Hello everyone, just some quick feedback from my end.
I just downloaded every entry that we had, went through the documentation and screenshots. Some things that really stood out quite positively, is that a lot of you were not shying away form experimenting this month. I think for the first time in the history of the Mix Challenge, I've read somebody using "Virtual Sound Stage" (which is basically a stage panner and early reflections creator). I've read about using the Brauerize technique (personally, it just never clicked with me), another one trying to run everything through outboard gear first and then committing to the sound. Then there were the ones that dove deep into the topic of this specific drum kit setup, while others went straight up go-to Pop Rock mixing techniques for the drum set.
It is great to see so much variety, and I do hope all of you could take away new ideas and improved your skill-sets in the process. `
Learning by doing, this is what this community is about.
Then I'd also like to address the criticism from Franz. In hope that I'm really not misunderstanding what you (Franz) try to convey here, then the following rant would be completely in vain. But it is a frustration I can completely understand, yet found a somewhat acceptable solution for myself throughout the years. It is not necessarily elegant, but maybe this helps offering a different perspective.
Franz wrote: ↑Mon Dec 21, 2020 23:57 CET
WHO has dared to send his music to a level of -16 Lufs ?
Anyway, these levels are automatically set to standards (CD or streaming).
In practice this is what happens: I first performed my "mix" at a level of - 17 Lufs. The sound is flat and of no great interest
(note that I still do not understand why ask to make a mix when the reference of the songprovider is completely mastered).
Then I build the "stems" which will serve as a basis for the mastering process. This is finished at a level of - 9.4 Lufs.
So a difference of more than 7 dB. Do you have any idea what that represents? An increase of 3 dB doubles the sound level, you see the rest ...
Anything that can be done in mastering within these 7 dB is STRICTLY impossible to achieve in the "base" mix.
The result: once mastered the song "sounds" !!!........
We do have this topic ever so often as to why "I" seemingly decided on -16 LUFS ILk absolute max rather than let people mix how they want.
The answer lies indeed in the current trend of "more dynamic releases" all over the web, audio CDs and co. Basically, going back to the days of early CD releases from the 1990s. Actually,
@Mork posted something in the Gossip Thread months ago: a blog post by German monitoring and signal analysis company RTW --
Worldwide Loudness Delivery Standards (Summarized). A very informative read, as this is currently the best overview on this topic (as in - collected in one place, rather than scattered all over the web). And yes, I will probably create/overhaul a Rule Addendum that "summarizes" this topic again. I just need some more spare-time.
Now what does that have to do with the overall topic at hand, and your problem with the mix vs a "master"?
A small dive into my work in parallel to the Mix Challenge:
While working on music for advertising in the last couple of years, I usually have to adhere to a certain loudness limit. The problem is, "perceived" loudness and therefore "feel" - the very thing you bring up with your argument. A more dynamically compressed, and therefore often also louder production, does sound subjectively fuller and better. And most clients will be like
"make it loud, make it fat, it needs to groove!". But is it really better?
Well, if the music is just "Background Blasting" for an ad, then that's one thing. Here, you can limit the dynamic a bit more (especially if the music is ducked away, rather than just... mixed along). Sometimes on purpose even, like it is the case for video games (prime example: DOOM Eternal from 2020, where the "in-game" music is heavily compressed to "cut through" the chaos that are sound effects).
But if we talk regular music, then it is a bit different. Especially if the de-facto future is, that heavily compressed/limited productions will be "pulled down" on radio, on TV, on streaming platforms, etc. (technical term: Loudness Normalization)
In fact, something that some of you might not have known yet... one of the first things that radio stations do if they get new music to broadcast, especially if it's straight from a CD or a very loud distribution format, is to pull down the loudness of the track by 6-8dB to not overload the compressor arrays from the station. In case of the infamous "Metallica - Death Magnetic" release, that was a necessity. If they had run it straight up through their setup, then it would have sounded even worse than it did -- especially during the climax section in the end (due to the CD being nearly -4dB RMS in that section!)
So what
can we do about this?
