Hi,

Happy weekend and hope everyone is having a good time. I'm feeling generous today, so I'd like to share 3 tips for ITB Mastering, important ones that I think some, especially the new people could benefit from it or maybe could consider and implement in their workflow. I've been using them for 2 years now and while they could be subtle and figured there's a huge difference (at least to me) with or without them. Before anything :

1) I'm not a certified mastering engineer, I have around 40+ tracks I've mastered for clients that mostly independent artists and local bands so far but I wouldn't call myself a mastering engineer for few reasons. First, those tracks most of them I "have to" master because I'm just doing the favour after telling them repeatedly that I don't do mastering but these people are not really familiar with music production stages and roles for them it's all the same, so I did it for them anyway. Secondly, me personally I believe the mastering stage is more about another person that doesnt know anything about the track perspective, proper gears combined with world-class monitoring which I don't have. Others are allowed to think otherwise, it's just me and where I'm currently at my music career.

2) Is mastering your own song/track wrong ? NO! should it be a separate stage ? NO! The only right thing is the one that works for you, everything comes with PROS & CONS of course.

3) These tips are just stuff I've picked up randomly from famous superstar mastering engineers from their videos and interviews, I've tried and it works for me and I hope it could work for others. Although I don't do full final mastering, I still do fake mastering everyday for the mixes to get them up to proper listening levels for preview. Those are sent to clients and the mastering guy for reference.

4) I'm just sharing this in the spirit of going against the bullshit gatekeeping attitude in our community. I still believe there's still some black magic and voodoo that I've yet to discover for mastering process. It's true, no kidding.

------------------------------------------------------------------------------------------

1. THE BASS SHIFT

Spoiler: CLICK TO EXPAND

This is something I've ripped off from the Lurssen Mastering duo, some of my personal favourite MEs. If anyone familiar with their mastering approach and chain, the people at Lurssen are using their custom build console that have few gears in it that work together as one big processing unit. One of the gear is a GML 8200 parametric EQ that 95% of the time (according to them in few videos I've watched) is set on a setting that they rarely change and it's always ON. The settings also implemented directly into their IK MULTIMEDIA LURSSEN MASTERING CONSOLE plugin (on the Solid-State EQ) and the user cant change it. The settings is simple :

a) -0.5db, Q=2 @ 30 Hertz
b) +0.5db, Q=2 @ 40 Hertz

This is the usual "Bass Shift" move in mastering that I've seen few MEs doing including Jonathan Wyner (Izotope's Are You Listening) in one of his videos (his EQ of choice is SPL PQ). What it does is it simply make the bass shifted inside to the midrange a bit. It's subtle and hard to hear with the EQ turned ON and OFF but in a full mastering chain with compression, limiting and whatnot, turning it ON will make the bass solid and a little perceivable. It will make the track sounds tighter especially when combined with the compression after. It is placed before the leveling compressor (Vari-Mu, Opto) in the chain to excite the low-mid.

To replicate it I suggest the mighty Pulsar Audio Model 8200 EQ plugin :



Notes :
a) there's something about using GML style EQ for this move, I think it has to do with its parallel topology, before this plugin I've been using the DMG EQUILIBRIUM for it. I have tried the Pro-Q3 but the bass shift is not as satisfying sonically to me compared to using the EQ with Parallel Topology plus Analogue phase. I could be wrong, idk.

b) I suspect there are some phase reactions happening that I've yet to understand for now that made it sounds like that. As for now, I don't really know for sure, but it 100% works for me, all the time. I believe the move has to do with making the bass more pronounced for smaller speaker caused by something with the phase twist and some added harmonic distortion from the plugin. It's subtle, but most stuff in mastering is.

c) the effect isn't really perceivable with the EQ turned ON/OFF but with compression after, the effect is more pronounce and the low end becomes more solid, tighter than before, assuming the compression setting is correct.

d) other EQ plugin with Parallel Filters : DMG AUDIO EQUILIBRIUM (8k2 or 250 curves + Parallel Peaks on DSP settings), MAAT thEQBlue (Parallel FF/FB Mode), TDR NOVA, SONORIS PARALLEL EQUALIZER.

