Master Plug-in Chain

[reycantong]

Can any of you guys care to share your plug-in chain in your Master Out in PT8? I'm a fan of that loud (but not distorted ) and brickwalled final master.

[albee1952]

I'll start you off with a curve ball here as I usually do not use a master track. I prefer to do my mastering as a separate step. Now if the project is going to be mastered by a pro(which I recommend highly), I would do nothing since the pro maybe could not undo poor choices on my part. But for those quick and dirty (low budget) projects, I will bring the final stereo file into a mastering session(I usually do this in WaveLab but PT works too). I might add a tiny amount of EQ(usually just a little bump at 12KHz), then Waves L2(set for a ceiling of -.02 and no more than 3db of reduction) and add dither as the last step(all my sessions start out 24 bit). If the L2 doesn't get the mix loud enough, instead of hitting it harder, it sounds better to add another L2 in the chain(most of my mixes are plenty loud already but YMMV). While you CAN do all this in your session, I like it separate for a few reasons. First, I can concentrate on it more since the mix is done. Next, I don't worry about the limiter getting too hot for the internal mix buss(maybe I worry too much) and the master fader is BEFORE the inserts so a fadeout can make the mastering limiter act weird as the track fades. I'm sure there will be other points of view just as valid as mine so steal from all of us and find what works best for you.

[Sven62]

Originally Posted by albee1952

I'll start you off with a curve ball here as I usually do not use a master track.

Then how do you know how hot the mix is?

[hnbike]

albee, do you bounce to disk as hot as you can without clipping, or do you leave a lot of headroom for mastering? Thanks

[DrFord]

Sven-
To Albee's credit, he doesn't need to know the exact number of hotness because as long as it doesn't clip, you're fine.

On my rig, if the client has budget for me to master (which I confess I am amatuer at, at best - nor do I have the right gear or room setup for that matter) I will do as Albee said and master in a seperate session.

I will however prep my first session with a master fader, as I also like to see whats going on with the meter, as well as I usually have a frequency analyzer running so that I can check visually what I am hearing. (I find it easier on the master fader so I don't have to throw it up on each chanel.)

For mastering, I use PT mostly, but if T-Racks Deluxe 3 is giving me a hard time I use it in standalone mode. They still haven't really figured out how to make it not hog the processor and stall out pro tools.

As far as signal chain, It usually looks a bit like this

1. Frequency Analyzer
2. McDSP Analog Chanel (Unless I used this plug in a lot in mixing [I do for rock tracks alot but usually not on Synth-ish techno or club style Hip Hop]
3. Compressorbank set to emulate a fairchild (Which I will soon replace with the T-Racks Fairchild as they are going to give the modules individually by the end of the month - very stoked about that)
4. Depending on the track (and this will come before or after the fairchild) I put up a compressor with a color I desire set to a 1.5:1 ratio and a slow attack / auto release to give a bit of energy to the mix if it needs. Pick the compressor I want to impart the color for the song.
5. Linear Phase EQ (I have been importing into T-Racks for this because it's the only one I have)

(2nd Insert bank - Thank you PT8)
6. Brickwall Limiter set to a very soft knee and late attack. Used to just tip the track and not to smash. I use McDSP ML4000, but sometimes I use either the L2007, WaveArts Finalplug, or the Sonalksiss version (name escapes.)
7. Brickwall Limiter set to a hard knee (usually 100% on the knobs) to do the final squash.

This is if I am not doing parallel tracks with different settings.
Sometimes I do a parallel track with Sonalksiss' Stereo enhancer plug, and then I cut the middle channel and just enhance the panned part of the mix. It also allows me to set a "mono under set fequency" so I set that and usually roll it off so it doesn't mess with my mix. I use this plug in sometimes in the mix on stuff so best not to over do it.

There's also parallel compression and just having a track that is not compressed... really there are so many techniques to use... and I really am not skilled at mastering nor do I have the ears yet.

