Getting rendered mix volume to match reference track

Abram4235

Member
Messages
5,073
I'm finishing my first song in reaper. I rendered down my tracks at -6db. When I reopen in reaper and attempt to use a limiter to increase the volume, I cannot get my song as loud as the reference track I am using, or as loud as any other commercial song. I have applied a compressor on my original mix before rendering but beyond that, I have no idea why I cannot increase the volume to match a commercially released song. Any ideas would be extremely helpful.
 

Multicellular

Member
Messages
7,880
There is no simple answer to this. Doing it well requires good gain staging and compression and eq at the mixing stage. Also make sure your reference track isnt ridiculously loud. Read Bob Katz book about mastering. Read the thread at Reaper forums (seriously, dont let the title turn you off, one of the best things written about recording around, not really Reaper specific)

http://forum.cockos.com/showthread.php?t=29283

Most people will use a brick wall limiter. And it is good to use an rms type loudness analyzer like Toneboosters Loudness, once you understand what that all means.
 

oldhousescott

Gold Supporting Member
Messages
2,736
Another trick is to not try to make up all your loudness in one stage, at the final limiter, but to spread it across a couple of compressors and the final limiter. You can also use a multiband compressor/limiter before your final brick-wall limiter but you need to be careful with those as it is easy to destroy a mix with one. There's a freebie comp-limiter by VladG called Limiter No6 that has most of this rolled up into one plugin, and it actually sounds pretty good.
 

zenmindbeginner

Gold Supporting Member
Messages
1,803
The loudness war takes a lot of skills and experimenting to wage... experience helps too.

I'd leave off the compressor and use a limiter on the master mixbus instead.
 

Rex Anderson

Member
Messages
5,066
Start by not leaving the level so low (-6 dB) on your mix. When you mix, do your compression plug in (if necessary) and put a limiter after it. Use the limiter to crank up level and have it peak at -.2 dB. You may need 6-10 dB of limiting.

Waves L3 works well for this.
 

Abram4235

Member
Messages
5,073
I really appreciate the advice so far. I have used eq and compression on most of my tracks including separate BUSs for guitar, bass, drums (no vocals yet). I have not used any plugins at all on my master during the mix. I am running into this problem after rendering into wav. Do you think that's where my issue is? Or should I add compression and limiting during my mix as well as after rendering or does it make a difference?

Also, after rendering, when i reopen my daw, my first track is my rendered wav file of my song and the 2nd track is my comparison track. Should I add compression and limiting to the master track or to the track that my wav file is on?
 

The Funk

Member
Messages
4,533
I really appreciate the advice so far. I have used eq and compression on most of my tracks including separate BUSs for guitar, bass, drums (no vocals yet). I have not used any plugins at all on my master during the mix. I am running into this problem after rendering into wav. Do you think that's where my issue is? Or should I add compression and limiting during my mix as well as after rendering or does it make a difference?

Also, after rendering, when i reopen my daw, my first track is my rendered wav file of my song and the 2nd track is my comparison track. Should I add compression and limiting to the master track or to the track that my wav file is on?
As far as readying a track to be sent to a mastering engineer, you are doing everything right.

Having the levels at -6db is the right thing to do because it ensures headroom. Look at your reference track and your track through a spectrum analyzer. Are the EQ profile's similar? This will have an effect of perceived loudness. Mastering is about EQ and compression. Just slapping a limiter on it is not going to get you there. Also, do you really want to be there? I hate the loudness wars. I like listening to dynamic music loudly, but I have a knob on my stereo for that.
 

Rex Anderson

Member
Messages
5,066
Are you sure you are not doubling the processing on your reference track? You have to turn everything off when you listen to it. Then go back and turn everything on to listen to your track.

6 dB of headroom is a lot. I recorded so individual tracks peaked at -6 dB fs and then mixed so my tracks peaked at -2 or -3. If you need to boost EQ in mastering more than that, you need to remix.

You need to get that gain up close to 0 dB full scale at some point.

Bob Katz (mastering engineer) says to use upwards compression to bring up low level passages in songs. He says you can hear limiting because the ear wants to hear dynamic crescendos not get stopped cold in their tracks.

Use limiting to make the track louder until you can hear it squashing the level as dynamics try to get louder and then back off so you can't hear that happening.

Pro mastering engineers have lots of methods and great tools to get things loud. Read up on the Digital Domain website and buy Bob's book on mastering.
 

