• Welcome to the General Discussion forum for UAD users!

    Please note that this forum is user-run, although we're thrilled to have so much contribution from Drew, Will, and other UA folks!

    Feel free to discuss both UAD and non-UAD related subjects!

    1) Please do not post technical issues here. Please use our UAD Support Forums instead.

    2) Please do not post complaints here. Use the Unrest Forum instead. They have no place in the the General Discussion forum.

    Threads posted in the wrong forum will be moved, so if you don't see your thread here anymore, please look in the correct forum.

    Lastly, please be respectful.

How do you get constant 'loudness' across a whole CD

Barney

Active Member
Hi,

A Basic question for y'all - I am coming to the conclusion that i do not understand dbs as much as I think I do, particularly in terms of 'mastering'. Put in quotes because of the ongoing discussions on this forum.

My question stems from trying to finalise a ten-track CD to give to my mates. I have spent a lot of time mixing and now have applied the Precision Limiter to the end of the chain. I have bus-compressed gently pre-limiter. Because of newness to these tools, I have used presets such as STD A, Happy Medium etc, across the ten tracks, using my ears. Music is retro-style acid-house cum Balearics, all a bit 90s....

In all cases the master level in Cubase is around the same - very little variation.

So, the net result, instead of having ten tracks that sound approx the same volume, is that there are differences. I can hear the groans from those in the know already. My point being, if the Limiter says 'peak at -0.1 db', and in theory makes everything a bit louder and seemingly more compressed as well (hey, you can hear it when you switch it on) what is the interim step I am missing to get 10 tracks of similar apparent loudness.

Apologies for such a rudimentary question,

Thanks,
 

Cass Anawaty

Shareholder
You have to adjust the level of the individual tracks--it's not an automatic process. Well, when everything is squashed beyond belief, I suppose it is. :lol:

You can use RMS meters to get you close, but it's really finalized by ear.
 

Never1

Member
I think you're mistaking peak levels with steady state RMS type levels. A soft song could have peaks that are louder than a rock song, but the rock mix will sound louder, by virtue of the rms levels that are riding constantly higher than on the soft acoustic tune.

When mastering (or leveling in this case) you want to treat each song appropriately. Compress and limit til it sounds right and never go further. Once you've got 'em all finished, you need to line em up and compare them. if you want them all to be equal, the disc can't be any louder than your softest track. But, it's better, IMO, to have some differences here and there. Let the soft be softer and the loud be louder, as long as the differences are only a few db apart. It gives contrast and a sense of flow and dynamics to the album.
 

Barney

Active Member
Folks... thanks for the replies, and I do agree with most of the points made... re 'Compress and limit til it sounds right and never go further' - that is just what I have done for each individual song. I am getting to know things like RMS and peaks.

I may also choose to have a quiet track, but would like to know what to do if I choose not to.

So let's say I have compressed and then limited four songs, in the same genre. Further to that, these four are sonically pretty similar, and I have checked on a spectrum analyser that there are no sonic 'holes'. My challenge is that I have done pretty much the same thing to all four, and I don't know what the missing link is. As one is quieter, so therefore needs to be made louder, I need to work out what to do as next step.
Given that the PL will peak at -0.1 db, and this is on already-compressed material, is it a case that maybe I need to crank up the gain into the PL (ie not use pre-sets) to get higher RMS level / volume? Or do I increase the output from the compressor? Or am I missing another step to increase volume on the quiet track.

Sorry if this is all obvious - however as I learn more I find the more I need to learn more. Happy to use my ears if I can be given a start point.

Thanks guys,
 
I had the same problem as well. After i started mixing and mastering my music using the K-System from Bob Katz, the difference in loudness between tracks issue disappeared instantly. You need to read though to understand how this system works, and you'll need a db meter to calibrate your monitors accordingly, but it's definitely worth it.
 

Fundy

Established Member
The K-System is pretty staright-forward though.

You have a 60 decibel range between the 0dB unity gain point.

K-20 = -40 to 20dB
K-14 = -44 to 14dB
K-12 = -48 to 12dB

The 0 point on this scale midway through is your crest level for the chosen reference point, therefore K-14 is -14dB RMS at this point. This allows for consitency across the board. To ensure you are averaging about this point aim for 2 - 3 decibel gain reduction on the maximiser/limiter.
 

