I’ve been noticing that my final mp3 isn’t quite as loud as some of yours. I’m normalizing up to -.1db, so it’s not like I’m leaving a lot of headroom.
So, I messed around a bit with compression to see what that’ll do. It certain pumped everything up to a more in-your-face stratosphere. But then the mixed lost a little bit of it’s subtlety.
My questions is: Is that the trade-off we face with final compression of a mix? To make mine as loud as many that I hear on the site (and on, say, radio or albums), do i need to compress the mix up to a point that fills the spectrum? I feel a little bit of a cringe when I see the tips of the waveform squared off. Is that something to completely avoid?
I feel a little on the fence about that. It seems to lose something when you push everything so there’s almost a full spectrum when looking at the mix. How do you guys strike a balance?
Just so you know what I’m looking at. I use Adobe Audition for my final mix. I’ve got several compression tools at hand: single band, multiband, tube modeled. I also have a million mastering presets in izotope 5. What do you recommend?
I recommend limiting. Maybe that’s what you’re talking about, but just in case, this is an excerpt from the link above.
You can turn your compressor into a limiter by putting the ratio to its highest value, or at least a higher value than 10:1. By having such a high value, any sound that reaches the threshold is completely squashed down. The threshold therefore acts as the limit of the sound signal and limits the signal from going any further.
I’m not advocating the limiter over the compressor in any way, as both processors serve their purposes. It’s just good to know that although a limiter sounds pretty destructive on paper, cutting off large pieces of your signal, in the end it can actually sound a lot cleaner. Not only great for mastering and overall level boosts, but can help with various mixing tasks as well.
Hope that helps. Not that I’ve achieved anything close to the “perfect” volume. Have you looked at Waves Mastering plugins?
Try some limiters (Fabfilter ProL, Oxford – very gentle sound). And before it you can use soft compression with ratio about 1,5-2:1. And of course you must have some headroom of your mix before mastering (3 db) Good luck!
Every song is different. Really. So is every sound. So it’s better to gain a better understanding of the entire purpose and process of mixing and mastering, along with the tools used to get the jobs done. There are also many standards and expectations to be considered depending upon how your audio will be used. So giving general sweeping advice to a question like this one, is something I don’t normally do. I have, however, just sent you an email which may help get you where you want to be with your audio quality here on AJ.
Best of luck out there! You have some good ideas, and a good sound.
Personally its taken me quite a while to find the right tools and method for mastering to bring things up to an acceptable level so I can compete with everyone else. And I still consider myself a total novice, but I use my ears, and I’m happy with the results I get.
I use waves plugins mainly. A couple of stages of gentle compression to begin with, then EQ (either multiband compressor or a pultec or mastering grade EQ) then limiting with Something like Waves L2/L3. After that I would use the Aphex aural enhancer, or Steven slates FG-X to counter the negative effects of limiting / compression to help transients pop a little more again.
At the end of the day, purely looking at the waveform will probably only depress you, as you realize you’ve had to squeeze the living daylights out of your beautiful music with all its dynamic range, in order to get it loud enough to compete with everyone else.
I would suggest you start by adding several layers of GENTLE and transparent compression, then limiting, and above all else, use your ears.
N.B. Everything below is all stuff that works / has worked for me. I don’t consider myself a professional engineer, but I think my mixes/masters sound pretty decent. All of the points made above before me are also perfectly valid. It’s a broad question, but there are quite a few starting points.
I’d recommend checking the Audio tutsplus network as there are some fantastic articles on mixing/mastering there. Anyway, here goes:
An old cliche, but if your mastering isn’t working out it can often go back to the mix.
Also, it’s not necessarily about filling up the waveform and having your file look like a NIN track, it’s about having balance and clarity. Loudness doesn’t actually entirely come from the mastering stage, perceived loudness stems from the quality of the arrangement (which is where balance and clarity starts from).
The best masters / mixes and the tracks that sound the loudest aren’t necessarily the ones who have the best ‘loudness button’ in their gear, but rather, the mix is so well balanced and everything is so clear, that mastering accentuates that even more.
One of the biggest mistakes that many composers / engineers / producers make is to try and add stuff to make things ‘louder’. Instead, the real trick is to make one element sound really really good. There’s a reason why rock bands only have one bass player! Hans Zimmer is also a master of this, listen to Inception. It’s just layer upon layer of instruments, but each playing in carefully defined frequency zones. This is one of the many reasons his music sounds so massive (the other main one is how it’s recorded). Anyway, point being, arrangement is important. Moving on….
If you’re only using your eyes to mix/master (looking at waveforms or frequency analyzers) then you’re not going to have success. The graphics are there to inform you visually of what you are hearing, not to tell you what you are listening to.
One of the biggest things in getting a mix ‘loud’ is bass balancing. One of the first things I do if my mixes sound squashed and horrible is to actually look at that instruments that aren’t ‘bassy’ instruments, and EQ a lot of the low end out of them. Bass fills up tons of room in your mix really quickly if you don’t keep it under control.