Multicellular

Member
Messages
7,880
I really appreciate the advice so far. I have used eq and compression on most of my tracks including separate BUSs for guitar, bass, drums (no vocals yet). I have not used any plugins at all on my master during the mix. I am running into this problem after rendering into wav. Do you think that's where my issue is? Or should I add compression and limiting during my mix as well as after rendering or does it make a difference?

Also, after rendering, when i reopen my daw, my first track is my rendered wav file of my song and the 2nd track is my comparison track. Should I add compression and limiting to the master track or to the track that my wav file is on?
The reasons for running plugins on the master versus after you render are, absent actual errors in your DAW, more work load and processor speed.

Work load i.e.- A lot of people like to do it after render because it is easier then to have multiple tracks to A/B/C/etc. i.e. if you are doing an album. Second, it makes you commit to a mix, which might be good or bad, depending on your psychology in particular how far you go to chase diminishing returns.

So I don't think your issue, based on what you've told us, is using 'mastering' plugins on the master bus versus post-render. Rather it sounds like your gain staging up to the point of mastering isn't enough.

-6 is fine, as @The Funk noted, if you are sending to someone else to master. But it sounds like you are trying to master it yourself.

If so, you are looking for -1 peak, but RMS is really more the target you have to deal with, -1 via brickwall limiter is easy.
 

Abram4235

Member
Messages
5,073
The reasons for running plugins on the master versus after you render are, absent actual errors in your DAW, more work load and processor speed.

Work load i.e.- A lot of people like to do it after render because it is easier then to have multiple tracks to A/B/C/etc. i.e. if you are doing an album. Second, it makes you commit to a mix, which might be good or bad, depending on your psychology in particular how far you go to chase diminishing returns.

So I don't think your issue, based on what you've told us, is using 'mastering' plugins on the master bus versus post-render. Rather it sounds like your gain staging up to the point of mastering isn't enough.

-6 is fine, as @The Funk noted, if you are sending to someone else to master. But it sounds like you are trying to master it yourself.

If so, you are looking for -1 peak, but RMS is really more the target you have to deal with, -1 via brickwall limiter is easy.
Yes exactly, I am trying to master it myself and Im leaving -6db of headroom during my mix before rendering and then trying to boost it again. My individual tracks are on the quiet side though, I think this might be contributing to the loss in volume as you can only increase low output so much. I think I'll go back and look at my mix again and try to increase my individual tracks more, then leave less headroom at my master track. Then render, add compression, a limiter, and see how that works out.

Thanks for all the feedback! I'll report back to let you know if I've figured it out.
 

Rex Anderson

Member
Messages
5,066
Part of your problem may be that your individual tracks were recorded at a low level. That means the mic pre was not hit hard enough and things sound wimpy. Bringing up level after recording does not fix that problem. Record your tracks so they peak at -6 dB FS.

It's hard to get powerful sounding recordings with gear that is not the same as your reference CD's. It's not just loudness level. It's the quality of the whole recording chain (including the studio acoustics) used by pro recording and mastering engineers.

You can't make chicken salad out of chicken sh*t. No offense meant to you, I don't know what gear you have and have not heard your recordings. Maybe your tracks sound great and you just need some help with mastering methodology.
 

Multicellular

Member
Messages
7,880
Yes exactly, I am trying to master it myself and Im leaving -6db of headroom during my mix before rendering and then trying to boost it again. My individual tracks are on the quiet side though, I think this might be contributing to the loss in volume as you can only increase low output so much. I think I'll go back and look at my mix again and try to increase my individual tracks more, then leave less headroom at my master track. Then render, add compression, a limiter, and see how that works out.

Thanks for all the feedback! I'll report back to let you know if I've figured it out.
-6db would be fine peak headroom, as much headroom as digital has. But I suspect it is not peak that is the issue but RMS - average loudness. This "I think this might be contributing to the loss in volume as you can only increase low output so much.." Makes me think you don't understand that. -6 db with a HUGGEE dynamic range could loose or distort things on render, but I've only really heard classical mixes that had that issue in digital.

'Loudness' as perceived is not so simple as peak. Humans generally can compensate for that, it is called the volume knob, button of their listening device.

I could, for example, make 2 tracks that both have -6 peak loudness, one with a high rms, and one with low and you'd perceive the latter as being much louder. Again, same peaks.