Actually,
learn how to "cross-check" how things would sound after distribution. That can be as simple as running your mix through a heavy compression array (that is volume compensated), and/or using a tool like Audified MixChecker Pro, AudioTools Speakers or even just a special tuned "band-pass EQ (with certain random boosts and cuts here and there) to get an impression how things translate to other mediums on top of that (if you don't want to re-render things over and over, to check on other devices).
One of the recent bigger projects I've worked on outside of the Mix Challenge, was "production music" for film and radio. I was also asked to create a CD Master as limited bonus release. And while we agreed on a certain distribution loudness that works on every platform (we decided on -14LUFS ILk for the whole album, Spotify and Youtube ready), we also checked how things would sound if they are over-compressed on purpose and/or played back from a specific medium/device.
For this, I've used a combination of VladG's Limiter No6 / Tokyo Dawn Labs Limiter 6 and Audified MixChecker. I ran two different gain modules pre-limiter to either reach about -10LUFS ILk (which would be louder than Spotify Loud), or get to a more moderate -16LUFS ILk (+-1, which would not only cover "Mastered for iTunes", but works great for a vinyl release as well). MixChecker was for checking how it would sound on different devices.
This resulted in me hearing a couple of things, that I could address in the mix before I raised the loudness to it's final desired target:
a) did I use too much compression during the mix (vocals for example, though mostly you notice that with snare drums)
b) is there too much bass that needs to be addressed or can/should it be fixed during mastering
c) can the music stand on it's own without excessive pushing of loudness, or does it need a small push with the help of a gentle summing bus process (glue)?
I
am an advocate for pulling back on excessive loudness. But I am used to how things sound on various volumes.
And this is a learning process:
Learn how to use your volume dial in the studio. Learn your mixing environment, already slightly compensate during mixing rather than trusting the mastering engineer (the next step in line), learn the pro's and con's of summing bus treatment (which
is allowed in the Mix Challenge, as long as it's musical, and some genres really benefit from that!). Switch volumes ever so often, listen to your music on "louder than comfortable" for a couple of seconds, listen to it super quiet, take a day off, listen again on smaller speakers, then again on your studio setup. This all comes with experience.
Which is why we do this monthly game, to experiment, learn from each other, based upon the same given parameters - without super tight deadlines. I get that you're frustrated with the fact that the demo mixes are usually already heavily processed. But this is just that... "demo mixes", to give us an idea how things could go, or went too far already (you decide).
Our goal is to make the best possible mix (and only mix) that there is, that translates well to every medium out there.
Mastering should finally go away from
"make things loud, fat, warm and groovy" and shall be treated again for what it originally was... looking that things are in check signal strength wise, check phase issues (would the track work in mono or near mono if you walk around the room and listen from a laptop or tablet computer, which is still important!), add a minimum(!) amount of fairy dust if needed, and prepare the content for it's final medium (Vinyl, Tape, CD, Streaming). Instead, it is still
"make it as loud as possible - the streaming service will adjust anyway". And then one is surprised why your edits/productions sound bad compared to everything else. This is IMHO not the route to go - not anymore.
I don't know if I ever brought this up on the Mix Challenge community, but please compare Carpenter Brut's "Turbo Killer" CD version to the official music video. The music video is more quiet, true, but also way less aggressive ear fatiguing due to less over-compression. This is a modern day example for things going a tad too far -- and one of the rare examples where you have an easy accessible A/B comparison.
So yes, call me old fashioned. Call me out on triple checking everything and that the rules are dreaded. It is fine. But there is a reason behind the madness. I don't decide on things completely out of the blue - some of these rules are based upon personal experience from roughly 25 years of being in the audio realm. From mistakes I made myself, stuff that was I was never taught in tech classes. This is why I try to establish a certain baseline for everyone, which makes learning easier, and focus on what is really important.
And again... who actually holds you off of using a volume dial? (especially after distribution)
The biggest question you should ask yourself is:
"does it 'groove', no matter the volume"?
With that said, I'll post the Statistics Sheet later tonight. My recent nights have been a bit too long.
I'll keep you folks updated.
