Karco's Comprehensive Guide To Making Better Music (Part 1 of 2)
2009-06-29 22:13:24 by KarcoReviewAlt
Did I say comprehensive? I meant excessively long!
New article! This is becoming a semiannual thing. :D This one's a follow-up to my first article on this account, though see how this article covers everything my previous two articles did and more - this is intended to be an all-purpose, independent article. All of this is original writing done just for this article, so I suggest you read it all even if you've read other articles of mine about some topics before!
Note that this article centers around the genre Trance, but many of the fundamental concepts can be applied to nearly any genre there is. Like always, if I've made a mistake here or got something wrong, or if there's a broken/incorrect link, please let me know! Likewise, if you found this article helpful, spread the word and/or leave a comment, it's nice to receive encouragement and I might just write more someday. ;D
Since this article's so long, a table of contents is below! If you want to skip to a section, use the find function of your browser and the code located to the right of each section or subsection. Also, don't pressure yourself to read the whole thing in one sitting - this is well over the length of my previous two articles combined. Stats at the end!
//PRODUCTION VALUE [PRO]
_ /Volume Levels [VOL]
_/Frequency Balance [FRE]
_/Master Channel Effects [MAS]
_/Thin Synths [THI]
_/Defined Sound [DEF]
_/Comparing with Professionals [COM]
_ /Sounds [SOU]
_/Chord Progressions and Melodies [CHO]
_/Your Own Character [YOU]
_/Recognizable Sounds [REC]
//IS YOUR TRACK SIGNABLE? [IS]
//REFERENCE TRACKS [REF]
It's more difficult to be a success in music than in most other options, simply because there are so many other people - many of them are just like you, the reader, with the same style, goals, and labels in mind, and there are only so many good labels to sign so many people, and only so many people with so much money and time to listen to your music.
Even if you do get signed, there's no guarantee you'll be successful. As a newcomer to the scene, unless you've been signed to a well-known label with a good reputation, there's more than a fair chance nobody will be excited about your track and want to buy it. It's not unheard of that a track by an obscure artist on an obscure label may hardly sell at all - and not relatively to the better-known tracks, but literally, only a few times, ever. You will not only need to find a good label willing to sell your music, but also promotion - often big names, with radio shows and DJ gigs that will promote your track by playing it to an audience. These people, just like the labels, have standards you will need to meet if you want them to play your track.
If going professional isn't a goal of yours, there's something for you here as well! This guide was originally titled, "Is your track signable?" but I decided to expand it to be much more than that, such that any kind of reader will have learned something by the end. This guide, then, is meant to help you improve as a producer, and distinguish yourself and your music from everyone else, no matter who you are.
Production value should be something everyone strives for. It's the polish that separates the professionals from the amateurs, or the best from the mediocre. Not only that, but all signed tracks are well-produced in the label's eyes - end of story. If they didn't think so, the track wouldn't be signed, or in the best-case scenario, they would help the artist rework the track into something better. For obvious reasons, high production value is a given with all signed tracks - if your track clips even before the master, you're deluding yourself if you think it could get signed, and no matter how much character your track has, or how well you manage your energy, a label will not readily sign a track with low production value.
As an amateur submitting among amateurs, production value is the first way to help yourself stand out from the masses. Experience plays a large role here; chances are as an artist of one, two, or more years of experience, you won't have the knowledge to make a successful track. Don't let that discourage you though - be patient and give yourself time to develop.
Setting volume levels is mixing in its most basic form, and the first step you will take towards your goal: a polished mix. In this stage of the track, you as the mixer will need to set your priorities as to what you want in the front of your track, and what can take a seat in the back. Consider the intended character of your track when choosing volume levels - is your bassline particularly characteristic? Your clap? Your FX sounds? Let your more generic sounds play more quietly to leave room for the more creative sounds of the track. Some sounds can achieve the same effect or better at a relatively quiet volume, and some sounds must be held back in order not to overpower the rest of the track. Noise FX is a good example for both of these cases - notice its subtle use at 4:14 in the Tritonal remix of Ferry Corsten - We Belong. Your judgment to make these calls can be aided by reference tracks, but in the end must come from experience.