HTH

[albee1952]

DrFord has some great ideas that I will be playing with in future projects. As for knowing how hot my mixes are, its really just a habitual thing where all my mixes will usually just hit -0 when imported into WaveLab. Sometimes they hit too much and I rebounce after backing down the volume of all my audio tracks(including Instrument tracks) by 2-3db(make a group of all these tracks and pull them all down with a single edit-leave the aux returns alone as they don't need adjusting in this case). I'll have to give some thought to adding a master as a check of the main level so I can save the extra step of rebouncing. Its amazing how habit directs our aproach. In any case, I try to always work at the same monitor volume and the same record levels which is why I am usually in the same ball park already. I also always listen to the mix at least once thru from the next room or at a really low level to listen for anything that either si too hot, or gets lost at the lower volume.

[DrFord]

Something else I have heard mentioned in this forum is bit bottlenecking, which if I understand correctly is when too many of your tracks are too hot and you are summing too much at the top most bits, and well it gets bad. So for what that's worth you may find it better to mix not quite hitting the unity gain line. Also, headroom for a mastering guy isn't a bad thing.

Also, if I may touch for a moment, I am not necessarily fond of crazy loud mixes... I like hearing transients from good players... especially good drummers, and the louder your mix is, the less transients you get. There can only be so much water in the glass, so to speak.

So just remember that depending on your product, peak loudness is not as important as perceived loudness also known as RMS. And that can be obtained by gentle compression like the example of the 1.5:1 slow attack compressor or the fairchild.

And, reference mixes are very important. When you are mastering and (like me) you aren't the best... (I always do this) I pull up a mix from someone like Dr. Dre (if hop hop) or Maroon 5 (if pop rock) or of course Christina Aguilera (cause she is not only sexy and immensly talented but her music has enormous budgets and the best people) and I check my levels and sound against the tracks that make sense.

Another good reason to have a mastering session. You can apply your mastering on a track with the stereo file instead of having to put it on the master fader. So that way you can check your loudness (gain plug in can check for RMS I believe) on the reference track, as well as an analyzer to see how they are shaping the EQ curve, and apply similar enhancements to your own track. You also then have the ability to match the volume of the reference mix with the fader, because remember that unless you are listening to Metalica's latest album... louder by 1db will sound "better." So make sure you match the volume.


Albee-
What do you do differently from what I do? I could always use new methods pointers too!

Thanks

[reycantong]

These are great pointers for me. Thanks guys. More...

By the way, what can you say about the Izotope Ozone Presets?

[albee1952]

Hey Dr, I spelled out my routine pretty much and it looks to me like you have a lot more options with your methodology. I realized a long time ago that mastering is a skill that I am just not really interested in digging into that much. I purposely keep it simple to avoid mistakes(learned the hard way). I really like the concept of parallel limiting and will have to give that a try. I also hate the "slammed and squashed" sound that we must suffer with these days and only use the gentle L2 to get the apparent volume up to a point where it stands well next to some "average" commercial releases. WaveLab has a nice spectrum and frequency analyzer that can average an entire song file and I will run that to check for any peaks that I may want to massage. I have also taken to checking/removing DC offset as the very last thing before I save/close the file. Sometimes there is none but others, it does find offset and as I understand it, the offset can actually cause clipping when all levels are okay. Also, the trick(?) of setting the ceiling at -.02 is because I have read that some converters are a touch more or less sensitive to clipping. And slamming with a lot of mastering limiter can basically eliminate any and all headroom. Lots to think about here and lots of good advice from DrFord so instead of diving in the deep end, just wade in a little at a time and don't forget to check your masters on several speaker systems. A good mastering house has spent more $ on speakers and acoustics than many of us spent on our cars(or maybe our houses).

[DrFord]

True that on the mastering houses.

Good mastering guys have special coveted gear that is hand modified anyway... but yea their rooms are usually small, and have perfectly flat frequency responses using ultra high end monitors. And beyond that they end up changing the cones on their monitors like twice a month...