Mastering Chain - what say you?

Pinky

and The Brain...
Today I spent some youtube time checking out mastering chains that the professionals use. In the past I tinkered with some filters, limiters, compressors, and whatever else I had laying around that I could chain in Soundforge to learn how they all played (or didn't) together. The results were underwhelming so I went back to a basic EQ --> Waves L3 chain [now with a RMS meter] and focused my effort on getting the mix sounding right and doing minimal adjustments at the mastering stage. This is on my own tracks, for the bit of mastering work I've done for others I have to adapt to the song or artists needs. Usually for extra loudness I use one of the compressor plugins I have in front of the limiter, with varying degrees of success, which is one of the reasons I'm starting this thread.

I'm curious what others do in a purely virtual mastering setup (all plugins/daw). Mastering is definitely my weakest production area, the only technique is what I've gleaned off the internet or discovered on my own as described above. It's time to lean on the better mastering people here to see what tools/techniques I need to add and learn. Links are welcome if you know of a good tutorial to share. :)

I have a decent set of 8" studio monitors (Samson Resolv A8, I'd say they're comparable to the Yamaha HS8) on iso acoustic stands, and a small 8" energy sub wired in with the crossover set to prioritize only lowest frequencies the Samsons don't handle quite well enough. The room lacks formal treatment, but speaker positioning and other articles in the room (it doubles as a outdoor gear storage room, so lots of backpacks, climbing rope, and sleeping bags, etc) aid greatly in helping to dampen enough I haven't felt the need to add anything. It's also only a medium sized room so for near field my setup is decent enough. Playback wise this setup has served me well for the past couple years, as I've done several albums now with it and the results have been good on all playback systems.

My primary DAW is Sonar Producer X3, as I mentioned I do mastering in Soundforge though I'm considering perhaps trying it in Sonar. I have quite a few Waves plugins, a bundle from IK Multimedia, and numerous other VSTi and mix/match plugins I've been accumulating when they go on sale. I managed to snag a few useful Soundtoys when they went on sale recently. I mention this to say that I may not have exactly what you use, but probably something comparable somewhere in the library.

Hope this was enough info without being TMI.
 
Last edited:
I went back to a basic EQ --> Waves L3 chain [now with a RMS meter] and focused my effort on getting the mix sounding right and doing minimal adjustments at the mastering stage.

I'm with you on that...minimal audio mangling during the mastering stage.
I've tried going with more shit piled on...with a variety of combinations...and now I pretty much do a touch of EQ and some level adjustment. I've got some new EQ/comp/limiter plugs aimed at the mastering stage that I'm eager to try out on my next crop of mixes...lots of stuff from the Plugin Alliance roster of plugs.

It think it's more about enhancing what's already there...or at least that's what I'm going for. My thought is...if I'm trying to really fix something during mastering...it's probably something I should have done during mixing.
 
Im with you on that mastering is my weakest point, I recently got the Izotope ozone 7 mastering plugin, I am still learning to use it but so far I like it. my room is treated to some extent but not perfect. I have another set of speakers in the main room I use to bounce things off of. I am still lost to the whole mastering process.:D
 
Im with you on that mastering is my weakest point, I recently got the Izotope ozone 7 mastering plugin, I am still learning to use it but so far I like it. my room is treated to some extent but not perfect. I have another set of speakers in the main room I use to bounce things off of. I am still lost to the whole mastering process.:D
(1) Try not working on your own material -- It's much like chasing your own tail.

(2) I'd - I'd suggest maybe not using Ozone... But that's just me. It's all maul-the-band (multi-band) this and maul-the-band that... Rarely ever used in the mastering process (no matter what the marketing people say). And I'll say it -- I've never used anything that can cause so much damage to an otherwise perfectly decent mix with such ease.

Want some really great plugs to mess with...? Tokyo Dawn. The Nova, Kotelnkov, the EQ (I can't remember the name although I have every version -- "Slick?" perhaps?). Freaky cool - and free for the regular versions. They do some odd stuff - But in a pretty cool fashion.
 
(1) Try not working on your own material -- It's much like chasing your own tail.

(2) I'd - I'd suggest maybe not using Ozone... But that's just me. It's all maul-the-band (multi-band) this and maul-the-band that... Rarely ever used in the mastering process (no matter what the marketing people say). And I'll say it -- I've never used anything that can cause so much damage to an otherwise perfectly decent mix with such ease.

Want some really great plugs to mess with...? Tokyo Dawn. The Nova, Kotelnkov, the EQ (I can't remember the name although I have every version -- "Slick?" perhaps?). Freaky cool - and free for the regular versions. They do some odd stuff - But in a pretty cool fashion.