If there is a conflict of interest between sounds - e.g. kick vs. bass, a sidechained bass with lots of mids and highs vs. hihats, or synths vs. vocals, always know that a mix is three-dimensional, that a sound is located at a point in three-dimensional "space." These dimensions are frequency, timing, and panning. Separation in one of these dimensions will solve problems between clashing sounds - for example, a kick and bass could play at different times, or on different frequencies at different volumes, without fighting each other. Be aware of this and take advantage of it to separate your sounds and fill up your mix!
To many novice and intermediate-level producers, placing a maximizer or compressor with some kind of "Maximum Loudness" preset on the Master Channel gives the track a desirable sound. The mix instantly becomes thick and loud, doesn't clip, and blends together. Similarly, an EQ might be used to compensate for a large lack in bass. These, and other dramatic uses of the Master Channel, are all bad practice.
A mix is the sum of its parts. Therefore, any problems in an imperfect mix originate from the individual sounds. Though it may be more work, getting your hands dirty in the individual sounds will, without fail, result in a cleaner mix than blaming everything as a whole will. If your only your hihats need more sparkle, why boost the treble on everything? If your mix is lacking in treble, why boost all your sounds by the same amount? Some may need more and some may need less - such is the nature of mixing. Master channel effects are never the answer.
Even the Master volume level should be left alone at 100% - if you think your track is too loud, chances are it's because only a few of your sounds are too loud. Don't bring the Master volume down, let alone compress the whole thing - instead, what sounds seem especially loud to you? Take those down a touch instead, and then reconsider your mix.
The use of the Master Channel is better known as mastering. Mastering is to be done at the end of production, with the finished mix - master early and, accidentally or not, you risk compensating or mixing under a false impression because of it during the mixing process, resulting in a potentially ruined mix. With compressors, limiters, or maximizers, this results in a "squashed" sound with very little dynamic range (the range between high and low volumes within a section). EQs and other effects will simply make your track extremely difficult to mix.
Frequency balance is a big part of holding high production standards, and is the trait behind the thick sound heard in many of your favorite professional tracks. Typically when putting a track together, frequency balance follows if the rest of the rules about mixing and synth choice are followed.
Telling when your frequencies are imbalanced depends on having good ears. The more experience you have, the more immediately you can recognize when your frequency balance needs work. Otherwise, checking frequency balance is a very good application of reference tracks - frequencywise, how does your mix compare to that of a professional's?
As a rule, never blame the Master channel for any imbalance of frequencies you find. Just as I previously noted that a mix is the sum of its parts, your problem lies within your individual sounds. Does it seem that you have too many playing in the same frequency range? Too few? Are any sounds too overpowering? Too quiet? And so on. It's very important to reduce volumes just as often, if not more, than you raise them, so be sure to do so - as a rule it's always better to go quiet than loud, unless you want your track to clip.
Often novice producers will fall for "smiley EQing," which might seem like a good idea to make your track sound better. Smiley EQing is placing an EQ on the Master Channel and boosting the lows and highs, and cutting the mids - the first problem with this is that it's a Master Channel effect, which I went over earlier, and the second problem is that by using smiley EQing, you lose all hope of having balanced frequencies - in fact, the use of smiley EQing will take you in the exact opposite direction, and you'll lose that coveted "thick" sound you were going for.
A word of warning: be especially careful with your frequency balance when listening on a new pair of monitors or headphones, as your ears may still be adjusted to your old pair and you may feel the need to compensate with dramatic amounts of EQ to what you're expecting to hear. After all: different sound systems reproduce different frequencies differently. Personally, I primarily use Dinka's remix of Jaytech - Pyramid whenever looking for a reference for mixing or for comparing monitoring systems, as Dinka's exceptionally clean mixing, defined sound, and frequency balance make for my ideal kind of reference. Doubtlessly, this track isn't the ideal reference for everyone - the best way for you to find yours is to listen to lots of music. The qualities I mentioned above are traits to watch out for when looking for a reference track.
