Mastering Mastering?

LCVL · 2025-01-17T09:10:03.812Z

I’m not trying to convince anyone.
Actually it’s pretty simple. If you work in professional audio, ask a professional mastering engineer if a limiter is enough for the job these days.

jm2c · 2025-01-17T11:10:58.150Z

Again, professional MEs are not mastering music they made themselves. Anyhoo, I’ve said all I have to say on the subject.

BLKrbbt · 2025-01-20T05:24:18.635Z

@LCVL

@jm2c

Very spirited discourse! Well done… the passion show in you both.

As in music, there are no rules.

And all things are true at once.

knobgoblin · 2025-01-20T17:44:01.241Z

Out of the thousands of projects I’ve mastered only maybe 3-4 have been mixed well enough that only slight limiting was needed. Generally it’s 80% eq work 20% dynamic processing and saturation

bwo · 2025-01-20T18:10:02.698Z

I was trying the learn from machine learned / early ai for mastering quite a while ago but eventually I started using presets and found those way better for learning and understanding what is actually going on. I use a number of mastering tools as part of my process these days so my final mastering stage really is just the louder than lift off silver bullet vst maybe a little ozone if I run into some trouble harmonics… that and a tiny touch of maximizer to make sure I get somewhat coherent levels across tracks. Silver bullet is great though… basically all I need

zhora · 2025-01-20T19:14:24.933Z

IMO the most important thing about mastering is that an experienced person, with good full range monitoring, listens to my work and makes adjustments to make whatever I am trying to release sound coherent and as well as technically possible. If you are doing “mastering” yourself on your own music, understand that you will be missing a critical part that makes the process valuable.

szzdz · 2025-01-21T00:03:52.515Z

freeing up cpu that some plugins need for higher oversampling. on my project 2bus, ill use pro-l in 4x oversampling, but 16x on my master export, as my laptop cant handle it within the project

thumperjack · 2025-01-21T04:49:12.442Z

I can’t get enough of 1k cut into compression, it’s all the clarity I ever wanted without fucking the phase to fuck

Sternenlicht · 2025-01-21T06:02:07.807Z

I use the cliper from newfangeld eventide before going into Ozone. It has an A/B comparison so you can check before and after clip, that really helps to just clip enough without totally destroying the signal. These A/B stuff is really helpful also in Ozone.

J3RK · 2025-01-21T20:41:58.170Z

I use Ozone. I do NOT use the AI stuff, or the presets.

It definitely pays to try each individual function and learn how they work, and then apply them in MODERATION.

Subtlety is undervalued in mixing/mastering IMO these days.

I believe someone mentioned this above, but push a setting up to an extreme to hear what and how it’s impacting a certain aspect of your sound. Then back it off to almost no effect, then push it up very slowly to get a nice nearly transparent effect.

Use the dynamic EQ, and learn how to use bell and shelf curves to cut certain low frequencies a bit, and boost some low-mids for a little kick, tame the 15-20KHz a little etc. Use the vintage tape effect at 15IPS with a bit of input gain to get a nice saturation effect. Use the Exciter VERY sparingly. Use its learn feature in a dynamic area of your track to pick basic bands to work with. They’ll usually need to be tweaked a bit, as the learn is decent but not perfect. Find the sweet spots, then pick a triode, tube, or analog mode, bump up the range a bit, drop down the mix-send level a bit, then tweak from there.

Use the Imager! It’s great! Use all four available bands. Grab the area between 0 and around 100Hz, and pull the slider down to a negative region to basically remove stereo impact from the lows. This can help solidify the lows, get rid of a bit of phase-related muck etc. Leave the next band alone more or less. (bump slight above or below zero to taste) Boost the next two (mids and highs) up maybe 20%. Turn on the stereo imager, set to mode I. Set about 8ms. You’ll get some nice stereo field effects, and it can crisp up your highs a bit as a side-effect.

If you want to use the multi-band compressor to add a bit of punch, or tame a range, use it SPARINGLY. Put it early in the chain. I usually go right after the EQ. Set the thresholds HIGH. Like just barely in the peaks. (after learning the frequency bands) Set the ratios, knees, attacks decays to your liking. If you want more punch to your sound, keep the attacks quick, and don’t overcompress or you’ll lose the transients. Try to keep the threshold just within the active peak region of each band. Don’t bring it down into the RMS level unless you’re using the compressor as an effect for some reason. You want to see the active compression just barely pushing down on the peaks of what you’re compressing. Go too far, and it squishes it into a brick of sound. It’s an overused term, but you generally want your music to breathe a bit.

