.  Betcha you never noticed that – but, if your eyes are finding their way to searching for their destination on the page, these "little things" help.

 

Another key difference between "recorded audio" and "live performance" is that all of the sounds that you hear on a recording come from just one source: the speakers.  All of the sound is being produced by a membrane that, at any given instant, is either being pushed or being pulled by an electromagnet.  Whereas, in a live performance, the sounds come from each instrument, individually bounce around the room, and arrive at your ears from different directions (unless you are in the back of the hall such that you are actually just hearing the PA system).  "Surround sound" is still only an approximation of this effect.

 

Your eyes can detect a far greater range of "brightness and darkness" (in film terms, "f-stops") than film or video can, because your eyes are never still.  They constantly scan the scene in front of you, and you assemble "your perception of it" in your brain's visual cortex.  Photographers and videographers must work within the "terrible limits" of their medium to produce an image that your eyes will accept as "real."  And, they must understand how you will go about perceiving it.  Specifically:  they know that your eyes are always attracted first to "the brightest and the most-contrasty thing in the frame" (because "that's where the tiger is ..."), and that they then "follow a path" through the total picture, always wanting to finally return again to the starting point.  So, they compose the image "to guide you."  Your hearing works in much the same way, and so a good recording will also "guide you."  This is "psycho-acoustics": the art and science of designing an audible presentation based on knowledge of how the human brain will perceive it and appreciate it.

 

As you listen, your brain is constantly "listening to" one particular thing ... just as you can discern your friend's voice in the hubub of a room where many people are talking, whether or not his/her voice is "the loudest sound in the room."  (If (s)he whispers, "gee, this band sucks ... let's just go back to the hotel and make love," you will hear it ... :) ... no matter how 'loud' the band is.)

 

A good recording will, in any particular "short passage of time," guide the focus of your primary attention on one thing, but it will also present your secondary attention ("other things that you could also decide to focus your attention on right now") with a fairly small selection of other things (the "context") that you can "easily hear, too."  If a particular sound ceases to stand out at all anymore (i.e. "over very-recent time"), it soon becomes part of "the hubub in the room."  You can no longer focus your attention on any particular part of "the hubub" without great effort.  So, if you wanted that sound to continue to "stand out," but it doesn't (anymore), your brain becomes confused and distracted.  (For this reason, supporting sounds should be clearly "ushered in" and then "ushered out again," so that the listener doesn't suddenly wonder where-the-hell they went!)

 

All sounds, both instantaneously and over (short) time (your "attention span"), should clearly fit into one of these three tiers:  primary attention ("listen to this!"), secondary ("you could also decide to listen to this, instead"), or tertiary ("hububububub ... don't mind this").  And they should clearly enter, step forward or back as the case may be, then exit.

 

When your auditory perception gets de-railed, you call it "muddy."

 

So: the final range of output levels should fit comfortably into the physical constraints of both the recording and the playback systems, but the mix should also gently guide the listener's ears to focus upon first one thing and then another and another and the next.  A recording "sounds loud" when all three things – lead, background, and "hubub in the room" – sit properly together while "also, of course," making good use of the available bandwidth of the recording medium and the playback system.  Your primary attention is being guided from place to place, supporting sounds make clean entrances and exits, and "everything else" still fits.  The recording will sound good – and, will sound "loud (and clear)" – at any volume level.

 

When editing a movie, directors pay attention to the overall brightness, contrast, color-balance and so on from one scene shot to the next, so that the screen does not appear to "flash."  (Commercials, on the other hand, do the opposite.  So do horror movies.)  "Contrast and context," again.  They also pay attention to what area of the screen you are probably looking-at when the show "cuts" from one shot to the next, so that you are not "looking around to find it" and thus being forced to become consciously aware of the editing process that has taken place.  Instead, and in ways that "you never think to think of," your attention and focus is constantly being guided.  If done well, you never notice.  You will simply sit back and enjoy the show, and wonder when your sweetie will tell you that the show sucks and that both of you should just go back to the hotel and make love.   8-)

Link to comment
Share on other sites

  • 3 years later...
@triplets -  I'm not really sure how to answer that. (sorry).  Since I'm not outsourcing mastering, I'm essentially just trying to emulate loudness with comparable songs.  All I can say is it seems my track is seems low.  My limited understanding of this all (trying to learn) tells me that my RMS levelsis too low by the time the peaks are just shy of clipping.  Is there a value I should be shooting for? 

Logic has a multimeter that can tell you the RMS levels with yellow bars and yellow numbers. RMS is the average loudness measurement.

Import one your references into Logic and look at the multimeter to see their RMS levels.

here is another thread where someone is confused when playing his bounces from the finder.

 

There is a very good reason to be confused, Apple had no real answers to this when I contacted them last week., because other apps bounce finals that sound identical in volume playback via the finder, example you play the pro tools sessions while previewing the bounce in the finder and both sound identical.

 

Maybe what's going on is logic sounds louder during the session compared to other apps or and daws, but what's odd is no clipping or distortion is noticeable, I am able to pull of finals., and regardless of the fact the file is finals are fine and its all that really matters, it is a bit odd that Logic of all DAWS is louder then the finders playback of the final.

Link to comment
Share on other sites

Have you checked the volume settings for your system out and the volume settings for Quicklook? Because when I put everything at unity, Logic is exactly as loud as Music (formerly known as iTunes) or Quicklook or any other audioplayer. No wonder "Apple" couldn't help (well, most of them don't really know much about audio anyway), because it is within your setup that the difference is made. It really is. O, and perhaps by now also perception bias: you have convinced yourself that it is there, you are "on a mission".

But there is no mission.

 

1138745188_Schermafbeelding2021-04-20om13_13_39.thumb.png.b795f11fd2f678e6870f1e2359f25739.png

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...