Thin or "tinny" synths are quite a common problem, so I've decided to write about them here. These are often heard from people still inexperienced with, or experimenting with, the synths they're using. Thin synths can detract not only from a track's frequency balance but also from its character, and are many times caused by relatively raw waveforms or otherwise poor synths. For example, an artist might decide to stop using Nexus and start creating their presets from scratch - truth be told, that's a very big jump to make and synthesis is a vast world. Whatever your story, don't aim for anything out of your league and take small steps. Even if you want to end your reliance on Nexus forever someday, be patient and give yourself time to learn what it means to make a good synth.
You can tell your synths are sounding thin by comparing to reference tracks, or by listening to the frequency balance. How do things add up? Are there holes in your mix? Are your mids weak? Does your track lack the "thick" sound professional tracks have? These are all signs of thin synths. Bear in mind these don't always mean your synths are thin - use common sense. Many of these problems could also have something to do with your frequency balance, so don't be too quick to blame your synths.
Very frequently I come across tracks or demos that sound muddled and messy. The problem isn't necessarily a busy mix, though - what the artists behind them forget or don't know is to aim for a defined sound.
A "defined sound" is almost always a desirable trait until a messier or wetter sound (e.g. Jaytech - Pyramid (Original Mix))can be pulled off for the sake of character (more on that later). This takes a higher level of skill though, and even these mixes are defined to an extent - basically, always aim for a defined sound in some sense.
When sounds blend together too much, the sounds need to be distinguished from each other to regain the defined sound. Hihat sounds blending with the clap or each other, the kick and bass blending together in the bass regions, supporting synths blending in with pads, sounds that are just barely audible - these are all examples of sounds that haven't been defined from each other. Consider this blending as a different sort of clashing or sounds "fighting" with each other as a kick and bass would, and then rework your mix accordingly, with the three dimensions of time, frequency, and pan in mind. Alternatively, you may replace one sound with another if it still fits the track's character.
When comparing your mix to that of a professional's, be aware that nearly every professional track you're going to find has been mastered. This means your track is going to sound much quieter and weaker than theirs will - this is normal.
I personally discovered this when producing my Instrumental #4 remix, in the style of Daniel Kandi's remix of Out There (5th Dimension). Compared side by side, my track paled in comparison - the difference was the master.
Don't try to replicate professional loudness without a master. Wait until the mix is finished, and then use the master channel to master your track, or find someone to do the mastering for you.
Worst of all is to use radio shows or rips from them as a reference when mixing your tracks - in a forum post written by Tony of Above & Beyond, note where he says a track played in their radio show Trance Around The World is typically compressed eight times. Such a squashed example will only lead you to squashing your track yourself. The ideal reference is a store-bought WAV, though 320 kbps MP3 files are passable as well in my experience.
What makes your track special, or unique? Why would anyone want to choose to spend their time listening to your music over that of everyone else? Why would a label want to sign your track over another? How do people remember your track? Most of the time, the answer to these questions is character.
Character is the overall nature of your track, and is the product of all of the choices you make in production - tempo, genre, composition, synths, drums, FX, the list goes on. The other two main sections of this article, production value and energy, also determine the character of your track. Character can make your track a brick in the wall, forgotten - or the hit of the decade. Character is the reason why tracks like Darude - Sandstorm became famous. Jono Grant of Trance group Above & Beyond mentions not once (at 15:00), not twice(at 12:00), but three times (at 4:28),how character is important in a track during the International Music Summit 09.
Overall, creativity determines character.
It goes without saying that the sounds in a track are the biggest contributors to its character. This means two things: one, it's a good idea to find a character early on in production so you can pick sounds to fit, and two, choose interesting sounds. As an example, a distorted, bitcrushed bass synth will do much better both on the dancefloor and in listeners' ears than a sine wave sub bass (almost always a bad choice) would.
