New Mastering Challenge - May 2013

[Compyfox]

I think the trick in this particular example is to first tame everything that is too prominent, then enhance, then raise the loudness.


In a nutshell (for me at least):

Tame:
Dynamic EQ and LowCut

Enhance:
Linear Phase EQ
Parallel EQ
M/S Stereo spread, mono compatibility

Loudness Raise:
Brickwall Limiter


There you go, 5 individual plugins maximum. Actually 4 if I wouldn't have used a Dynamic EQ on the bass drops, and only three if I wouldn't have added "some" fairy dust to the signal.

So it is possible to take care of most issues, or enhance a good mixed track with only 3 modules.




I went away from multiband compression years ago (only used it to "tame" too strong frequency sections, and only about 2-3dB GR maximum). The times of uber-loud productions and "evened dynamic content" are over. Let the music breathe again.

[Davias]

Is there a strong difference between dynamic eq and multiband compression (except the shape of course) ?

[HanafiH]

The sub "rumble" is not random noise or a mistake. It's very obviously a drum/tone put there on purpose. It's clearly syncopated and makes sense, thus it's part of the track and shouldn't be removed, it could be tamed a little but I suspect that if you hear it at exaggerated volume levels it means you have room modes which exaggerate it's length and strength.

It's the singer stroking a heavy mic on a long boom, which resonates when he does, in my opinion.

It occurs at around 26s and around 44s, which is a pretty spectacular syncopation if it's deliberate.

edited for spolling and adding the file below

http://www.filefactory.com/file/5zmbux3 ... _poot_flac

This is a very short pitch-shifted and filtered version to illustrate what I'm talking about - it sounds like somebody blowing across a milk bottle in this file. It only happens twice and could be (now I've heard it like this) the bassist fooling around. It isn't musical, it's off key.

[HanafiH]

For me the mix was almost perfect

I tend to agree with Compyfox when he said he'd send the mix back for other reasons. It's suprisingly hot in the mids for a pro mix and that makes EQing it a real challenge because you either boost and wobble the phase or cut everything else and reduce the signal quality. Each to her own taste, though. My mix is too hot in the lows, to hide the flaws chewed over elsewhere in this thread.

[bmanic]

The sub "rumble" is not random noise or a mistake. It's very obviously a drum/tone put there on purpose. It's clearly syncopated and makes sense, thus it's part of the track and shouldn't be removed, it could be tamed a little but I suspect that if you hear it at exaggerated volume levels it means you have room modes which exaggerate it's length and strength.

It's the singer stroking a heavy mic on a long boom, which resonates when he does, in my opinion.

It occurs at around 26s and around 44s, which is a pretty spectacular syncopation if it's deliberate.

edited for spolling and adding the file below

http://www.filefactory.com/file/5zmbux3 ... _poot_flac

This is a very short pitch-shifted and filtered version to illustrate what I'm talking about - it sounds like somebody blowing across a milk bottle in this file. It only happens twice and could be (now I've heard it like this) the bassist fooling around. It isn't musical, it's off key.

It's curious you are still convinced it's some mistake/noise.. To me it sounds very clearly as a 808 type kick without any attack. It's just a sub note and it IS in tune with the song (first note is the 5th from the key of the song, the 2nd is a 6th).. not sure what you are on about there. It's also rhythmically exactly in place so why on earth would the singer blow into the microphone at specific, rhythmically sensible times while he is not singing? The first one is a syncopated "chakka cha-BOOOM" fill like thing. The other one a syncopated "pa-boom!" thing.

Seriously, you need to upgrade your monitors/headphones if you can not hear it for what it is. Listening to your pitched version makes it even more clear that it's a synthetic tone. Blowing into a microphone doesn't sound anything like that in my experience.

Anyhow, you are of course entitled to your opinion but I'm 99% sure you are seriously off on this one. Of course we can't know unless somebody asks the artist/producer.

Cheers!
bManic

[Compyfox]

Is there a strong difference between dynamic eq and multiband compression (except the shape of course) ?

To me, yes.

Dynamic EQ changes turns down (or up) the signal past a given threshold and selective frequency (the EQ setting). It leaves the dynamic content otherwise in tact (no compression) - so you could see it as frequency selective volume control in this case. Which is a good technique if you want to let the density of the signal as it is.

A multiband compressor reduces the dynamic content according to threshold and ratio. Granted, even with a ratio of 1,5:1 it would be subtle, but it messes with the signal regardless. So it's not a plain volume control of some sort, you just reduce the dynamic and hope that the endresult is good.

[bmanic]

For me the mix was almost perfect

I tend to agree with Compyfox when he said he'd send the mix back for other reasons. It's suprisingly hot in the mids for a pro mix and that makes EQing it a real challenge because you either boost and wobble the phase or cut everything else and reduce the signal quality. Each to her own taste, though. My mix is too hot in the lows, to hide the flaws chewed over elsewhere in this thread.

Well in my experience this mix was pretty well done. Sure, it WAS a bit hot in the mids (a bit nasal even and quite stingy at around 2.1kHz) but those kind of problems are very, VERY common in mixes I've got for mastering. It's perfectly possible to create an awesome master from this mix.