2. THE HF & LF PEAK LIMITING

Spoiler: CLICK TO EXPAND

Countless of time I've heard MEs saying "this is my secret weapon" when referring to their MASELEC MPL-2 or ETEC DBTL20 unit suggesting how important it is for their workflow. It's technique started back in the days for vinyl mastering to limit the High and Low frequency content of a track so it will translate nicely when transferred/cut on the lathe. I've heard them saying all those good music we've listened growing up, all the smooth high end that made people think "vinyl is better than cd" is basically thanks to the High Frequency Limiter doing its job perfectly. Just to name MEs that been using this processing in their chains, Pete Lyman (Infrasonic Mastering) and again the Lurssen's duo.

It does what it says, limiting the High & Low frequency. I've seen people placing this at close to the end of their chain, some placed it before their final analog limiter and most of them place it before the second compressor (usually the VCA style) for 2 compressors setup chain. The HF Limiting is a lot like the De-Essing process but typical De-Esser works at the targeted frequency range of the harshness but HF Limiter affected range is the whole High Frequency area starting from the cut-off frequency selection to the Nyquist.

a) It helps pushing the track to the limits of the high end content with an EQ placed before it doing the usual High or Air Shelf boosting. Pushing with the EQ to bring up the high end content, while the HF limiter is pulling down the excessive peaks, and you find the right balance in between. Me, I will spend some time for this, smooth high-end content is a make or break for the mix/master. It's like playing with fire sometimes, you can get burned too. Moderation is key.

b) 100% of the time, the pros will have HF Limiter is at a default set and forget settings in the room, one that they know if the signals crossing the threshold for a certain amount, it's signalling too much high end content for the overall balance. So the HF Limiter must be set up with the user understanding their monitoring environment, the room, the speaker and at the right working SPL (Sound Pressure Level). If the high end frequency content is too much, it will trigger the HF Limiter aggressively, if it's not triggering at all, it's not enough. It takes time until one will get a good working setting that can be trusted.

c) using a reference track as a benchmark can also work if the above option is not possible or for those that using headphones.

d) the LF LIMITING works differently, as it's more or less a soft-knee clipper that work as peak leveler just to cut off some of the Kick/Bass Drum peaks before the signal reach compressor. It's useful for low-end management, if one likes their compressor on slow settings that lets through the low end peak also to manage pumping for fast attack compressor settings. Simply put, it's to "relax the dynamic processing later in the chain" according to TDR in the manual. I don't recommend pushing signal into it like you'd for HF limiter. Another important thing is, LF Limiting is optional. Some music just don't need Low Frequency Limiting.

Plugin recommendation is one of the best limiter plugin ever, the Tokyo Dawn Lab Limiter No.6 GE but stripped down to just HF LIM and CLIPPER modules :



Settings for HF LIMITER :

a) Frequency is set to 2500Hz, replicating the frequency of the MASELEC MPL-2 unit. It works great.
b) most of the time the full stereo mode works fine, but the plugin is capable of doing M/S mode as well.
c) 3db of Gain Reduction is MAX dont go more, 1-2db GR is awesome. You are not suppose to hear what it is doing, use the DELTA function to listen to the artifact and keep it minimal.

Setting for Low Frequency (CLIPPER module) :

a) Mode is set to LF CLIP mode.
b) I like the SEPARATION at 50%, it's the middle ground between a Wide-band (0%) and full multi-band (100%). For bass heavy modern music, it is better to use 100% settings. It depends. 50% is a good starting point IMO.
c) 6db for KNEE and 0.5-1db Gain Reduction is more than enough. Again, you are not suppose to hear what it is doing. I personally use it with the compressor after, closely listening to the compressor behavior on the low end, setting up side-chain filter and set them all up together. Listen in DELTA mode and make sure the artifact is tolerably low.

Notes :
1. For both HF & LF Limiting, good rule of thumb is if you can hear what they are doing, it's already too much.
2. Normal human being isn't capable of listening to the peaks (fast transients), we can only hear average of them collectively in certain amount of time and perceived them all in once. Even though we can't hear it, it's there, really important for the sound and needs to be managed. Tools like compressors and limiters are invented specifically for this job.
3. What's the difference between HF Limiter and using ProQ3 on Dynamic mode for the high end ?
- Dynamic EQ on ProQ3 under the hood is a soft-knee RMS detection compressor with program dependent att/rel & ratio function working in parallel and mixed back 50/50 with the original signal. It affects the dynamic balance of the whole track.
- HF Limiter is a high-ratio peak detection compressor with fast attack that supposedly catches every transient that get past the threshold. The process is applied directly to the signal. With the right usage, it shouldn't affect the dynamic of the track at all.

