Here are notes from Ronan Chris Murphy’s seminar “Production Tricks for Maximum Impact.”
These are quick and dirty, maybe I’ll expand and clean them up a bit later. But just wanted to get them out.
First, credit where credit’s due – Here are two links to learn more about him, and contact him if you want him to do some work:
OK, here goes. He did note that these things don’t apply to avant garde jazz or chill ambient music, and that there are no absolute rules in music. That said, read on.
The first FIVE seconds of a song are hugely important – that is, if you are trying to engage people in a mass way, or trying to get the attention of an A&R person or a music supervisor. Song intros that are just twice through the chord progression will not get your song heard by anyone.
Create interest in these first 5 seconds. Could be a cool riff, a weird sound, a great groove, an interesting synth, an engagine melody, a wall of guitars, highly distorted drums… think outside the box, and listen to hit songs to find examples.
We cannot engage in multiple things at once. Whether it’s two people talking to us at once or two melodic elements at the same time, we can’t engage in multiple things at once. When it sounds like a song is complicated with many elements, listen more carefully: only one or maybe two things are actually being FOCUSED ON at once. Example: OK Computer by Radiohead. It is thought of as complex, but it is really actually simple. Another example: Sgt. Pepper’s Lonely Hearts Club Band. Lots of instruments, but whenever one is added, another one disappears to allow it space.
If you change the sound stage, people think a song is more complex than it really it.
Throwing the kitchen sink at a song NEVER works. More is not more. More is less.
Use bold elements. Then take them away, and add others. And so on.
After a certain point, things usually start going downhill fast, the more elements you add. Example: if you have drums/bass/guitar/voice and then you add keys, well, yes you are adding keys but you have to remember you are also taking away from the drums, bass, guitar, and voice. New elements or instruments almost always compromise the existing ones both sonically and with respect to the listener’s ability to hear what’s going on.
(Multiple layers of a single element count as a single element, i.e 50 layers of background harmonies singing a single chord count as one element.)
The most hugely successful songs are SIMPLE.
Figure out what you want the listener to focus on, and create space around that. Remove competition. Feature one thing at a time.
Sustained sounds rob your mix of punch.
Sustain is the enemy of impact. Sustain and punch are OPPOSITES. If you want punch in your mix, look for things that are sustaining, and change them to be less sustaining.
This includes the piano sustain pedal, a snare that rings, an arena-size kick drum, guitar notes that sustain rather than mute, reverb and delay on vocals (or on anything), etc.
This is not to say that sustain is bad or should never be used. Phil Spector’s Wall of Sound was intentionally just voice and tambourine over a massive indistinct wall of sound (comprised of a ton of musicians in a single room, like three pianists, 5 bassists, etc, some people not even mic’d directly). If that’s what you’re going for, fine. But anything that rings for a long time after the transient results in a) less punch and b) muddier sonic information.
Drums are frequently some of the biggest offenders. A snare’s ringout is almost never in the key of the song (ever think of that? I didn’t). Toms vibrate sympathetically even when not hit. Ride & crash cymbals are common problems, and compression usually makes it worse because compression brings transient and sustain levels closer to each other, thus reducing the effect of the transient and increasing the effect of the sustain.
Big hit records (except for 1982-1990) have a lot less reverb and delay than we think. A lot of them lately have little to NO reverb. Pink and Katy Perry: completely dry vocals. Rihanna is an exception.
NEVER add reverb on the master channel. Unless you’re doing spacey ambient experimental music.
Exception on delay: Delay which is in time, i.e. delay as part of the groove (as in dance/EDM), is more a part of the performance, than a factor adding sustain.
Keep it simple.
Keeping things simple makes you look like a badass (as a mixer or producer).
One of a great mixer’s most powerful tools is the MUTE button!
If anything in your mix sounds awesome soloed, it will not sound awesome surrounded by a huge, dense mix.
Record in mono. The enemy of big wide stereo mixes, is stereo instruments.
The widest stereo result comes from recording everything in mono and then using those mono recordings to define the far edges of the mix. Individual stereo tracks cannot be pushed as far to the edge of the mix; they all live in a muddy middle of the mix. (And then there is no place to put the vocals, which should be in the middle.)
If given stereo tracks, see whether it’s possible to either collapse them to mono or actually use only half of the stereo field. The individual track will sound worse, but the mix will sound better.
There are some exceptions to this, such as a piece made only of live piano and vocals, where you may want a stereo piano sound which helps define the room space.
The louder your vocal has to be, the wimpier your track sounds.
We tend to calibrate volume to the vocal. Therefore, if the vocal has to be pushed way loud in the mix in order to be heard, the listener will turn the track down, because the vocal is too loud. This also turns everything else in the track down, and makes it sound wimpy. It is better to carve out space for the vocal (and keep it dry) so that it can be at a low volume relative to the track, and still be heard. This allows the guitars, drums, and everything else to seem much louder.
Often, in mastering, a stripped down ballad on a CD will end up sounding way louder than all the rock songs on an album, because of this (above).
When mixing your stuff and comparing to commercial releases, compare at the same volume.
It doesn’t help to compare your un-mastered mix to a mastered commercial release. It will always seem quieter, and you won’t be able to troubleshoot as easily. Compare your track at the same perceptual volume as the commercial track, and then listen for specific differences.
Bass, synth bass, anything with lots of low end should always be recorded mono and placed straight up the middle in the mix.
Cut low frequencies as high as possible, until it starts sounding bad, then back down again. (I think he said he cuts up to as high as 80Hz, until it really starts taking away from the track.)