There are some cases where simple waveforms or other simple sounds work wonders - for example, Benny Benassi - Satisfaction uses a modified sawtooth wave. Many Trance tracks use sine waves, and tracks out of various genres of EDM sometimes use drums that sound like they came straight out of a drum machine. Note that the best of these examples, though, will modify the simple sounds to give them another dimension of character - returning to the example of Benny Benassi - Satisfaction, the sawtooth bass is sidechained and EQed to give it a different timbre than the typical sawtooth wave does.
Be aware that character is not only affected by sounds chosen, but also by sounds placed up front and sounds placed down back in the mix. A prominent hihat sound sticking out of the hihat line will definitely affect character, for better or for worse. A kick taking the backseat for the bass synth, or vice versa, will also alter the character. Take advantage of this to further build the character of your track towards what you intend it to be, while keeping production value high.
Of course, sounds of conflicting character simply won't do. It would be extremely difficult, for example, to justify extremely lush pads and a raw-sounding square wave bass in the same track. This is the problem behind the complaint, "it doesn't fit." Be sure the characters of your sounds match each other and the intended character of your track.
Perhaps counterintuitively to some, there is such a thing as good character and bad character. More logically is that from creativity comes good character, and from unoriginality, bad character. However, there is no "correct" good character - as long as your track is original, stands out, and is memorable, all is well!
This applies most significantly to composition. Notice how Daniel Kandi recycles many of the sounds from his track Liberate in his remix of Rapha - Pandora. Admittedly, many sounds are also changed, but the biggest characteristic difference between the two is in their composition. Liberate is positive and uplifting, while Pandora is darker and more emotional. Always consider composition when considering character, and if you can make it match with your sounds, the character will become even more prominent.
Composing is a vast world. For the sake of the article I'll assume that you, the reader, know basic theory, as it's important that you do in the first place, whether by habit or as knowledge. Without basic knowledge of theory, your tracks will come out to be dissonant nonsense, lacking any purpose and direction.
One should have two goals when writing a chord progression. The first is to write something with energy (which I will cover later) and direction, with a last chord that smoothly flows into the first. It takes some experience and intuition to judge these, or you could compare to a reference track - but be sure you don't end up accidentally sounding like your favorite artist instead of yourself! This goal is vital to your chord progression, as if you fail here then it becomes impossible to manage energy well. More on that later.
The second of these two goals is to avoid all generic progressions. I could hardly attempt to name them all here, but the big ones are i-VI-III-VII, VI-VII-i-i, and all of their variants. As for the rest, the best way to avoid them is to do three more things: 1) avoid plain major and minor chords, which are very flat and one-dimensional by nature, 2) avoid flat rhythms if it fits the character of your track, meaning you should scrap A-A-B-B-C-C-D-D for something like A-B-B-B-C-D-D-A, and 3) get creative! Don't settle for an "average" chord progression even if it doesn't strike you as generic, but instead aim for something you're truly proud of. That's the kind of progression that will go over well in listeners' ears, be they fans of your music or labels that could potentially sign your track.
Once a good chord progression is established, good composition will follow - your goals here are similar to those for chord progressions. Write something with energy and direction, and don't be generic! The more energy and direction your melody has, the catchier it's going to be, and the more people are going to remember it. This is why I think tracks like Gareth Emery - Exposure are much more likely to be successful - it has a creative chord progression and a catchy melody, not to mention the rest of its tremendous character. Once again, it takes experience to judge direction and energy, but reference tracks may also be used.
Generic rhythms can also happen. The rhythm pattern of four dotted quarter notes and two eighth notes repeating, or the pattern of two dotted quarter notes and an eighth note, occur in some form in well over half of the music I've ever heard. This alone isn't bad, but overuse of it is - trance can get especially notorious for the number of tracks using the first of those two rhythms with a long-release supersaw lead. After enough of them, it became unoriginal. Likewise, offbeat basses are almost always generic and offer nearly no character at all. Don't take the easy way out. Beware of these unoriginal rhythms as they, too, can detract from the character of your track if misused.