3. SMOOTHING THE SQUARE WAVE

Spoiler: CLICK TO EXPAND

This is the most exciting one (at least for me, lol) I picked up from Brian Lucey (Magic Garden Mastering) from one of his videos. One of my favourite ME and the man is known for being extremely particular about every single component of the gear in his chain and handpicked every single NOS tubes used in them. The idea is about balancing the amount of 2nd order and 3rd order harmonics generated during the process. Apart from him balancing the amount of solid-state vs tube based gears he used in the chain, the one I want to talk about is how he used the saturation section of CRANE SONG HEDD QUANTUM ADDA converter for the final limiting process.

The idea is when doing clipping and limiting, the harmonic generated from the process are mostly 3rd order heavy which is known to be brittle and edgy in character, so in order to balance it out, he used the Pentode (Tube) style saturation on the unit to add 2nd order harmonic to the signal to theoretically smoothen the flatten/squared wave-form. Here's the link to video where he talks about it -> LINK. Fast forward to minute 20:00 or just watch the entire video.

To replicate the process, here's some steps I've used as suggestion :

1) this process requires using 2 Limiter plugins.
- the first limiter, is your usual favourite Limiter plugin
- the second limiter, is True-Peak brickwall limiter. I highly recommend A.O.M Invisible Limiter 3.

2) do the limiting as you normally would with the first limiter, except set the Ceiling to -3db to give you some headroom. Make sure at this point you are already happy with the sound of everything and it's 98% close to the final product you have in mind. Limit the track as you'd normally would for brickwall limiting but if possible TURN OFF the True Peak mode.

3) add a tube-based saturation plugin after the limiter to add the tube saturation to generate 2nd (even) order harmonic saturation. More about this below.

4) Use the second limiter as your final True-Peak Brickwall Limiter. You need to push around 2db (the headroom you left earlier on the first limiter) to get to -1db ceiling (or theoretically 0db). I highly recommend using the A.O.M INVISIBLE LIMITER G3 for it is known for being the most transparent, produced close to no harmonic at all with lowest artifact ever limiter because this step suppose to be as transparent as possible. 0.5-1db Gain Reduction MAX. It's just to keep you safe from getting over the ceiling to the digital clip territory. Expect no change in sound whatsoever, just a brickwall limiter doing barely nothing. Here's a setting I recommend :



a) SOFT LIMITER = OFF
b) BRICKWALL LIMITER = ON
- set ceiling (THRESHOLD) to -1db or whatever works for you
- TRUE PEAK AWARE - ON
- CH LINK = 0%
c) INPUT GAIN accordingly, 0.5db Gain Reduction if possible or just keep it below 1db. Make it as transparent as possible.

5) Other limiter plugin produces harmonic (it's 3rd order) during limiting, I have tested a lot of limiter plugin and so far only the A.O.M's is clean enough. Limiter that produces 3rd order harmonic is kinda making the whole steps/trick useless if you think about it. Although the amount of harmonic generated by normal limiter plugin can still be considered super low at 0.5db Gain Reduction limiting, it's always a good idea to keep things as clean as possible, especially when you can.

THE TUBE SATURATION

Any tube saturation plugin is good as long as they have flat frequency response but since we at the Mastering stage, picking the right plugin is always half of the job. The BLACK BOX HG2 or OZONE EXCITER module have Triode & Pentode style tube saturation (I like the "Warm" one on Ozone) and I believe they are good enough to be used in mastering.

EDIT : forgot to add TONE PROJECT KELVIN it has pure 2nd order harmonic generator. All the tubes mode are awesome too.

I have tried different plugins before but I always come back to this one plugin because it's just sound good to me. Weirdly enough it's not even a tube saturation nor based on a Tube style hardware unit. It's FET-based, at least half of it. FET is short for Field-Effect Transistor, it's a (solid-state) component made to replace the Vacuum tube that is known for being inconsistent and hard to maintain. Well, you can google it. My point is, when moderately used, FET saturation is closely similar to Tube, except it breaks up quickly and clip the sound aggressively beyond that threshold. My choice of this plugin for this job is the VERTIGO VSM-3 by Plugin Alliance/brainworx because it sounds the best to my ear and easy to work with.



This is the settings that works for me, it produces clean 2nd order heavy tube-like saturation, while having the flattest frequency response possible, when listened on DELTA (Reaper users benefit only, lol) you can hear it's adding nice low-mid to mid noise-like low volume sound meaning it doesn't affect the track much, just rounding off the aftermath of a heavy limiting.