Song volume was well balanced, the overall balance of each instrument was "pro" level. So many here get hung up on what frequency, what compression, etc. etc. when the most important thing is the balance between the various instruments. For a typical pop-ballad/pop-sugary-thingy it's an almost "textbook" mix, in my opinion. It just needed a bit of gluing together and perhaps a tiny bit enhancement of the middle vocals (a gentle 0.5dB wide boost from around 3kHz would do just fine.. then taming the harshness at 2.1khz with a narrow eq).

But yeah, I am of course basing this all on my headphones as I don't have access to the studio at this time but I've used these for more than 3 years now and do feel quite comfortable in trusting what I hear on them (except dynamics perhaps.. it's still a real challenge to judge any kinds of dynamics on these).

The most important thing when it comes to a good balanced mix like this is that it all comes down taste. It's quite difficult to completely screw this up. So from that point of view you can ask yourself and Compyfox why you'd say to the client that the mix sucks and you want a new version. Is it REALLY because it has serious technical issues or is it because it doesn't conform to your taste of how such a mix should be?

Cheers!
bManic

[bmanic]

For all people reading this and having not a single clue of what it's about. Here is the ORIGINAL unprocessed mix from the Sound On Sound page.

Unprocessed mix, 44kHz .wav

OP can add this link to the first post so that it's easier for people to find it. I'll keep hosting it for a while.

Cheers!
bManic

[Compyfox]

The most important thing when it comes to a good balanced mix like this is that it all comes down taste. It's quite difficult to completely screw this up. So from that point of view you can ask yourself and Compyfox why you'd say to the client that the mix sucks and you want a new version. Is it REALLY because it has serious technical issues or is it because it doesn't conform to your taste of how such a mix should be?

Like I said, I'd get back in touch with the musician/engineer to raise an instrument or two - since I personally feel that "lacks" in the whole concept. As you pointed out, that is a personal opinion, maybe even preference of style. Sometimes this can improve things, but it can also go totally haywire.

But we don't have access to that - the challenge is all about a final stereo mixdown, available for all to edit to their needs.




In a "closed door" situation, I would talk with the participants, get feedback, weight in the good vs bad, share opinions. But in this case, I just work with what I get.

Now imagine a future challenge to be with STEM mixes. This will be one hell of a funzone. Why? Because everyone does see something different in the mix, will add a special ingredient or two. And then mix to taste rather than focus on what's already "there". We then don't talk about mastering anymore - we talk about (creative) editing.


So even with this (current) particular challenge, we can't find a definite winner. Because there are no guidelines as to "how it's done right". And it will never be.

It's as if you say a carpenter "dude, your desk seems to be built correctly, but I don't like the look of it". It's always a matter of taste.

[Hermetech Mastering]

I'll have a go but all the EQ and Compression will be done with the analogue outboard loop. Is that allowed?

[Compyfox]

I don't see a problem in that. As long as it's 5 modules max.

[Hermetech Mastering]

OK, went ahead and did it anyway! Treated it like any other client's music (i.e. tried to respect the music as much as possible, and not do anything differently because it was a "shoot out").

I'd definitely be asking the client about the two bass booms, as to me they don't sound very natural with the rest of the track, but of course I would respect the client's wishes if they were intentional.

FLAC Link:

http://darkflame.hermetech.net/HermetechMaster.flac

Chain:

1) Out Crookwood transfer DAC into analogue chain.

2) Into pair of Chandler Germanium Compressors set for vibe and to "groove" with the track, no more than 1.5dB of GR.

3) Thermionic Culture Pullet EQ and Chandler TG2 combo, slight boost around 2k to help the vocal "pop", slight cut around 340Hz to enhance clarity.

4) Dangerous Music Bax EQ, HPF at 30Hz, Low Shelf +1.0dB at 74Hz for body, High Shelf -1.5dB at 3.4kHz to reduce mid range build-up/add warmth, LPF at 28kHz.

5) Into Crookwood capture ADC. Did NOT clip the converter. Back in digiland I reduced the Mid (of M/S) by 0.2dB for very slight stereo width enhancement.

6) Into Elephant limiter, max GR was 2.9dB on loudest peak, final RMS should be around standard K12, perhaps a little hotter in places. Am guessing if this were a commercial release the client would be expecting it hotter, -8 to -10 RMS, but for the sake of sound quality, and because we don't have the client to answer to, I didn't go there.

7) Sonoris Dither (down to 16 bit, as capture is 24 bit).

TIA if you take a listen/comment,

Gregg

[Edited for typos, was an English teacher for 12 years!!! ]

[Hermetech Mastering]

PS that's five right? Compression, 2 EQs, M/S, Limiting.

[Compyfox]

That's 5 modules indeed. I don't count SRC and Dithering, or monitoring/analysis either.

[Hermetech Mastering]

Cool, it's pretty much my standard/fixed chain.

I thought the mix was pretty good, and wouldn't have asked for any revisions, (but I would have mentioned the bass booms).

PS: OT but downloaded your white paper on "new" K-System CompyFox, and am looking forward to reading it. I use my own "bastardised" version of the K System for metering/SPL in the room. If I stuck to standard K System 90% of my clients would be asking for it louder!