How Does My Mastering Compare to Yours

[Calvin]

Hey guys,

I recently had a couple people ask me about comparing notes on our mastering techniques. Since I know that this is a topic that lots of us like to find information about I decided to make this a very large session for comparing notes and results. I decided that I'm going to post a before and after of a simple accoustic song on my website for you to download and hopefully compare to your own mastering technique. I would love to hear all coments and criticisms if you have them.

A small note: We all know that for best results we should take our music to a real mastering house, this thread is just to help perfect our own technique. Please save those type of comments for other threads.

For you to listen to I will be posting a little accoustic song that I recorded that is simply accoustic guitars and vocals. I know that this isn't a good representation of all kinds of music but for the sake of listening for quality, I wanted something that had pauses and lots of transients. Since we are talking mastering I wanted to try to keep some sort of quality so all of my mp3's have been encoded at 192k at the highest quality in wavelab. If you also would like to post your version of mastering from the original but don't have space please e-mail me, I can put it up on my server for you to link to.

Here is a link to my original mix.

Here is a link to my mastered version. It was mastered completely in pro-tools using the waves mastering bundle.

In a few minutes I will be posting a very lengthy post about how I mastered it, with pictures. Once again I appreciate all comments and I'm doing this so we can all learn a little.

[Calvin]

I always start a session with 2 stereo tracks and a master fader. I like to use a technique that some call the NY compression technique. The idea behind this technique is that if you use one track to do all the processing (compression and EQ), and one track you leave unprocessed, when you mix them together you get the best of both worlds. The processed track is good for using a lot of compression and EQ on to even things out and maybe highlight a certain part of the track that was a little under mixed. Granted this is no way to make up for a bad mix, but it does help a little. The unprocessed track always sounds the most natural because you haven’t gone threw and killed all the transients with compression. So when you mix a little of the original track back in to the compressed and EQed track you get a track that still has transients with a little bit more natural sound, but is very even and full.

When using this technique in mastering we don’t want to use it to nearly as high of a degree of processing as you would in mixing however the same concept still applies. Here is a picture of the Mix view of my mastering session. As you can see on one of my stereo tracks I first put the Multiband Compressor then the Linear Phase EQ. I usually put the Multiband Compressor first because the entire point of this track is to even things out, and that is usually best done when you put compression first, then EQ.

At this point I usually solo out the track that I am doing all the processing on to start with my adjustments there. The first thing you need to do is to make sure that none of your mastering plug-ins you have on your track have dither enabled. Since dither is a process of cutting off those last insignificant bits to bring it down to 16 bit, we don’t want to do that till the very end. Usually the first thing that I do with the Multiband Compressor is to set the compression ratio down in the lower right hand corner to “Hard”. I do this once again because I am just trying to get rid of the worst transients however this is changed depending on the music and my taste at that moment in time. I then go threw and set the threshold of each band to be just under the maximum level for that band. You can see that Here. You can also do some wider changes to EQ here if you like.

Next I move on to the Linear Phase EQ. Once again make sure that you have the dither turned off. If you don’t you will immediately hear the difference when you turn it off. This plug is one of the easiest to use, just EQ to taste. Here is a picture of my EQ for this track.

Now your ready to bring back the original track in the mix. The first thing you will notice is that the latency from the mastering plug-ins is huge and causes a very bad echo. To adjust for the latency caused by the plug-ins you need to go to very bottom of the processing track in the Mix View. Down there at the bottom of the track you will see where it normally tells you how high you have your fader set. By holding down the “Ctrl” button and clicking on that number, you will get it to cycle threw three different values. The fader level, the peak level of the track, and the latency on the track caused by plug-ins. As you can see in my Mix View, my latency is 6519 samples. This value seems to change sometimes, I don’t really know why, so you should check this every time. Now switch over to your Edit View and above your tracks you can select your nudge value. Pull this down so that it displays samples and set it to your latency value. For me this is 6519. Then simply select the region on the unprocessed track and nudge it backwards in time by pressing the “+” key on you number pad.