Some people attempt to emulate the sound of their favorite artists during their early years as a producer. That's fine for learning how to produce well, as a sort of pseudo-mentor, but to make it in the music industry you need a sound of your own that sets yourself apart from everyone else. As I mentioned before - character is the reason most people will listen to your music. A thread on Anjunabeats' forums, titled "Producer's Trademarks," is the perfect example.
What will it be for you? Your chord progressions? Bass synths? Breaks? Peaks? Drums? Fills? The possibilities are endless, and the only thing to beware of is that you don't accidentally start using someone else's sound, for whatever reason. Deadmau5 started with his own character, and countless others saw its potential and copied it - don't do the same and be your own artist.
Be patient in finding your own sound - you can hardly expect it to come right away just as you could hardly expect to start producing signable music right from the start. If you're still learning and working on improving as a producer, don't expect to find your sound anytime soon - it will change with time as you continue to learn.
It's very easy for an inexperienced producer new to VEC or Nexus to get excited by the quality of the sounds and start using them raw - good as these sounds are though, they, in fact 1) were made by someone else, and 2) are used all the time, just as raw, by many others. This leads to raw sounds from many banks (VEC, Nexus, FL, to name a few) sounding generic. Just like using generic chord progressions, using generic sounds in your track causes you not to sound like yourself, which is the goal, but instead like everyone else, or like whoever made the sounds - in the case of VEC or Nexus, Manuel Schleis. Without your own sound your character dissolves completely, and you lose all chances at succeeding anywhere with your music.
There are four things to be done about recognizable sounds. They are:
1. Layering. This is the addition of sounds to each other, the sum of which is used in their place. For example, two kick sounds, one for bass, and one for punch and high-end response, might be layered to achieve a superior sound. Similarly, your generic Vengeance kick can be transformed into a powerful Trance kick by combining it with other generic Vengeance kicks. Layering takes practice, but done right, can be applied to just about anything to just about any degree. It's not unheard of that a bass synth is made up of four different synths layered together. Often I layer a sawpad on top of another pad to give it more high-end sparkle. Different types of sounds can be combined: a closed hihat can be layered on top of a kick for high-end response. The applications of this technique are endless, and anything goes! Though be sure you know what goes into a good mix before you start layering, that way, you know what you want out of your sound before you begin.
2. Samplework. This mostly applies to drums and drum loops, which can be cut up, modified by envelopes, timestretched, pitchbent, and so on. Kicks can be gated for a different character, loops can be sliced and combined with each other. If two layered sounds are out-of-phase, simply change the pitch of one and the problem is solved. I've seen a case where Andrew Bayer, half of the Signalrunners, processed a simple three-word voice recording to make an entire track - drums, synths, and all, and it was quite convincing. Hear it here, it's called Time To Forget - read his blog for more information. Just like layering, anything goes.
3. FX. The complement to samplework, this applies mostly to synths, as only a few of your FX options are going to help any sort of drum sounds. That being said, options abound here: reverb, delay, chorus, distortion, compression, phasing, bitcrushing, stereo widening, EQ, and so on. I don't regret repeating myself in saying anything goes - get creative, learn how to use each of these tools, and experiment. One way or another, just about any synth can be matched to your intended character through FX.
Noticing any common themes? For the fourth time: anything goes when altering your sounds. Never forget that character is the goal and creativity your only limit.
4. Choosing the right sounds. This is very handy - by choosing the right sounds to begin with, you can sound less generic from the get-go, and have less to disguise. A rule of thumb I use when going through samples or presets is to always start from the last one and move up. This way, the first sounds you like will be the last ones almost anyone else would consider, covering your tracks. Though this may seem counterintuitive, also avoid sounds with lots of character in their raw state. Examples of this include FX hits and loops - their sound, naturally, will be much harder to change into something new.