Some nasty comments coming my way, for me simply expressing my opinion. Also being called a troll, which I find laughable. Look at my post on Queen's gig at the famous Elland road gig in 1982, which I attended for proof of that.
First of all I am a songwriter, musician and do a lot of mastering for various bands world wide, although I don't do mixing as that is an art in itself, which I have no interest in and get others to do mixes for me. I can give you links for my stuff too, if I need to prove I do have knowledge on mastering and the loudness war, which I detest. IMO British albums in the last 10 years or so have become unlisteneable, but we are talking about re masters here, which is a completely different subject. I also have a lot of respect for Bob Ludwig too and also Bob Clearmountain.
I will try to explain the mastering process, without trying to patronise any of you. Mastering engineers get a mix of a song, which will come in at a volume level of -9db, even lower in some cases, I prefer it to be -12db. They will also provide an instrumental mix and a vocal only mix. Those are only if the mix is particularly bad and those mixes will be used to fix things. A mix will initially be done at a low volume, based on the Fletcher Munson theory, which is quite technical, perhaps beyond me, but does work. Chris Lorde-Alge, who mixes virtually every rock song in the USA is a genius, he mixes songs at a volume which can actually be spoken over. It took me 30 odd years to do the same and only because he does it. No mixes or masters are ever done on headphones, no matter how good they are.
Because we are talking about re masters here, it's a slightly different process. They will do them from the flat track mix, which is the un mastered version of the origional album mixes. AKOM was recorded digitally, so there will be no clipping. The mastering engineer will then use the mastering tools he has, looking for sound that are too prominent, or too quite. He will then use subtractive EQ to achieve those sounds. In mixes, for example, if you can't hear the bass track (Queen and most bands record 3 bass tracks, DI'd, amped and distorted, then blend them) you don't turn the bass track up, if it can't be heard, you turn all the other tracks down. This applies for every track, turning tracks up is a massive no no. So back to mastering, subtractive EQ is the equivelant when you master. There will be conflicts, which should have been sorted out in the mix, but may not have been, i.e. frequency masking. Every instrument or vocal will occupy a certain space in the frequency range. However, if it is a bad mix a good mastering engineer can fix it, but may need the extra vocal mix and instrumental mix I mentioned earlier. Subtractive EQ, is making what's called notches (cuts) in the frequency range, for various tracks to be heard more. The kick and bass track, for example, may be conflicting with each other. that should have been fixed in the mix, via what's called sidechaining. Where the kick will 'duck' when the bass kicks in at the same time and vice versa. EQ can also be boosted at certain points, boosts are done at a wider range in the frequency range, cuts are much smaller, for precision. Nowadays you can hone right in on certain sounds, like hi hats, and you can actually change the sounds of those and even the snare, it really is that precise.
Obviously there is more to it, but in a nutshell, that is what they do. the last part of the process is the limiting of the track. That is where the 'loudness war' comes in. Limiting makes things louder, it is not the same as compression. Limiting will make the mix louder and it is how far you push that, that when listening will give you the muddiness, clipping, distortion whatever you want to call it. One of the best tests for listening to a song is the go out of the room test, far better than listening over headphones, that test will tell you more things about a song, i.e. can you hear every instrument in the mix?
The loudness war is due to 'dynamic range' which is sadly missing in most music today, sadly and annoyingly, IMO. Dynamic range is the difference between the loudest part of a song and the quietest part of the song. Off memory rock music should ideally have a dynamic range of 9db. The best example of this is Bo Rhap, very quite in some parts very loud in other parts. I brickwalled Bo Rhap, just for my own purpouses a few years back, as if it was done today. As you can imagine it sounded dreadful. But I would imagine listening in a car or a noisy workplace it would sound ok, as you could hear everything. Radio stations compress music for this purpous. Which is why we should wait for the new songs to be released before we judge them.
One last point, the masters and any CD cannot be clipped. Nothing can go over 0db in the digital domain. In the anologue days, it could and was used to great effect, to achieve that now, is a totally different process. So when you listen to a re master, it should be listened to, for a fair opinion, with flat EQ and there should be no red lights shown on any LED displays you have.
