Assorted Mastering Questions (Resampling, Compression, Gain Increases, etc.)

[Confusitron]

I am nearly to the point of having some songs I've recorded fully mixed and will be entering the mastering process. I have a few mastering questions. I am using Cubase SX.

1. For saving a mixdown, should I be saving the audio in its native sample rate (I am using 16 bit 48 kHz audio) and then resampling to 44.1 kHz later, or should it be saved as 44.1 kHz audio in the first place?

2. Once I am mastering the final mixdown, should I apply fades to the beginnings and endings of track after gain increases have been made? I see that a gain increase could possibly "expose" the actual end of a fade if it were applied before gain increases were made.

3. Should I apply compression to the entire completed track at say a 2:1 ratio? What kind of attack and decay should I use? Should I have the compressor predict the attack and decay times?

4. How can I avoid distortion in the audio even if there is no clipping occuring (as a limiter is used)? I have had trouble with this in a past recording and need to completely avoid it. It happened (of course) at the loudest section of a song. Would the compression assist in avoiding this?


Thank you.

 

[Cyanide-Depende]

I'm not a mastering pro, but I'm fortunate to know one the the best ssl mixers/mstering guru's around. This was the advice he gave me when I has simlar questions, (although he is going to master my final mix, not me)

Hope this helps

I am nearly to the point of having some songs I've recorded fully mixed and will be entering the mastering process. I have a few mastering questions. I am using Cubase SX.

1. For saving a mixdown, should I be saving the audio in it's native sample rate (I am using 16 bit 48 kHz audio) and then resampling to 44.1 kHz later, or should it be saved as 44.1 kHz audio in the first place?

I've been told record, sequence, sample everything at the highest resolution you can (I do EVERYTHING at 24bit 48kHz)

3. Should I apply compression to the entire completed track at say a 2:1 ratio? What kind of attack and decay should I use? Should I have the compressor predict the attack and decay times?

Depends on the sound you're looking to get, from my understading. I would start there and tweak untill you like the sound, then get a profefessional opinion.

Thank you.

You're Welcome, lol, I tried

P.S.
Sorry I couldn't really help with the other questions, I didnt want to bullshit you, but I'm sure someone else can help with those as well

 

[Confusitron]

I've been told record, sequence, sample everything at the highest resolution you can (I do EVERYTHING at 24bit 48kHz)

Well, I guess I would have done that IF everything was not all ready recorded at 48 kHz 16 bit audio. What I was saying is, since the tracks will be resampled to 44.1 kHz 16 bit audio once they are put on CDs, should I master the tracks at 48 kHz 16 bit audio and THEN resample them to 44.1 kHz 16 bit audio, or should I make my final mixes into 44.1 kHz 16 bit audio FIRST and then master them?

 

[Cyanide-Depende]

Id master them at 48kHz then, I was told to have everythign at the highest quality level possible for mastering, and downsample it Only when you finally put it on the cd

 

[mshilarious]

Very generally, I would, probably in this order:

Edits (de-noise, de-click, etc.)
EQ
Compress (if necessary)
Limit
Sample Rate Conversion
Dither (which you don't need)
Apply fades

That basic order could change if required. For example, you might want post-dynamic processing EQ, although I always put the limiter last to make sure EQ isn't causing overs.

I could see an argument for doing fades earlier, but I can see arguments against too. Unless you are doing a long fadeout, it happens so fast it probably doesn't matter too much.

I dislike long fadeouts anyway, I think musicians should only be allowed one fadeout for every fadein! If they can figure out how to start a song, they should be able to figure out how to stop it

Should you compress, and if so, at 2:1? Totally dependent on the material. Also a compressor at this stage might be selected more for its "color" rather than a pressing need for compression.

 

[masteringhouse]

Dither should be last in the process, after fades assuming that you are fading a 24 bit file.

 

[GreenDank]

Should I apply compression to the entire completed track at say a 2:1 ratio? What kind of attack and decay should I use? Should I have the compressor predict the attack and decay times?

I don't even goof with attempting to compress/expand on my computer. It only seems to change everything I attempted to tweak to begin with. Perhaps you'll have better results, but it really kills a lot of the dynamics. Anymore, I just amplify the track to as loud as it'll go without distorting, and go with that. No, the end product doesn't sound like my favorite band's latest CD, but neither will will my mastering job, unfortunately. It requires too much high end gear and godlike ears.

 

[LeeRosario]

I myself am not a mastering engineer by trade, but from those I have worked with...

1. For saving a mixdown, should I be saving the audio in its native sample rate (I am using 16 bit 48 kHz audio) and then resampling to 44.1 kHz later, or should it be saved as 44.1 kHz audio in the first place?

2. Once I am mastering the final mixdown, should I apply fades to the beginnings and endings of track after gain increases have been made? I see that a gain increase could possibly "expose" the actual end of a fade if it were applied before gain increases were made.

3. Should I apply compression to the entire completed track at say a 2:1 ratio? What kind of attack and decay should I use? Should I have the compressor predict the attack and decay times?

4. How can I avoid distortion in the audio even if there is no clipping occuring (as a limiter is used)? I have had trouble with this in a past recording and need to completely avoid it. It happened (of course) at the loudest section of a song. Would the compression assist in avoiding this?


Thank you.

1. You save in whatever samplerate and wordlength you work in. So if your project was in 16/48khz, then you save in that format.

The main reason being that a well equipped mastering house has better means of doing conversions. Downsampling and upsampling at home is a good way to deteriorate the quality of your mix. In this way, you can also avoid dither at home since no conversion was made.

However, if you do this at home, you can treat it the same way. Export the mix at the session's resolution and save conversions and dither for mastering.


2. Fades generally are a last process thing. In compressing and processing tracks, you might of brought up low level noise that will need to be tied off with fade ins and fade outs.


3. Compression on entire mix (and at all for that matter) is always situational. You might not have to compress easy listening material like you would with rock material. These days, it almost seems common to apply moderate compression in the chain to rock mixes, while jazz and the like continues light.

I would only advise to start with high thresholds, low ratios and long attack/release times (unevasive settings) and work your way down from there.

4. To avoid distortion,watch your input levels to plugins. Your end result might not be clipping, but if you hear clipping, chances are you have clipped one or a few of your plug-ins at some point in the production of your audio.

If you work out head room in your plug-ins/outboard gear, then you should be clear. Later on you can rely on maximizers (like an L2) to bring back some volume.

 

[mshilarious]

Dither should be last in the process, after fades assuming that you are fading a 24 bit file.

See that's what I meant. In theory as soon as you fade 6dB you'd wipe away the dither otherwise. On the other hand, the signal-dither ratio drops pretty sharply, and that used to annoy me because I could start to hear the dither on the end of the fade, if for some reason I listened to the fade at a high volume. Not that I do that, but maybe somebody will.

Since I started used the Apogee dither, I can't hear it at all (sometimes I suspect it doesn't actually do anything ), so it doesn't matter to me. I really try not to obsess about this stuff.

 

[masteringhouse]

See that's what I meant. In theory as soon as you fade 6dB you'd wipe away the dither otherwise.

If you dither before adding fades. That's why you want to dither after the fade, when reducing the wordlength from 24 to 16bits.

Since I started used the Apogee dither, I can't hear it at all (sometimes I suspect it doesn't actually do anything ),

Like compression (unless used for effect) the best dither is the kind that you can't hear. So that's a good thing. I would think if you removed it when reducing wordlength it would be something that you can hear though (grainier sounding).