Thanks John, I actually sent you an email awhile back, I may need your services on some projects I have coming up. :D
 
(1) Try not working on your own material -- It's much like chasing your own tail.

(2) I'd - I'd suggest maybe not using Ozone... But that's just me. It's all maul-the-band (multi-band) this and maul-the-band that... Rarely ever used in the mastering process (no matter what the marketing people say). And I'll say it -- I've never used anything that can cause so much damage to an otherwise perfectly decent mix with such ease.

Want some really great plugs to mess with...? Tokyo Dawn. The Nova, Kotelnkov, the EQ (I can't remember the name although I have every version -- "Slick?" perhaps?). Freaky cool - and free for the regular versions. They do some odd stuff - But in a pretty cool fashion.

I have a good EQ plugin (Waves GEQ) but think my compressor selection is poor. There's also plugins that add analog "warmth" to the mix, which I've never felt the need to have, but perhaps I should try a few and A/B to see for myself. Any trial/free suggestions along those lines? :)
 
I note MM's advice not to master your own, which is sound advice . . . fresh and uncontaminated ears can hear things that you can't.

Having said that, I nevertheless do my own mastering. I don't touch EQ at all . . . there's no point . . . if I mixed it in such a way that it seems to need EQ, I'll go back and remix until I think it doesn't. That's not to say it might not benefit from further EQ . . . it's just that I can't hear anyfurther requirement.

I just top and tail the tracks (i.e. making sure the entries and exits are as I want them), and work on levels. There are are two parts to this: the first is getting a suite of songs up to a reasonable level generally; the second is to get the songs to sound consistent with each other.

Thus my chain is very simple . . . I dump all the tracks into Soundforge, do an RMS normalisation on them (using the Soundforge tool), then use a WAVES L1 maximizer to limit the ceiling to -0.3 and do the finer adjusting between tracks by varying its threshold.

Note that Soundforge and Waves L1 are ancient: I got both on a CD when I first got Logic and the ISIS sound card way back around the turn of the century. My version of SF is 4.5, but it still works a treat.
 
I note MM's advice not to master your own, which is sound advice . . . fresh and uncontaminated ears can hear things that you can't.

Having said that, I nevertheless do my own mastering. I don't touch EQ at all . . . there's no point . . . if I mixed it in such a way that it seems to need EQ, I'll go back and remix until I think it doesn't. That's not to say it might not benefit from further EQ . . . it's just that I can't hear anyfurther requirement.

I just top and tail the tracks (i.e. making sure the entries and exits are as I want them), and work on levels. There are are two parts to this: the first is getting a suite of songs up to a reasonable level generally; the second is to get the songs to sound consistent with each other.

Thus my chain is very simple . . . I dump all the tracks into Soundforge, do an RMS normalisation on them (using the Soundforge tool), then use a WAVES L1 maximizer to limit the ceiling to -0.3 and do the finer adjusting between tracks by varying its threshold.

Note that Soundforge and Waves L1 are ancient: I got both on a CD when I first got Logic and the ISIS sound card way back around the turn of the century. My version of SF is 4.5, but it still works a treat.

Similar to my strategy and mindset. I've learned over the years how to give myself proper distance from my own mixes so I can hear them fresh. Additionally, I tend to mix much of the time on headphones then master with the studio monitors, auditioning that master on my HD600 and various playback devices. Mastering on the monitors helps a lot with hearing the mix with fresh ears.

My underlying concern is learning the tools for 'heavier' mastering, dance and metal especially. These genres almost require a heavy handed mastering technique, and since I typically live on the opposite end of that spectrum I haven't learned to harness compression at the mastering stage as I should.

As a side note - I find that even in the case of industrial, dance, and metal that using some side chaining really helps reduce the need for mastering gymnastics to get a good volume while retaining some punch. This is assuming I have access to and complete control over the entire mix.
 
My typical mastering chain is really simple and not at all fancy, but it works for me. If I'm doing a full album I'll usually have each song on its own track with an instance of ReaEQ on each so that I can make minute adjustments to bring them all just a little closer to sounding the same - like they belong together.

Then they go to the Master where the chain usually goes ReaEQ>ReaComp>ReaEQ>lt_diode. Sometimes I'll try a second ReaComp, but I use that less and less lately.

The two instances of ReaEQ are partly there to give an overall frequency curve to the album as a whole, like recording to a specific tape formula and speed might. Not that I'm trying to emulate tape really, just that I want all the different pieces to sound like they all come from the same place, to have a sort of sonic signature to the album. Aside from the "bookend" high and low pass filters, there might be a band or two of very small, wide dips or boosts to maybe add a touch of bite or air or girth or whatever I think it needs.