That's why, to do it well, you have to do it in stages. Simply slapping a compressor and limiter on, may get you, inelegantly into the territory, but if you are asking these questions, you might be able to learn a lot more and you'll be glad you did.
 

MiguelDamas

Member
Messages
297
Yes exactly, I am trying to master it myself and Im leaving -6db of headroom during my mix before rendering and then trying to boost it again. My individual tracks are on the quiet side though, I think this might be contributing to the loss in volume as you can only increase low output so much. I think I'll go back and look at my mix again and try to increase my individual tracks more, then leave less headroom at my master track. Then render, add compression, a limiter, and see how that works out.

Thanks for all the feedback! I'll report back to let you know if I've figured it out.
Don't bother going back and screwing with your gain staging if you're happy with the mix. Nothing good can come from that. Just normalize the mix to -1dBFS or whatever it is you want to start with. Normalizing will bring the highest peaks to that value and keep the dynamic range (there is no compression or limiting involved, all samples get bumped up by the same amount, just like increasing the track volume).

From there you can start compressing, EQing and limiting but it really is an art and a science. There's a reason we pay people to do this. I'm not saying don't do it, it's fun and a great learning experience but don't expect to get your stuff as loud and as clean as the pros - it's not easy.

Add a couple of reference tracks to your session and make sure you're only adding plugins to your track, not the master bus. Go back and forth and try to match them not only in loudness but frequency balance. If you want to get really loud be aware that low end takes up a lot of energy. If your track is heavy on the low end and your reference tracks aren't you won't be able to get it as loud.

What also helps is A/Bing your mastered track and your unmastered track (volume matched, of course) to make sure you're not completely destroying the mix in your quest for volume.

I've been learning by trial and error and I'm somewhat content with my limited skills but I only master my demos to get a feel for what they would sound like after mastering. I would never master songs for release myself. So yeah, try to be realistic with your expectations and get your hands dirty.
 

Rex Anderson

Member
Messages
5,066
One more thing...

Be sure you are working at 83 dB SPL when you are making EQ and loudness comparisons. Google "Fletcher Munson Equal Loudness Curves". The ear's frequency response is most linear at 83 dB SPL.

Also, be sure to compare you mix to your reference CD's on a lot of different systems, in your car etc.

Normalizing every track to peak level can destroy dynamic range between songs. You don't want a slow mellow ballad as loud as an all out rocker.

Folks keep saying peak at -1 dB FS. That is losing .8 dB in level. The only reason we don't master at 0 dB FS is because some crappy D/A converters will clip and distort. -.1 or to be safe, -.2 dB FS is fine. Squeeze out every last drop....
 

oldhousescott

Gold Supporting Member
Messages
2,736
If you convert to MP3, you'll need more than .2dB headroom to keep from getting "overs". -1dB is not even enough for some MP3 formats. Apple has a tool to let you compare the original WAV file to a converted AAC file which helps determine if you'll clip on the output.
 

Multicellular

Member
Messages
7,880
If you convert to MP3, you'll need more than .2dB headroom to keep from getting "overs". -1dB is not even enough for some MP3 formats. Apple has a tool to let you compare the original WAV file to a converted AAC file which helps determine if you'll clip on the output.
Yep
 

Abram4235

Member
Messages
5,073
So I have figured out that I was not using enough compression really on most of my tracks. Also, I had way too much kick drum and bass frequencies going on when comparing it to my reference track and the plugins were in the wrong order (eq compression) and i rendered at a higher volume. I am now able to get the volume pretty darn close to my reference track. Thanks everyone and I will continue to learn!
 
Last edited:

Rex Anderson

Member
Messages
5,066
If all your tracks need that much compression, you might want to try playing with more even dynamics. As a recording engineer/producer, I often coached musicians to perform with more controlled dynamics to suit the recording/playback medium.

Good compressors used as an enhancement to sound quality is one thing, needing to use a compressor to control dynamic range is something different.
 

Abram4235

Member
Messages
5,073
I'm doing more research on "normalization" as well. The reference tracks waveform is much more full as opposed to my tracks. Additionally, a bit of compression helped my rendered track look more like the reference track (not a lot of compression, just enough to control the peaks). I am noticing though that pro level waveforms appear to have been normalized in order to look and sound like they do. I noticed normalization has been mentioned a few times in this thread. Any thoughts on any of this?
 




Trending Topics

Top