Simple Loudness Question

[nick_w79]

Hello,

This has probably been asked a million times before in several ways but I'm not sure if I'missing something simple.

I understand the loudness thing is old news but I have just mixed a demo for my mate's band but when my mate played it on his stereo he claims he had to set the volume twice as loud as normal.

When I mastered it I just did some light limiting to 0dB in Cubase 5 and it sounded pretty sweet but when burnt to CD and played on a stereo I guess the level is really low.

The reason I ask this is not because I need a lesson in how to make my masters super loud cos I'm not overly fussed about that but because I think I may have lost some level in the process somewhere.

I exported the mixdowns from Cubase as wavs and then had to import into iTunes to burn to CD as I had no other option. Could I have lost some level there?

Thanks for any help.

Nick

 

[Massive Master]

Simple Loudness Question

No such thing.

 

[moresound]

Listen to your mixes on other stereos and such would tell you how your mix sounds like in the real world out side of the studeo.
If it's not loud enough out of the studio....turn it up the final mix then burn again.

 

[nick_w79]

If it's not loud enough out of the studio....turn it up the final mix then burn again.

I've limited to 0dB so I don't see how I can turn it up. That's what I mean. I didn't know if it could be something after the final mixdown that could cause it to lose some of the level

 

[moresound]

Are you employing compression at all? If so it may be to much or you need to bring up the make up gain of it . For any other plug in you may have in the chain check all of them.

 

[nick_w79]

There's no compression in there but it is limited to 0dB so I don't see that the level should be as low as it is.

 

[mattr]

You keep on saying "limited to 0db" but this doesn't really mean much at all, other than implying a peak ceiling of 0dB... it says nothing in terms of the actual perceived loudness, RMS level, etc... you could have something "limited to 0dB" which only peaks once or twice during the song with hardly any action from the limiter and an average RMS of around -20 dBFS, or you could have something "limited to 0dB" which is smushed right into the limiter with tonnes of gain reduction and up to around -8 dBFS RMS.

 

[bouldersoundguy]

I've limited to 0dB so I don't see how I can turn it up. That's what I mean. I didn't know if it could be something after the final mixdown that could cause it to lose some of the level

You need to learn about a few things:

1. decibels full scale
2. peak versus RMS
3. operation of a mastering limiter

In the digital world levels are described in dBFS. The number will always be a negative as it refers to the loudest level in digital, 0dBFS.

Peak level and RMS are different ways to describe loudness. Peak is the literal highest level of a signal, or the furthest part of the waveform from the center line. RMS (root mean square) is a type of averaging that is more like how we hear loudness. A peak can be quite high but contribute little to the RMS level of a sound.

You can't make the peaks higher than 0dBFS, in fact you should probably make them just slightly lower, perhaps -0.1 to -0.3dBFS. So a limiter is used to bring down the peaks so the overall level can be raised without clipping, thus raising the RMS level. This is more or less how to make a recording sound "loud" within the limits of digital recording. You can make it sound as loud as you want relative to other recordings, but in doing so you sacrifice the natural peaks which give the recording much of its impact. My typical target RMS level is -16dBFS. I've heard commercial music recordings with RMS levels as high as -11dBFS. Resist the urge to push your RMS levels too high or you'll lose that impact that comes from the peaks.

Some limiters work in the conventional way, lowering a threshold to reduce the peaks, then raising the level with a gain control. Others, often specifically mastering limiters, either have automatic gain compensation or have a gain control driving into a fixed threshold.

 

[nick_w79]

You keep on saying "limited to 0db" but this doesn't really mean much at all, other than implying a peak ceiling of 0db... it says nothing in terms of the actual perceived loudness, RMS level, etc... you could have something "limited to 0db" which only peaks once or twice during the song and an average RMS of around -20 dbfs, or you could have something "limited to 0db" which is smushed right into the limited up to -6 dbfs RMS.

I see what you mean. I don't know what the average RMS but I get about 3 - 6 dB of reduction throughout most of the track with a bit more on some of the higher peaks and from looking at the output meter (bad I know) the output doesn't seem to drop too far from 0dB at any point and just from this I expected the level on CD to be a bit higher than it apparently is.

 

[nick_w79]

I think perhaps I need to learn how to compress and limit in a mastering sense. I tried to compress but I found I lost a lot of dynamic range (like bouldersoundguy says) so I abandoned it and just used the limiter which gave a better sound.

Maybe I was to heavy handed with the compression and should try again.

I was just concerned that my problem might not have been as simple as the perceived loudness from the master but a problem elsewhere.

I know loudness isn't the most important thing but I just find it annoying when people you mix for say "it isn't very loud is it?"

 

[mattr]

RMS = Root Mean Square, i.e. the square root of the average of the squares of the peak values (well, its a bit more complex if you're finding it of a wave). Basically its another type of average level that is reference to a lot in audio. An average level tells you a lot more about the overall 'loudness' than a peak level.

 

[newcomputeruser]

loudness wars

Hello,

This has probably been asked a million times before in several ways but I'm not sure if I'missing something simple.

I understand the loudness thing is old news but I have just mixed a demo for my mate's band but when my mate played it on his stereo he claims he had to set the volume twice as loud as normal.