Almost more importantly, though, these two EQs allow me to change the behavior of the compressor. It's what I call "pre-emphasis/de-emphasis". Mostly it means that I shelf down the extreme low end before the compressor and then shelf it back up on the other side, though I sometimes will also add a bit of boost to the top end before, and cut it back down after. A little but goes a long way here, and you can't always "undo" a big difference because of how a compressor does what it does, but it can make a real difference if you have a little patience dialing it in.

The compressor itself is there to just smooth things in a little tighter, to make the instruments jostle each other just a little bit so they can all fit through the little speakers and sound like a happy family. I use a long RMS time, set the pre-comp parameter half as long as the RMS, turn attack and release down to 0, and set the ratio to 1.1, which is as low as it gets. Then I set the threshold to where it just barely does anything at the louder parts of the song, and then crank up the knee parameter so that it's almost always doing a little bit, but never actually reaches that full ratio. The effect is subtle but addictive. It's almost too easy to end up crushing the dynamic range because its so transparent you kind of dont hear it fall apart the way smashing it into a brickwall would do.

But it doesn't really catch all of the transients. By this point I should have handled most of that already, but you still sometimes get that one or two brief spikes where everything accidentally hits together and creates a peak that is just way bigger than all the others, and those just waste headroom for no good reason. That is part of what the lt_diode plug is for. It's really nothing more than a saturator plugin that sets a hard limit and has a bit of curve going into it - "soft clipping", but I wrote it myself so I know what it's doing, and it allows for asymmetrical clipping, which makes me happy. I tend to look at it like the "rails" on the mix bus.

The real trick is in setting the thresholds and then controlling where each song sits with respect to those thresholds. Once I get the plugs on the bus set to where their working together to get what I need, I use the faders on the individual song tracks to adjust their relative levels and how hard they're hitting the comp. For me, that usually means first roughing it in almost by eye - looking at the waveforms and adjusting the take volume so they all peak around the same, or seem to average about the same, or otherwise kind of look close, and then skipping around almost randomly while playing and listening for things that are way off between the different pieces. The goes and whys of this part are so program specific that it's hard to even generalize. I guess most of the time I look to get the drums and/or vocals to hit around the same level from song to song, and let the rest of the instruments fall where they may, but it really is about listening and thinking about how like if you just "drop the needle" some random place on the album it should sound like it is actually the same record as if you dropped somewhere else.

That's a lot of words for such a simple setup. If I use a second ReaComp, it's set faster, with higher ratio and threshold, but usually a smaller knee. I still usually still will leave A and R at 0, and use the pre-comp halfway through the much shorter RMS. This would basically be to kind of help the clipper not have to work quite so hard, but I've found that any time I think I want this, I can usually do better by going back to the mix.

Every once in a while, especially on some of the really crazy experimental noise stuff, I'll drop Otium BassLane on at the beginning of the chain in order to center the sub frequencies partly so smaller speakers can share the load, and partly for headphone compatibility.
 
My typical mastering chain is really simple and not at all fancy, but it works for me...

Almost more importantly, though, these two EQs allow me to change the behavior of the compressor. It's what I call "pre-emphasis/de-emphasis". Mostly it means that I shelf down the extreme low end before the compressor and then shelf it back up on the other side, though I sometimes will also add a bit of boost to the top end before, and cut it back down after. A little but goes a long way here, and you can't always "undo" a big difference because of how a compressor does what it does, but it can make a real difference if you have a little patience dialing it in.

The compressor itself is there to just smooth things in a little tighter, to make the instruments jostle each other just a little bit so they can all fit through the little speakers and sound like a happy family. I use a long RMS time, set the pre-comp parameter half as long as the RMS, turn attack and release down to 0, and set the ratio to 1.1, which is as low as it gets. Then I set the threshold to where it just barely does anything at the louder parts of the song, and then crank up the knee parameter so that it's almost always doing a little bit, but never actually reaches that full ratio. The effect is subtle but addictive. It's almost too easy to end up crushing the dynamic range because its so transparent you kind of dont hear it fall apart the way smashing it into a brickwall would do.

But it doesn't really catch all of the transients. By this point I should have handled most of that already, but you still sometimes get that one or two brief spikes where everything accidentally hits together and creates a peak that is just way bigger than all the others, and those just waste headroom for no good reason. That is part of what the lt_diode plug is for. It's really nothing more than a saturator plugin that sets a hard limit and has a bit of curve going into it - "soft clipping", but I wrote it myself so I know what it's doing, and it allows for asymmetrical clipping, which makes me happy. I tend to look at it like the "rails" on the mix bus.

