Jump to content

Need infomation: CD duplication and normalization


Recommended Posts

  • Members

Basically ive been mastering some tracks pretty loud, around -13RMS in the chorus's and the compression is unaudiable which is great. Im just worried that to get that volume i had to attenuate the track by around 1.5db essentially droping the peaks too - 1.5db below 0db. This allowed the tracks too not cross over the 0dbfs mark with intersampled meters. Im just affraid that in duplication the track could be normalized back to 0 going over 0dbfs on ADC. Are my worries false or would my track get normalized at some stage?

Link to comment
Share on other sites

  • Members

Just worried me that the work i had greated would ruined. I can imagion some cheap duplication plants doing it. But then again the usually stand alone duplicators so i supose not. Thanks bear

Link to comment
Share on other sites

  • Members

contact your duplication house for exacting specifications for level and overs and whatnot. I forget the details but IIRC it's like they recommend -.03 or so, with ZERO ERRORS... I dunno... they will give you the specs. If they don't, run away very fast!!!

 

They will either "setup" your disc - levels, spaces, order... OR press your disc as you provide it if it's redbook compliant and contains no errors. Anyway, they will give you the detail. I'm rusty on it

Link to comment
Share on other sites

  • Members

 

They will either "setup" your disc - levels, spaces, order... OR press your disc as you provide it if it's redbook compliant and contains no errors.

 

If someone doesn't have a mastering engineer, then most duplication houses do have some type of in-house mastering to handle that for them prior to duplication. But either way, that's NOT part of "duplication" - that's part of mastering.

Link to comment
Share on other sites

  • Members

 

Normalization shouldn't take your disc over 0 dB.....all normalization does is take the highest peaks and put them up to 0 dB. And as mentioned, a duplication house doesn't (or shouldn't) mess with any levels, they just take what you have and duplicate it.

 

 

If controlling a master and you attenuate the peaks lower so that at ADC it doesnt exceed 0dbfs. By normalizing ur raising these peaks essentialy bringing over 0dbfs at ADC after duplication. Hence the question x

Link to comment
Share on other sites

  • Members

 

"By normalizing ur raising these peaks essentialy bringing over 0dbfs at ADC after duplication."


I believe "ur" incorrect, sir.

 

 

If ur final master has its loudest peak at -3db and reachs but does not exceed 0dbfs using intersampled metering. Normalization will bring ur mix over 0dbfs because the peaks before metering were 0dbfs so +3 db over 0dbfs. If you dont know what im talking about do some research into dbfs and intersampled metering. I would love some real help hear as im still devided.

Link to comment
Share on other sites

  • Members

So you cannot explain what you meant by your comment? That normalization wouldnt exceed the 0dbfs mark. If your source material already hits 0dbfs and its peaks are not normalized it will only exceed 0dbfs due to normalization. Explain otherwise?

Link to comment
Share on other sites

  • Members

I have Steinberg Wavelab and i'm still trying to figure out what you mean....anything you master before duplicating should stay the same after you duplicate it. And anything you duplicate yourself--burning a master disc, and/ or a copy of the master should not be different, unless you've got some sort of plug in on (when I first started using Wavelab, if I didn't bypass the plugins, it would apply them twice). Duping houses should not be tampering with any levels. Your worry about levels is a moot point, because if the duping house tampered with any of my levels on releases (which they haven't), i'd go bloody ballistic. It's not their call to make so don't worry about it.

Link to comment
Share on other sites

  • Members

"That normalization wouldnt exceed the 0dbfs mark."

 

Well, it can't put a sample above 0dBFS.

 

But your are right that if you are getting 3dB intersample peaks then normalization could (possibly) result in non-legal signals in the DA.

 

I don't normalize, so it's not an issue for me. And whoever duplicates your material shouldn't be tampering with the master.

Link to comment
Share on other sites

  • Members

Thanks for agreeing - instrospection im pretty tired but here goes. Put a intersample meter on the mix your mastering at the end of ur insert chain. now apply a limiter as ur first step, apply threshold till the metering tells you that you've peaked it. These peaks represents distortions in ADC from ur final product. As i say im pretty tired but look up dbfs and intersampled metering a good one for this example is http://www.macmusic.org/software/view.php/lang/en/id/1260/

 

Hope ive opened you up too some more mastering infomation.

Link to comment
Share on other sites

  • Members

Limiting and Normalization are two completely different things, From the conversation here it appears some dont understand the difference between the two. Someone who only burns CDs like a dupe house doesnt add either.

Consumer CD programs often have a normalization button in the software but this isnt the case at a dupe house.

 

 

If you're talking about Mastering thats a whole different issue. A mastering engineer will use Limiting to achieve the best sound quality possible given your perameters. If you want increased loudness they'll give it to you but they are working for you, You need to communicate what you want.

 

Normalization - How it works is if your highest peak in music is say -2 and the rest of the music is say -12 and you only have the one peak. Normalizing to 0 will only increase the level of the of the entire selection 2 db. The one peak will be adjusted to 0DB and the rest of the music will be adjusted from -12 to -10.

 

Normalization can only be set to zero or less and looks to the peaks for info in making adjustments. Once it finds those peaks it adjusts everything up the same amount. It has no affect on dynamics at all. It may or may not make anything louder depending on where the highest peaks read.

 

Limiter - A limiter is very different. A limiter has a ceiling that can be set and the volume of the music below that peak can be adjusted up until theres no peaks left.