For example, if you have a violin playing a high melody, you can afford to scoop out most of the low-end. Individually, that might take away from the quality of the instrument, but when you put it with a bass instrument…1) the bass shines a little more because its frequencies are not competing with the violin, and 2) what the violin lost when you scooped out the low end, is now supported by the bass anyway. This is just one example.
Another might be when dealing with distorted guitars layered with bass. When you record a guitar amp with distortion (for example, power chords) – there is going to be a TON of rumble and superfluous sound that you don’t need. You have to get rid of it by using a pass filter. Whatever you lose is going to be supplemented by the bass anyway, so it doesn’t matter.
The high pass filter is a very very good friend.
A lot of perceived loudness is not to do with the quality of the mastering (though it helps), it’s the quality of the arrangement and how balanced things are.
As for the actual mastering stage, everyone will have their own opinion on this, but here’s GENERALLY what I do.
1) Make my mix sound as close to what I want the final track to sound like as possible.
2) Export my mix to a single 24-bit WAV file, leaving 4-6 dB of headroom. If you don’t leave headroom, you’re not giving much leeway for your mastering tools to do your thing.
3) Set up a new project – ‘Track 1 Master’ or something, and import the WAV file you exported. I like to set up different projects for when I’m mastering. I think it’s important to separate the two processes. I know some composers who write with a mastering plugin on the master output all the time. Sounds great, but you’re also not hearing everything clearly.
4) Generally, I’ll use EQ first. I use Fabfilter Pro-Q, as it has a great spectral analyzer, and it’s GUI is amazing and very easy to use. You’ll also feel like you can really sculpt the sound. The problem areas with most music tracks are too much rumble (40Hz and below), low mids (200-400 Hz / the ‘mud’ zone) and high mids (around 3000Hz, the ‘irritation’ zone). I’ll generally isolate these frequencies and see if I can find anything that’s really bugging me, and cut accordingly. But this is just the starting point, depending on the track/style of music, I’ll have to look elsewhere too. I spend most of my time here.
NOTE : Getting a good mix/master is more often than not about what to take out, rather than what to boost.
6. When it comes to compression – if needed – , I’ll look at using a multiband compressor, the one in iZotope suits me, but there are plenty of others. With mastering, you just want to be careful how much you’re using. You shouldn’t be using it to cover up weak elements of your track. If the track doesn’t ‘feel’ right before you master it, go back into the mix and work on that instead (or the arrangement).
7) I’ll also check the stereo width of the track, mainly by using my ears and supplementing it with a graphic tool like the free Flux Stereo Tool. For example, whenever the kick drum plays, you’ll see a big vertical line in a stereo analysis, and then when – for example – a pad kicks in, the stereo spread will be wider. Ideally, you want a nice contrast between the vertical line, and the width of the sound.
NOTE : Remember you’re doing all of this, while constantly listening for all of the other stuff already mentioned (boomy basses, boxy mids, irritating hi-mids).
8 ) I’ll then – depending on the music – look at adding in a little saturation ( PSP Vintagewarmer 2 is a favourite). I generally do this for more synthy sounding stuff rather than orchestral works as it just tends to make everything a little less sterile.
9) If it’s orchestral music, I’ll think about throwing in a tiny amount of reverb on the final track (in addition to what might be there on the mix), but this really depends on the track and what mood I’m trying to convey.
10) The final step is to use a limiter. I like Fabfilter Pro-L, but there are plenty of others. This is the step where you can really push the volumes hard if you want, but where you also kill all the dynamic range that you’ve worked so hard to create in all the previous steps. If you’re looking for a total wall of sound, then you can push really hard (but everything will sound squashed). Less is more here in general.
Overall, just use your ears (and supplement them with the visual information that the plugins give you), and trust your judgement. You can aid your judgement by listening to commercial records all the time. One of my favourite things to do is to go out and buy all the Grammy-nominated ‘Best Mixing / Mastering’ albums. It can be a real education.
I’d also recommend saving the money to get a top-notch pro to work on your track and ask to be in the studio during the process so you can watch and learn.
It’s important to remember that there is no definitive process, just the process that works best for you. But once you’ve found that, it’s good to stick with it, and refine the skills you have with the tools you already have. It takes years to become really good at this stuff.
Hope this helps.
......An old cliche, but if your mastering isn’t working out it can often go back to the mix.
Also, it’s not necessarily about filling up the waveform and having your file look like a NIN track, it’s about having balance and clarity. Loudness doesn’t actually entirely come from the mastering stage, perceived loudness stems from the quality of the arrangement (which is where balance and clarity starts from).The best masters / mixes and the tracks that sound the loudest aren’t necessarily the ones who have the best ‘loudness button’ in their gear, but rather, the mix is so well balanced and everything is so clear, that mastering accentuates that even more…...
Such an important point! The arrangement and original mix is SO important. Unless anybody has found the elusive “Turd Polish Masterplug 1.0 VST ” ?