At the end of the chain put the Maximizer. Once again, as with any mixing/mastering tool, learn the sweet spots, and use it sparingly. You want to BARELY hear the effect. If you hear it too much, to where you feel like it’s “ear candy” you may be pushing it a little too far. Let people do that with their own EQs when they’re listening to the music. You want to provide a good frequency range, but if people want to sweeten it further, they can, and probably will just by listening on colorful sound systems. So as with the compressor, watch your threshold levels, don’t push the gain too far. Depending on your source material maybe 3db to 5db. I generally like the IIRC3 setting, and use the “pumping” setting as a default template to work from. Adjust the soft-clipping to moderate to low settings. Choose the band based on the tune you’re working with. Low, medium or high will depend on the type of music. If you’re say, doing some bass heavy minimal techno, use the low setting, but keep it moderate. Boost the transient impact a little bit, keep the attack low, (so use the “fast” setting).

You want your peaks hitting the -.01db level pretty regularly, but not constant hammering. You want your RMS levels to be healthy, but not overbearing.

That’s my process for using Ozone for this. It really does take practice, restraint, and just listening, re-listinening, letting your ears relax in between listening to not get over-fatigued, etc. Trying on as many things as you can. Headphones, monitors, phone, etc. then go back in and re-tweak a bit if needed. It’s not difficult, but it does take repetition and practice to get a process down.

This is just my process for my styles of music, but it works pretty well for me at least. Thought I’d post it.

survivors-guide · 2025-01-21T21:55:33.552Z

+1 for keeping it minimal when mastering. If using fabfilter plug-ins I highly recommend setting the metering to lowest setting, eg 3db in pro Q so you only cut or boost +/-3db max. Usually anything more than that should mean going back to the mix, with limiting being the exception. I rarely get a balanced master when I feel like I have to cut more than 2db.

Mid-side processing turned out to be my friend, as well as shelves with little dips/boosts before the cut.

Slightly off-topic, but does somebody know a 4-band imager plug (like the one in ozone)? I don‘t wanna buy ozone advanced just to have this as a standalone plug. I wouldn‘t need the stereo enhancing as I find it unpractical due to phase issues. But I’d really like to have some independent width control!

White_Noise · 2025-01-22T02:32:21.700Z

J3RK:

Use the Imager! It’s great! Use all four available bands. Grab the area between 0 and around 100Hz, and pull the slider down to a negative region to basically remove stereo impact from the lows. This can help solidify the lows, get rid of a bit of phase-related muck etc.

I’m don’t want to pick on you if you’re getting results you’re happy with this way… BUT I have had some instances in the past where I did this and the artist came back with their notes complaining about the low end being changed. I think it’s better in the newer versions of Ozone with Recover Sides settings, but just folding stuff down the mono nilly willy is not a good base policy IMO.

I set up and run the other bands, and then I audition what’s left in my low band and if there is anything down there that sounds off, then I might narrow it this way. But if the issue is only in one channel or only in certain parts of the song, I might do L/R eq, M/S eq, or some variation of dynamic EQ, or automate the control so it’s not on all the time, instead of affecting the entire track.

If you want to dig really deep into just this problem (how to make bass mono compatible without losing side information and/or affecting the stereo image) and nothing else, you can get Basslane pro from Tone Projects. I have only needed the toolset that offers a handful of times, but when it is what I need I’m glad I have it.

In fact, not to talk down on Ozone (because I learned mastering in there and I still use it on every master), but I think it’s strength is it is the best tool for learning mastering ever. It gives you access to a really competent version of most of the tools that an ME is going to use regularly, and is even best in class in certain cases (the Exciter is unmatched IMO, and their tape and vintage EQ are really good models too, and Maximizer is a favorite of a lot of MEs). But, if you want to have the BEST tools for the job, I think you need to pick up some dedicated other plugins in addition - specialty EQs, compressors, limiters, and some tools like Basslane as an example.

