[Moseph]

Originally Posted by Sad But True

So just mix to reasonable levels and the master fader will retain all your dynamics now matter what the volume, should've realised that.

This isn't exactly true, but ends up being generally true in practice. Like I said, in the digital domain (especially at 24-bits) you have a lot of dynamic range to play with above the noise floor. Most popular music styles (even the "most dynamic" stuff) don't need more than about 20 dB of dynamic range (peak). Since you're working with 90+ dB of dynamic range in the digital realm, you can knock down by 3-12 dB and diminish your volume without having to actually adjust the dynamic range (you're still a good 80+ dB above the noise floor).

Originally Posted by Sad But True

Do you put limiters on any individual tracks at all?

Sometimes. You might even say "often", depending on what I'm working on. Again, the process is really about figuring out what all the components in the mix need, figuring out what tools you have at your disposal, and then figuring out what they actually do.

Originally Posted by Sad But True

At the moment I'e got one on my kick so it's all huge and stuff but will this have any repercussions later?

Yes. Though it's impossible to say if you're going to end up caring or not. I realize that's a bit cryptic, but the nature of mixing is that each song will have its own set of circumstances.

Originally Posted by Convectuoso

How you get high RMS levels is beyond me, I'm not a Mastering engineer so I shouldn't give advice.

Conceptually, "getting there" isn't hard. RMS levels are just a measurement of volume: turning up will make it louder. The reason compression comes into play is because RMS levels (which more closely reflect how we perceive volume) are actually lower than peak levels, but the machinery doesn't care about the difference: it's gonna have clipping/distortion whenever anything maxes out. In the digital realm, this sounds pretty horrible, so dynamic range compression/limiting is used to tame the peak values while still pulling up the RMS volume.

That's all there is to "get high RMS levels." The hard part is doing it in a way that doesn't sound awful.

Originally Posted by Sad But True

This is what I'm trying to achieve but I do still like a mix to really hit home when necessary, so I'm trying to walk a line between nice dynamics and hitting hard, and with my limited experience and non-existent mastering abilities I'm struggling.

Here's the counter-intuitive thing: making it loud removes the ability to have it be hard hitting. It's the contrast between the soft parts and the loud parts that can give you a "hard hit."

Think of it this way:

(01) Assume you have a track that is squashed to death by compression. The quiet parts are maybe 6 dB (RMS) quieter than the loud parts. If you listen at a level that is comfortable for the quiet parts, when the loud parts kick in they'll be about 6 dB louder.

(02) Assume you have a track that has more dynamic range. We'll say a difference of 12 dB between the quiet and loud parts. If you listen at a level that is comfortable for the quiet parts, when the loud parts kick in, they'll be about 12 dB louder.

12 dB can be a lot louder than 6 dB, depending on where your starting point is (the ear tends to care less about the difference as your SPL levels get higher).

If you want to actually hear this in action, try the following:

(01) Do a YouTube search for "Loudness" and "Death Magnetic." When that Metallica album came out, the commercial release was squashed to hell, but oddly enough the Guitar Hero versions were pre-master. Some folks made a bunch of videos demonstrating the differences. When you listen to them, don't crank your volume, make it loud enough to hear everything well, but not so loud that you need to raise your voice to talk over it. Then focus in on the kick, snare, and bass guitar parts. What do you hear?

(02) Do a comparison of two albums of relatively similar styles: one from the past decade, and another one at least 10 years older. My recommendations are respectively Songs for the Deaf (Queens of the Stone Age) and Electra 2000 (Hum) respectively (okay, so there's only a 9-year difference, regardless it works because I'm cherry-picking). Line up two tracks side by side in your DAW and switch between them. In my example, QotSA is way louder than Hum. Then grab an analysis plugin, and find the difference in RMS level between the two. Turn down the "louder" track until the RMS levels are pretty much the same. Now turn up your speakers by the same degree. Pay particular attention to the drums (and usually the bass). What do you hear?