 

To be used proprely you have to first determine the RMS or average music level you currently have. If the RMS is say -12 and the threshold of the limiter is set to -12, You will basically flaten all dynamics out of the music bringing all peaks up to 0db. Anywhere from -12 there to zero there will be a lower percentage of peak flatening that occurs depending on the dynamics and threshold once the threshold level is crossed there should be no affect on the music. This is a key item because music with littel dynamics like say very dence rock music, will have a very sharp threshold of say 1 or 2db between no limiting and full limiting. Music that is very dynamic may have a very wide threshold of 4 or 5db or more between zero and full limiting

 

 

In Weemans case, He says his peaks reach zero. A limiter is usually used after a mix set to -.01 or -.02 so no peaks can reach zero and interfere with the burning process, You can use a limiter a limiter threshold to increase volume at the sacrifice of dynamics but it doesnt mean you Have to.

 

In either case a limiter Should be used as a safety factor to prevent bad CD burns caused by random peaks. You Can set a Limiter out cieling to Zero if you feel brave and want to risk bad burns but the difference between zero and -.01 cannot be heard by the ear.

With a threshold set to zero and output ceiling set to -.01 you probibly wont even know its been run, but the burning will not have errors created by stray peaks and playback should be jitter free as well.

Link to comment
Share on other sites

  • Members

 

Limiting and Normalization are two completely different things, From the conversation here it appears some dont understand the difference between the two. Someone who only burns CDs like a dupe house doesnt add either.

Consumer CD programs often have a normalization button in the software but this isnt the case at a dupe house.

 

 

That's been my point--normalization takes everything up to 0 dB. It shouldn't exceed it. If in doubt, put a brickwall limiter set to 0 dB on soft knee, it won't exceed 0 dB. I just put the files in the mastering software, do some EQ-ing and some limiting/ brickwall, and then render it. I've never had any problems with any overs.

 

 

If you're talking about Mastering thats a whole different issue. A mastering engineer will use Limiting to achieve the best sound quality possible given your perameters. If you want increased loudness they'll give it to you but they are working for you, You need to communicate what you want.


Normalization - How it works is if your highest peak in music is say -2 and the rest of the music is say -12 and you only have the one peak. Normalizing to 0 will only increase the level of the of the entire selection 2 db. The one peak will be adjusted to 0DB and the rest of the music will be adjusted from -12 to -10.


Normalization can only be set to zero or less and looks to the peaks for info in making adjustments. Once it finds those peaks it adjusts everything up the same amount. It has no affect on dynamics at all. It may or may not make anything louder depending on where the highest peaks read.


Limiter - A limiter is very different. A limiter has a ceiling that can be set and the volume of the music below that peak can be adjusted up until theres no peaks left.


To be used proprely you have to first determine the RMS or average music level you currently have. If the RMS is say -12 and the threshold of the limiter is set to -12, You will basically flaten all dynamics out of the music bringing all peaks up to 0db. Anywhere from -12 there to zero there will be a lower percentage of peak flatening that occurs depending on the dynamics and threshold once the threshold level is crossed there should be no affect on the music. This is a key item because music with littel dynamics like say very dence rock music, will have a very sharp threshold of say 1 or 2db between no limiting and full limiting. Music that is very dynamic may have a very wide threshold of 4 or 5db or more between zero and full limiting



In Weemans case, He says his peaks reach zero. A limiter is usually used after a mix set to -.01 or -.02 so no peaks can reach zero and interfere with the burning process, You can use a limiter a limiter threshold to increase volume at the sacrifice of dynamics but it doesnt mean you Have to.


In either case a limiter Should be used as a safety factor to prevent bad CD burns caused by random peaks. You Can set a Limiter out cieling to Zero if you feel brave and want to risk bad burns but the difference between zero and -.01 cannot be heard by the ear.

With a threshold set to zero and output ceiling set to -.01 you probibly wont even know its been run, but the burning will not have errors created by stray peaks and playback should be jitter free as well.

 

 

All good info. I'm still baffled by weeman's mentions of intersamples and whatnot amidst the liberal spelling errors and lack of clarification in which no one can give a proper answer because of the hard to understand nature of what he's asking...in understandable English for an answer, all you need to do is put a brickwall limiter on it with the threshold of 0 dB, and it won't exceed it. Normalization, in any case, shouldn't push any peaks over 0 dB. Regular limiting and compression, yes, you have to watch the monitors, but brickwall, no, and normalization no. Forget about any intersamples and EQ or whatever....the bottom line is that if you put a brickwall limiter on it at 0 dB and render it, you're going to be safe. If you're worried about clipping and overs with regular limiters and/ or compressors, you need to watch the meters and there's no other substitute for that. If the problem is in the rendering, it only renders what is going to be on the meters.....if it's giving you something other than what's on the meters, you have a problem with your mastering equipment or are doing something wrong with your plugins.

Link to comment
Share on other sites

  • 1 month later...
  • 2 weeks later...
  • Members

Those quick spikes are your enemy if you choose to 'normalize.'

That's why you should check for peaks in the audio file and deal with them individually.

 

Why would you want to normalize the audio? If you think you have a perfect mix, then leave it alone.

 

If I wanted a specific part to be at -12 dB and I don't know any better, I would use the normalize function because I read somewhere that I should. Now that dB level has been raised up beyond where I wanted it.

 

If My final mix is perfect and I choose to use the normalize function because I read somewhere that I should, then I will most likely ruin my perfect mix.

 

So it is not simply a matter of 'Oh I will just normalize it and everything will be alright' because in most cases you do not need to use that function.

 

If you know how to mix properly and look for those sneaky little spikes, you will do well to invest the time. Tossing a limiter on is another weak link in a chain with the possibility of adding more problems including phasing. The spike may be to fast for the limiter to detect and stop.

 

Learn how to mix properly! Substituting some trick you read about isn't going to make up for your lack of skills. They usually do more damage than you bargained for.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...