When I mastered it I just did some light limiting to 0dB in Cubase 5 and it sounded pretty sweet but when burnt to CD and played on a stereo I guess the level is really low.

The reason I ask this is not because I need a lesson in how to make my masters super loud cos I'm not overly fussed about that but because I think I may have lost some level in the process somewhere.

I exported the mixdowns from Cubase as wavs and then had to import into iTunes to burn to CD as I had no other option. Could I have lost some level there?

Thanks for any help.

Nick

read about loudness wars at wikipedia

limiting to 0 db does not mean it will be loud
just that it does not clip

compress the entire signal into a much smaller db range limited by 0dBFS and you will sound a lot louder still

 

[bouldersoundguy]

read about loudness wars at wikipedia

limiting to 0 db does not mean it will be loud
just that it does not clip

compress the entire signal into a much smaller db range limited by 0dBFS and you will sound a lot louder still

Search Wikipedia for the term "intersample peaks".

 

[newcomputeruser]

Search Wikipedia for the term "intersample peaks".

nonsense -- according to my graduate digital signal textbooks

at best somebody is playing semantic games
nothing can clip in the d-->a conversion
can a bad d/a conversion algorithm/design cause problems
you betcha but not "clipping"

clipping occurs in the d/a when the analog goes higher than whatever 0dBFS is set to

dont believe everything you see in wikipedia

 

[philbagg]

nonsense -- according to my graduate digital signal textbooks

at best somebody is playing semantic games
nothing can clip in the d-->a conversion
can a bad d/a conversion algorithm/design cause problems
you betcha but not "clipping"

clipping occurs in the d/a when the analog goes higher than whatever 0dBFS is set to

dont believe everything you see in wikipedia

I'd trust Uncyclopedia more than you.

 

[mattr]

nonsense -- according to my graduate digital signal textbooks

at best somebody is playing semantic games
nothing can clip in the d-->a conversion
can a bad d/a conversion algorithm/design cause problems
you betcha but not "clipping"

clipping occurs in the d/a when the analog goes higher than whatever 0dBFS is set to

dont believe everything you see in wikipedia

LMAO, I have to say you are the biggest asshat we've had on this forum in at least a few months. Congratulations

I'd be quite intrigued if you posted the names of these books, and whether they actually explicitly state that intersample peaks don't exist, or just they don't go into sufficient detail to mention them.

Intersample peaks are just a universally accepted side-effect of the interpolation used in the D/A conversion, which is far from a simple 'join-the-dots' exercise. Well, its only the peaks we worry about as they can be troublesome, but the same thing happens with virtually everything passing through the converter and is the whole basis of the reconstruction - remember, the sampling during AD conversion occurs at a regular rate and wherever they may fall on the waveform at that time, not at nice convenient locations at the peaks and troughs of a nice uniform sine wave like your high school textbooks might make out.

If you're really that adverse to trusting the content of that Wikipedia article, I would kindly suggest getting your head out your backside and having a quick Google on the subject, which would reveal hundreds of independently written articles, papers, etc, on the subject of wave reconstruction (and probably some written by the very people who actually design these DA converters).

And I have to say, you've been giving people a lot of mouth and been full of your own knowledge and abilities, yet I haven't seen you post any of your own music/recordings yet. How about you post a song or two of your own so we can see how great you are and give you the praise you're craving so much?

 

[dintymoore]

re: volume drop in Cubase

You might want to check the mixers in Cubase to make sure that, when you're exporting a master, you aren't lowering the volume there. I run all my outputs at zero on Cubase.

If you are going through a soundcard or other other audio interface and it has a mixer section, check the output level settings there. I use an M-Audio Delta 1010LT for instance, and in it's the control panel I set the channel outputs to +4.

If you're using a Mac it wouldn't hurt to check the Audio Midi Setup in the Utilities folder to see if there's any suspect settings there.

 

[nick_w79]

You might want to check the mixers in Cubase to make sure that, when you're exporting a master, you aren't lowering the volume there. I run all my outputs at zero on Cubase.

If you are going through a soundcard or other other audio interface and it has a mixer section, check the output level settings there. I use an M-Audio Delta 1010LT for instance, and in it's the control panel I set the channel outputs to +4.

If you're using a Mac it wouldn't hurt to check the Audio Midi Setup in the Utilities folder to see if there's any suspect settings there.

I just use the Laptop's internal soundcard at the minute so it might be something there. Will have a look (as much as my tech knowledge allows!)

I'd like an audio interface but can't afford one at the minute.

 

[moresound]

HEY mattr is that a dot picture at your signature line of your post,if so what the heck is it? I can't make it out.

Nice work on explaining intersample peaks. And if your age in your profile is correct how the heck did you get so smart in 18 years?
Keep up the good work of helping out newbies and trying to reeducate lost souls.

 

[philbagg]

And I have to say, you've been giving people a lot of mouth and been full of your own knowledge and abilities, yet I haven't seen you post any of your own music/recordings yet. How about you post a song or two of your own so we can see how great you are and give you the praise you're craving so much?

He doesn't have time/doesn't need the praise. He's a brain surgeon