Tips on making your sound or song sound fuller.
Some of you have made a song and wondered why it didn't sound like a professional recording or why one song sounded more legit then the other. So I am going to tell you a few things on how to make a song or sound better.
The first step would be something called spacial positioning. The reason why studios are set up the way they are is so that the waves are played off speakers and hit the walls to bounce away form the engineer sitting right in the middle. he needs to have the left and right speaker perfectly in in a triangle or a V from where he is sitting. Why? Because they do that to build a auditory illusion of when you listen to the song that it feels you are watching the band right in front of you. When you have a song, each track should be panned as if you were looking at a band on stage or you are the band on stage looking out, or an orchestra where people are sitting in chairs. lets say you have a simple track with Bass guitar and Drums. The first thing you would do is record the drums. if you want to get super nity,grity,hardcore,producer,multi-billion dollar precise. you would edit the drums to be directly in the pocket of the beat before recording bass. But for now we wont do that. next is bass and then over it is lets say guitar rhythm doubled. Now for the guitar track, no guitarist when playing is usually sitting directly in front of you playing. He is usually off to the sides. For me I usually do it like looking at the band. So would pan guitar tracks to the right of the track just a air to give that illusion. Bass you can leave in center or you can pan to the left. But most of the times bass is there to keep time and the groove and isn't doing much embellishment for you to pan it. This is your choice though. The biggest problem I see is drums. Because even when the drums are usually right in the center. you still need to pan every part of the drum accordingly. if you have a drum fill going from left to right you need to make sure the drums are panned left to right which the correct parts of the drum. You don't want your drums playing a hi-hat when its dead in the center. You want to replicate that the drummer is in front of you and playing nicely. So of course you can have the kick right in the middle as well as the snare and make sure crash/hi-hats/cymbals/rides/chinas etc are all panned if being used in the tracks, as well as the toms he if rolls or fills. This will give a more full live surround feeling in a mix.
The second part I want to talk about is Eq and compression. Let me just say, to be safe. That you will 95% to 99% need to apply a EQ or compressor to every track you have. not every track is gonna be so superb in timbre for you to use it. Also you need to eq accordingly so you don't kills some ones dynamic hearing be having an extremely high note go off in the ear. Make sure you separate tracks and eq and compress accordingly. You don't want to put all of the drum parts on one track and increase bass kick with an eq because then when the crash is it, it will sound like someone is laying a hand on it or something. You need to be able to maintain that Christmas and head room with the crash. Compression is the one thing you need into making your mix sound "tighter" you need to have this where in case I have heard mixes where the mass and drums sound nice and all of a sudden about 30 db louder then the drums comes a guitar track that blew my ears away. I wont get into compressor too much to explain to you just know that you should maintain a threshold that doesn't blow speakers and hears and manipulate the ratio accordingly to how you want it and how it "fits" into your mix.
More on compressors, because I know this was one of the toughest things I've had to wrap my mind around so far:
Compressors have at least a few basic settings, plus some more options depending on what model/plugin you're using. The key settings for a compressor are: Threshold, Ratio, Attack, Release, and Makeup. Attack and Release are pretty self-explanatory - the time it takes for the compression effect to kick in or let go when the audio signal passes the Threshold level.
The Threshold level is the dB level at which compression takes effect. Once the audio signal gets louder than the Threshold, the compressor begins to reduce the amplitude according to the Ratio control. The Ratio control can usually be set from 1 to 10 and will be marked at least by a label that says "1:n" or "Ratio: 1:n". This is the factor by which volume is reduced, the ratio between output and input: 1:1 is no reduction, so no change is made to the signal. 1:10 is the most extreme volume reduction, meaning for every 10 dB the input goes past the Threshold level, the output only becomes 1 dB louder. The key here is to not overuse the Ratio control, because settings like 1:10 tend to distort your audio, which is a Bad Thing (unless you really want that compressor distortion sound - it can be appropriate in some electronic music).
The Makeup or Gain knob is for amplifying the signal after compression has taken place. This is how compression can help normalize your audio levels. Let's say you have a signal which varies between -30 and -5 dB - that's quite a big difference, two and a half doublings in apparent magnitude. You can reduce this gap with a compressor. Set the Threshold knob to -30 and the Ratio to 2 (1:2). Now, for every 2 dB the input goes above -30, the output only raises 1 dB. A little math will show you that means when your input signal is at -5, your output signal is at -17.5 - we've gone 25 dB past the Threshold, resulting in a gain of only 12.5 dB: -30 + 12.5 = -17.5. Now you can set your Makeup Gain knob to +15 and your signal is at -2.5. Your signal now only varies from -15 dB to -2.5 dB. With more compression (a higher ratio and more gain) we could tighten it up further.
Sidechain compression is a topic for another day - essentially you can use an audio signal to trigger compression, but that audio isn't part of the output. Other differences your individual models of compressor might have are Root-Mean-Square (RMS) tracking vs. Peak tracking (you can read about that on Wikipedia), "hard" knee vs. "soft" knee or knee radius, or key filtering. These are less integral to the compressor's function but are all useful.
Wow, thanks a lot for the explanaition, both of you! I'm a hobbyist musician myself, though I'm not much aquainted with compression. The reason is that I'm making music almost purely the electronic way, so since I'm not recording much I haven't felt the need for learning how to compress correctly.
In other words, I'm lazy, but thanks for enlightening me a bit. :P
Also with that being said. with too hard of a ration control it is a very good idea to, if you have a very dynamic volume level in the track, to use automation with the compression. I have seen some programs not even have automation when automation is actually 1 of the key things that seperates a audio program from a Digital Audio workstation (D.A.W). So make sure you have a program that has automation,expandability,flexability,integration, and communication speed like real D.A.W before getting this precise in your final product.
Brandon Morris,
Steam Group: http://steamcommunity.com/groups/OpenGameArt
If you guys need any help on projects email me atBrandonmorris12@gmail.com. Pay is not mandatory and Im open.
Automation is pretty well integrated in the program I use (Cubase 5), so that shouldn't be much of a problem.
But since I mostly do VSTi's, I prefer to leave the insrtuments as they are and rather up the volume of individual tracks. The volume difference within each track doesn't tend to be too much, so adding compression on those will probably just lessen the dynamic feeling of my tracks.
For recordings, I can imagine compression is nearly a must have.
Yes Cubase is an actual D.A.W and a good one. If you havnt tried Reaper I would recommend trying that out, they both have similar style and feel.
Even if you do use vsti's which I have to. Sometimes I use a multiband compression on vsti's to compress the levels of the bass,mid and highs. Because I feel when I use multiband compression on vsti's it really brings out those instruments as a real instrument would. plus, its better to be safe then sorry.
Brandon Morris,
Steam Group: http://steamcommunity.com/groups/OpenGameArt
If you guys need any help on projects email me atBrandonmorris12@gmail.com. Pay is not mandatory and Im open.