Now you are ready to mix your two tracks to your delight. I usually end up with the unprocessed track slightly higher than the processed track, however on this song I actually liked it the other way around.

Wow this is getting a lot longer than I anticipated, sorry guys, but we’re almost done.

Now on to the master fader. The first plug-in I put on the master fader is the digi 1 band EQ. I use the High Pass setting to cut off everything below about 50 hz. While some clients will swear that they need this, you really don’t. Anything that goes on below 50 hz usually has transients that you can hear in the higher ranges. Also when you put your disk into a small to medium sized stereo system, they will actually sound worse because they are struggling to produce those low frequencies and taking away from the rest of the sound. Don’t worry, most stereos and have large low end boosts in them anyway so there isn’t a thing to worry about.

Next on the master fader goes the L2. The L2 is an amazing limiter that can be transparent or add that compression sound that some really like. I personally like to keep it transparent so I try to keep the attenuation dial below 6db. Anything over 6db becomes pretty obvious to my ears. The output threshold of the L2 needs to be set to .02 at least. This will keep your CD player from peaking when playing back your CD. If your going to use the dither on the L2 plug-in, you should have it set to 16bit and the shaping should be set to “Ultra”. I can’t hear much of a difference between Type 1 dither and Type 2 dither so I don’t bother with it much. You want the shaping set to “Ultra” because it is a mastering type dither much like the apogee uv22. Only in mastering do you ever want to use the ultra setting because if you use it to much the noise that it makes can become obvious to the listener.

I personally think that the Pow-r Dither sounds the best so I make sure that the dither on the L2 is turned off. You know for sure it’s turned off when the IDR sign dims. Also if you are using the Pow-r dither instead of the dither on the L2, it’s probable a good idea to set the out ceiling on L2 to .03 or even .04. I don’t know why but the Pow-r dither will on occasion peak out if it is set any higher. On the Pow-r dither you need to make sure that it is set to 16 bit. Here you need to do some listening to decide which setting is best for you. The three different settings change which harmonic the dither puts noise. I can’t remember which one is which but I use setting 3 the most. This will sometimes change due to the song but generally for me its setting 3. I also want you to know that Pow-r dither is a mastering dither just like the apogee and the Ultra setting on the L2. It should only be used when your doing your final dither down to 16 bit from 24 bit, no other time.

And that’s it. I hope that some of you find this helpful. I sure put a lot of time into writing it (my typing isn’t so good as I’m sure you’ve noticed ). I would love to hear comments and criticisms if you have them. This is the best that I’ve come up with so far but I would love to learn other ways of mastering. Thanks

[fredsparky]

Calvin, Awesome post! I'm going to spend a great time with this!!

[ptrbas]

Great post Calvin. Thanks!

Pieter

[MaKoRancid]

wow. This technique is very good. My mixes are much louder and nicer already!!! Thanks calvin!

edit : "then go threw and set the threshold of each band to be just under the maximum level for that band"

Are those thresholds in the picture the maximum for each band, or just for that song? And how do i determine the maximum level for each band?

[FYN]

Thanks, good information.

[Calvin]

Just like you see in the picture of Multiband band compressor, I tend to set the threshold just under the average level of the audio for that band. Just so that you can see that it is working and cutting off the peaks.

[gYs]

Nice info!
We have used that processed/unprocessed mixing on vocals a few times (compressed/not compressed) but I never tried it on the whole mix until now.

Just curious, do you know of a good multiband compressor other than the Waves C4?

[Calvin]

I know that there are several different pluggins out there that have multiband compression, however I think most of them are VST. I haven't had a chance to try much out other than the C4 and Linear Phase Multiband Compressor in the waves mastering bundle.

[gYs]

right, thanks!
I am asking this because the C4 has this smooth sound to it, at least in my opinion. This can be great, but for some songs it is just a little too 'nice' I think. But I must say I haven't played around with the C4 enough to know every single detail of it's sound with different settings.