The settings :

a) THD MIXER fully counterclockwise to the 2nd order Harmonic Balance. The Zener Section is OFF and doesnt affect the sound.

b) FET CRUSHER SECTION
- DRIVE = 3%
- MODE = FULL
- STYLE = SOFT
- SHAPE = 100%
- THD MIX = 10%

c) dont push the plugin, don't DRIVE the plugin (read the reminder below). With the settings above, the generated harmonics pattern :


d) it's more like a "feel" thing than can be heard thing. I've been doing it for some time and can hear it. It can still be heard on good headphones tho.

REMINDER/PRE-CAUTION
- don't push the tube! tube-style saturation only produces nice sweet 2nd order harmonics heavy saturation at its low to medium not-pushed settings.
- tube style saturation will crap out quickly if pushed and will start to produce 3rd order heavy pattern which will reverse all the process. Remember the trick is to balance the harmonics, not add more saturation.
- it's a delicate process, so test the plugin with sine wave, plugin doctor to see if it works correctly as intended for the job.

I hope these tips can be useful to y'all and if you have any question or anything, just ask me, I don't know much but if it's within my capability of answering it, I'll try my best.

Happy experimenting! Cheers.

 

Last edited: Jan 19, 2024

thank you so much stevie!
I am also a fan of Lurssen guys, I will surely test the trick #1 very soon because it seems to be the easiest to implement

#3 could be also interesting for me, will try to slowly implement it in my chain. I do already something similar but with the Waves L2 doing 1.5dB limiting after your usual limiter/s final chain, maybe you have heard of this, it helps taming LF and HF and move forward a bit your middle range. It doesnt work always, but when it does it's very good

 

Amazing job: loud and clear and explanation, really cool to share that much wrapped (and correct) information to the public.

Regarding the Bass Shift: in digital you can easily tune the boost/cut frequencies to align to the lowest fundamental (kick or bass) or just moving it a little bit up or down (touching that swe(e)t spot) and so amplifying the result with the same db move.
I don’t think that this technique alone will create lot of issues related to phase, while I think those would become more relevant if you add other moves around that area

Another guy who you might wanna check out, specially for low end techniques, is Bob Powers (but maybe you already stole all the secrets you can from him)

Grazie for your post ciaciao

 

Many thanks for sharing this. Very clearly reasoned and explained.

 

Awesome!

 

Nice, man. Wow!

I'm a mastering engineer and i haven't been doing the bass shift thing. I'm gonna start testing that out now. Thanks

 

For the 3 tip I'd also like to suggest Boost by UrsaDSP, it's by far the cleanest limiter I've ever used, FirComp2 comes next.

 

Correct me if I'm wrong, but isn't 8200 supposed to be an emulation of a completely transparent mastering EQ?

Boost is so incredibly good it feels like cheating lol.

 

stevie dude, you share some great knowledge. greatly appreciate this post, as well as your acustica posts etc. thank you!

 

There are some great tips here

Am I mistaken or the HF-LF limiting may also be a great way for cleaning up some busses/tracks when mixing an all-ITB song? Lately I've been experimenting with some very broad, very subtle settings on the Weiss Deess for preventing the harshness from piling up on DI guitars and on some VSTi, and the LF clipping seems like a nice addition.

 

that's nice

 

How do you guys feel about true-peak limiting? I think most sound bad but I never tried Invisi Lim G3.
I just find it to ruin the sound a little.

Also final export at -1 or -2 dB? I think -1 is not enough. Streaming services suggest -2dBTP.

For some reason newest Boost plugin keeps asking for activation, anyone else have that problem or know a fix? I just use the one before.
Thanks for sharing these tips.
I have been wanting to try the 8200 for a while now.

 

Newest one from R2R is timebombed, use the one before from TCD

 

yes, im aware of that "mid-pop" L2 trick. It's cool and I think I've been doing that a lot few years ago. Forgot why I stopped though.

I have watched Bob Powers old seminars videos or something on YT and forgot about it, I'm curious can you pls give me the link about what you mean by the Low End Technique by him ? could be cool if I don't know about it.

Yeah, Boost is clean and one damn fantastic plugins. I uses them on track these days. Boost uses multiband technique for the process, the low-end is being treated separately which has some sort of clipping stage before the actual limiting that why I think the amount of volume you can push from that plugin is a lot. Unfortunately it produces 3rd harmonics for the low-end only, it's really low but it's there, I have tested it before. I think it's clean enough if used solely for the brickwall limiting without pushing but AOM G3 is cleaner so I picked that one.