I think I've seen you describe this technique before in less detail. I'm going to see if I can replicate it with what I have.

Yesterday I did spend about 30 minutes setting up a new mastering workflow (template) in Sonar. I added the C6 compressor to the mastering chain, in front of the limiter. GEQ-->C6-->L3 (limiter)-->TT Level (for monitoring dynamics). The C6 has 4 mastering presets, I like how the "B" preset worked with the test track (something I had mixed for a collaborator). My goal was to liven it up, and see how loud I could acceptably get it without clipping. [fwiw - this artist likes their mixes bass heavy]

Here's an A/B of yesterday's results with the prior master:

Prior

Latest

The artist liked the results.
 
Last edited:
Yesterday I did spend about 30 minutes setting up a new mastering workflow (template) in Sonar. I added the C6 compressor to the mastering chain, in front of the limiter. GEQ-->C6-->L3 (limiter)-->TT Level (for monitoring dynamics). The C6 has 4 mastering presets, I like how the "B" preset worked with the test track (something I had mixed for a collaborator).

I had forgotten about T-RackS, which I got with my Amplitube 3 bundle. I didn't end up using any of the compressors, but I could see a purpose for some of them depending on need/mix. I did find that I like their Metering plugin (and have subsequently dropped the laggy TT level). After tinkering more with C6, it seems that this chain is precisely what I was looking for. Now I feel like I have better control of the dynamics. As a result, I can also push the mixes more without hitting their limit nearly as early. Don't worry, no interest in throwing my hat into the loudness wars. But it's good to know how to do it if I'm ever asked.

C6 is effectively providing me that first layer of compression askcat described.

I've subsequently gone back through and remastered all my latest (unreleased) material too. The songs truly shine now!
 
Was watching a youbube vid not long ago and the guy put a glue compressor in first: 1.1:1 and just a hint of reaction. Followed that with a tape simulator. Followed that with a bit of 12k shelf boost (to make up for the knock out of the tape sim). Followed that with a stereo widening effect. Followed that with a touch of demolition (distortion, color, whatever you call it). Followed that with a limiter. Listening through the process, you could barely (if at all) hear a difference in what he was doing. When he was done and A/B'd the tracks, it was night and day. I'm not remembering the entire chain because (if I remember right) he had seven different things going on in the chain...seemed like a lot, but it came out magnificent.
I just don't have the chops to try to put 7 pieces of kit on my mix bus and have it not sound like 7 pieces of shit...something always gets overdone. So I just stick with a fairly clean saturation setting and pump a fairly transparent limiter (Onyx Tube Limiter RE) until I can hear the mix start to change, then back it off. I've also played with the C1-L1 limiter section, and Ozone Maximizer RE (which is not multiband), but doesn't do much for my tracks (with my experience).
 
Ive never "mastered" before, this is new for sure. Im from an analog background but find Im doing more and more with the digital set up.

I upgraded my DAW (studio one 3) to the Professional level and it has a mastering suite. Reading responses here I think my understanding is in the right direction, not much is going to be getting changed during this process, more like a good wash before you sell a car. Make the tracks have a similar sonic character and mind levels. The mix should have done most of the work.
 
A couple of plugins I like are the filtrate Eq, voxengo marvel EQ, and limiter No6 (which even has intersample peak detection). I typically only use the final module on limiter No6.

I like reaxcomp for multi band compression.

I am finding that mastering is as much about feeding into fx to change the Eq curve as it is about using Eq to change the Eq curve.

Waves L1 is not a very nice limiter IMHO. It's really crunchy sounding/not at all transparent.

A typical chain for me might be
gain
Eq
(Mid side Eq if needed)
Multiband compression (does nothing at all if the mix is balanced well)
Limiter
Dither
 
Very very very very, unforgivingly linear monitoring. If you don't have this, you won't know what you are hearing.

I wish I had dealt with my monitoring a lot sooner.
 
Very very very very, unforgivingly linear monitoring. If you don't have this, you won't know what you are hearing.

I wish I had dealt with my monitoring a lot sooner.

Still don't know what you're talking about. L1 is one of the cheaper/lower end Waves limiters, L3 the more robust limiter and considered to a certain degree a gold standard among software limiters. What does this have to do with your monitoring?
 
I've heard that compression in small steps is better than large leaps. Any thoughts? Is it better to slam a limiter at the end, or stage it up with track comp, bus comp, glue comp and then into the limiter. In my (limited) experience, this just seems to make everything flat and lifeless.
Is it just me? I see people talking about not EQing before limiting. My limiter adds a flavor that I have to pre-EQ or it really shows through...
 
Back
Top