White_Noise · 2025-01-22T02:35:16.393Z

survivors-guide:

Slightly off-topic, but does somebody know a 4-band imager plug (like the one in ozone)? I don‘t wanna buy ozone advanced just to have this as a standalone plug. I wouldn‘t need the stereo enhancing as I find it unpractical due to phase issues. But I’d really like to have some independent width control!

It’s not multiband in the same way as Imager, but the other killer stereo width control plugin is DrMS by Mathew Lane. TBH, I don’t fully understand it, but if I cannot get what I want with Imager, this is the one I go to. If I weren’t already using Ozone for other things, this would probably be my default stereo width plugin.

J3RK · 2025-01-22T03:46:17.029Z

I don’t do anything willy nilly. However, when someone is asking for a magic sound good tool, (which doesn’t exist) I was making some suggestions for things to try.

I actually do get good results with the imager, but that’s because my sounds are already structured in a way that will work well with it being set up on my master chain.

I was just giving a brief recommendation for a possible chain, and how to use it subtly.

White_Noise · 2025-01-22T04:04:07.846Z

Apologies, I think your chain is a really good starting point! Overall, I’d say it’s shockingly similar to how I started out and did some of my first work for other artists. I just wanted to point out with that particular step I have had issues in the past and give some insight as to how I handle it now to prevent said issues. Sorry if I came off as trying to “correct” you or something, I wrote that quickly between dinner and tax class.

J3RK · 2025-01-22T04:31:59.668Z

No problem at all! It’s a good point actually. The imager CAN definitely cause some problems if it’s just slapped on.

I just usually have my panning all done and frequencies pretty well defined before it ever gets to that point, so adding it in the higher frequencies makes a nice spacious sound.

I didn’t think to mention that, so it’s good you pointed it out!

Edit:

Also, I do occasionally do SOME things willy nilly. But, in all the best ways….

knobgoblin · 2025-01-22T04:43:27.631Z

Another thing I’ve come to be wary of is having too many mid/side splits in the chain. 1 is invisible, 2 usually still is super transparent, but going beyond that can sometimes lead to a loss of definition in the stereo field, like the space between the mid and the sides gets hollow. Same goes with multi band processing. One or two can be very useful (though I try to steer clear of it completely these days) but start piling on multiband stuff and things can deteriorate. Generally when mastering for other people I try to keep to the motto of “Do no harm”

J3RK · 2025-01-22T04:53:45.658Z

Yes, I suppose you could carefully keep track of all your bands at all stages just right, but that would be a total pain too! I guess you just kind of have to decide where you want to direct your effort. Early in the process, or trying to “fix it in post”

My tracks are generally simple enough, and from few enough sources that I can do most of my work in the sound design phase, then just touch things up a little bit later.

To your point about too many multiband layers, I think having several, set to slightly different bands produces almost a quantization effect at all the overlaps.

knobgoblin · 2025-01-22T05:02:12.474Z

Ah, sorry, didn’t mean for that to be a reply to you or your process, just things that I look out for in my own signal flow at the mastering stage. I don’t worry about having a bunch of multi band or mid/side stuff at the mixdown or composition phase, whatever works there works. Talking strictly when you have a stereo bounce where someone spent a lot of time getting their mix to sound the way they like. I try to be respectful and subtle about what I do to honor their artistic vision and only use heavy handed tools when I feel like something needs “fixing”.

J3RK · 2025-01-22T05:10:43.089Z

Nothing to apologize about! I think having as many points of view, data sets, techniques to pick from and try is quite beneficial to someone learning to do this sort of thing.

It’s going to be a little different for everyone, the type of music, the tools, the various stages of the process, etc.

I’m always looking for more points of view from everywhere from true professionals, to equipment designers, to novices, to small time studio tip videos to other musicians. Doesn’t matter how big or small. It’s all data. I like to present this or that thing that works for me, but in no way do I think my way is best, an all in one solution or anything of the sort.

I do a TON of my work in the sound design portion of my process. I actually have very little to do afterwards later. I’m also not making pop hits. I make music for my own amusement and to share with whoever wants to listen. I want it to sound good, but I’m not going to expend huge amounts of effort to make it absolutely flawless.

I do a lot of what I do by ear, feel, etc. I don’t work like a “proper” engineer. I approach everything as an artist, including my circuit designs.