LFranco

Venerated Member
If you're new to "mastering" (love the quotes), I would suggest you get your best mixdown and gently use a limiter to get a little bit of loudness, stop when you think you're screwing up the mix and squashing it too much and then back off some more.

If you have other mixes in your project that just sound louder (due to higher average levels from those mixes), instead of making the previous mix(es) sound as loud as this one, bring this one's levels down and match the "nicey-nice-sounds/feels-great" mix, and apply a limiter to this one also, just so that it stays in the same ballpark.

Sure, your CD won't sound as loud as the latest, slammed-to-death release, but this is a good thing!! I believe in 10 years or so, people are going to stop acting like chimps and we will start seeing lower overeall levels in releases.

Making all of the songs in a project sound as loud as the loudest loud mix in the project is a tough thing to do; Mastering your own stuff, in general is like the saying "shooting yourself in the foot", but trying to make everything loud, and having all of the other songs follow (in loudness)is like that too, but more like "blasting your legs off below the knees with a mossberg".
 

Barney

Active Member
Folks,

Thanks so much for all the replies..... hey, ya never had to woory about this when recording was all on an Atari.... in my case then direct to cassette.

Another question to help me piece together the missing link.

Is 0db, really 85db SPL? ie the term db can be either relative or absolute? There the Katz method is a way of assigning the 0bd 'relative pointer' to an absolute level?

thanks,
 

Paradigm X

Established Member
Barney said:
Is 0db, really 85db SPL? ie the term db can be either relative or absolute? There the Katz method is a way of assigning the 0bd 'relative pointer' to an absolute level?
thanks,

ALL dBs are relative. Its just a log of a ratio.

the 85dB SPL you are talking about is referenced to 0.00002 Pa (Pascals pressure).

0dB on cubase etc is relative to the maximum level (digitally). IE just before clipping.

Therefore, you set the 0dB on the digital side to equate to a known level 'in the real world', using your speakers and room acoustics.

ive not read a huge amount about the k-system but makes sense (im an acoustician by trade). Been meaning to look into it.

cheers
 

Eddie Macarthur

Active Member
i use sony cd architect for cd master preparation. it lets me use plugins on each clip non-destructively.

i generally use izotope ozone in clips that i feel need adjustment, sometimes with a limiter over the master buss to avoid clipping.
then i quickly play a part of each file and jump randomly to another part of another track.
i listen to a few seconds of each track to estimate loudness. i go through this a few times and adjust as required.
then i burn a cd and listen through it, making notes.
then i adjust settings as required and try again.
it's something you have to do by ear, not by reading meters... but of course others will disagree!
ed
 

Barney

Active Member
Eddie - thanks - that's a very good shortcut... I now bring up the tracklist into MediaPlayer, and skip thru the tracks to the point in each (measured as seconds from start) that is the loudest. Quick notes taken, then back to re-burn hopefully not too many tracks.

It's not foolproof, but it does save the 'burn to CD, listen in car, listen in lounge-room' syndrome that takes up the hours.

And thanks everyone else for the replies.
 

Paradigm X

Established Member
You can find the rms level in wavelab and prob many other tools.

In WL, go to Analysis, Global analyise (or just press 'y') and then click analyse (or press enter)

The second tab shows you the rms value.

If you had two tunes at -12dB and -15dB, youd have to reduce the louder one (-12) to meet the same apparent loudness. A 3dB cut would be a good starting point...

The quietest tune will have to be the loudest overall, and all others brought down to meet it, as the peaks of all the tracks will be at 0dB (peak), and cant be turned up any more.

I find a good way to do it is to analyse them all and note the rms's , then import into cubase to arrange and adjust the volume of each track. then quickly jump between all the tracks at random and adjust to taste. sometimes some eq also helps. but thats just a diy home studio job, sure there are other ways.
 

Fundy

Established Member
Har-Bal is a good analysis tool which allows you shape the dynamics and harmonics of the input track and a reference track. Although I have found using another limiter works better than built-in one.
 
UAD Bundle Month
Top