Yeah, the plugin is clean, well not 100% the harmonic produced by it with a sine wave test (or the Plugin Doctor) is showing them below -100db which is a little too low and no way it's perceivable. I have compared the Pulsar 8200 with DMG Equilibrium, MAAT thEQBlue and TDR Nova (that capable of doing parallel filter EQ), I like the Pulsar version better. I just did simple EQ comparison with curves and listen and I ended up liking the Pulsar. I used to hate it on its release. lol. I think there is something that I don't know about it just. The Pulsar Massive (other plugin) was known for doing some cool stuff on the low end, Dan Worrall found it. I think 8200 has its own secret sauce.

For the hardware, it's clean enough for a hardware standard I guess since no such thing as 100% clean in hardware land, a lot of people swear by it that it has a box-tone tho.

Just FYI, the Pulsar Audio company was founded by the some of the people that made the Slate's plugins (alongside Fabrice Gabrice of course) which IMO are some of the best plugins out there back in the days. Virtual Bus Comp, FG-X all those good stuff. Still good now. So I assume they are talented people that has some secret sauce they have used that we all have yet to discover.

Dude, you too are sharing great info and insight about stuff which made me learn a lot. Thanks for that too.

Yes, they are useful for that job. Don't forget the Pulsar 8200, it has the built-in De-esser that can be combined with the Air Boost which is fantastic too for track usage.

For guitars I used to De-Ess (or simple Dynamic EQ) them before, but since last year I started doing steep Low Passing on them to remove all those sizzle and let the amp recreate the removed content. After that Low Pass again (not steep, just a bit) and sent to reverb with a lot of high end content, the Lexicon 480 SILICA BEADS preset. Much smoother High-End. Sometimes it works, sometimes it wont. Just some idea for you to experiment.

Then there's doing the exact opposite which is doing Expanding combined with High Shelf cut, Luca Pretolesi style. Acustica Audio JET. Goddamn that thing is awesome. Pro-MB can do it too.

for real ? I never do -2dBTP ever I think. Need to recheck them all back. I'm at -1db for quite sometimes, dont know about the newest standard. Thanks.

I believe it does for heavy limiting but I only TP limit around 0.3-0.5db GR with TP settings and the signal only hit the brickwall every once in awhile so that should minimize it a lot. The heavy limiting was done without the TP settings ON. That's one of the reason I'm using 2 limiters.

 

Last edited: Jan 16, 2024

Bruv, if you have the time, share some tips for us. I don't think we have a lot of MEs here.

 

Didn't they also work for Arturia at some point?
Either way, so far I love everything they released. 78 and Mu are amazing compressors.

 

As for HF limiting, I don't like the sound of no.6.
I have a better way, but I can't tell you because it's my audio teacher's secret.
It needs to work faster and you need to know that it's not literally limiting the high frequencies, but rather bringing about an extension of the bass, adding more dynamics to the track.
Some mastering engineers achieve similar results by using compressors with specific features in parallel. Or a transient processor.
Many MEs say less is more, but that term applies to experts who know all the secrets.
This is how great MEs create good sounds with just EQ and limiters.

 

As far as I know there's no clipping stage at all, unless you enable drive or have a super lower lookahead on a super bassy audio source. I've gone through the manual multiple times and through plugin doctor and I can say that it takes a while and a really good amount of gain reduction to start distorting anything. Idk if its 3rd harmonics tho, need to check that out

 

I didn't do anything, just set the big center knob to 0, turn off the punch, drive to 0. Lookahead is at 20ms which is a little more than enough for fast transients to be detected early. +1db Input, -1db Output :



Hammerstein : Yellow is G3 (third order), Orange is G5 (fifth order). So they are odd harmonics. They showing up on the lower frequency only, suggesting the limiting is done multiband. Mid and High are cleanly limited without the clipping.



Low End clipped :



Mid to High cleanly limited.

 

Hi Stevie, I wanted to tell you that.
I've been going to the sister site and here for over 12 years but have never created an account because I don't have the time to write. I'm doing it today to thank you for everything you do here for this community, especially for all your passion, the Acustica thing + your many knowledge sharings. Sincerely thank you for all !

And I would like to take this opportunity to thank everyone involved in the creation of the sister site and Audiosex, as well as all the participants here, without them my life would never have been the same.
Thanks for all ! Cheers from Paris and let's make some good fucking music !