EQing Trance Music

EQ is one of the most commonly used tools in mixing. This also makes it one of the most commonly misused tools, particularly when starting out. There is great temptation to apply massive EQ cuts or boosts to fix frequency imbalances, rather than making level adjustments, or even finding more suitable sounds in the first place. The vast majority of a great sounding mix comes from sounds which fit together well, set at appropriate levels relative to each other. if the mix isn't sounding decent before EQ and other tools have been applied, then the work of mixing is going to be much more difficult.

So we have a mix which sounds decent - how can EQ help? I picked up a very handy summary from Bobby Owsinski's "Mixing Engineer's Handbook" - cut to make things sound better, boost to make things sound different. This is how I apply it as well. Cuts clean out the parts of the sound we don't want, while leaving the overall character of the remaining parts of the sound relatively intact. Boosts introduce coloration which can spice up or change sounds quite a bit.

The first EQ I apply in a typical trance song is highpass filtering. Low frequencies take up a lot of headroom, due to the prominent kick and bass. So I remove the low frequencies from all the other sounds. This clears out the low end, leaving plenty of room for the kick and bass. Where I set the highpass cutoff frequency depends on the sounds in the song (including both the sounds being highpassed as well as the kick and bass). The filters are applied either directly to individual sounds, or to groupings of sounds. While the full mix is playing, I find the point where the filter begins to cut away part of the sound I want to retain, and move it back a bit from there. Sometimes I'll use a low shelf EQ instead, to use a much gentler slope - my goal is simply to significantly reduce the low end content which isn't essential to the sound within the context of the mix. When I started highpass filtering sounds I made some sounds too thin by setting the cutoff frequencies too high, in a misguided attempt to introduce separation and clarity into my mixes. As usual, experience helps to train our ears and get better at finding the sweet spot.

With the highpass filtering done, I turn to problematic or resonant frequencies, finding and taming them. Certain sounds (especially basses and pads, though sometimes kick drums) frequently have a lot of content around a particular frequency, in a way which isn't pleasing to the ear. To me, it makes the mix sound like it's under pressure. It can quickly cause muddiness and will mean the apparent loudness of the song is decreased, since the overpowering frequency will be hitting the ceiling of our levels. To locate these problem frequencies, a frequency analyzer can be used on the particular sound to see where there is excessive content. Another way is to set an EQ to maximum boost, with a very narrow Q, and sweep around the problem area to find the frequency where the resonance is greatest (be careful with your ears, don't do this at high levels). With problem areas, my first instinct is to try pulling the fader of the offending instrument down a bit and see whether that fixes the problem. If not, I notch out the problem frequency on that instrument. The depth and width of the notch are dependent on the sound. Sometimes it will only take around 4 dBs of cut with a very narrow Q to fix the problem. Sticking with the "under pressure" viewpoint, this feels like taking a small pin and deflating the pressure in the mix. Suddenly, the mix is cleaner, clearer, more open, and the offending sound's character hasn't changed at all. It's a great feeling.

At other times it's not so much a resonant frequency as a conflicting region - say, a pad which is overlapping with the bass sounds at 350 Hz, muddying things up. In this case I'll apply a much wider cut. I make it as deep as it needs to be - just to the point where the conflict is resolved. if the pad (or whichever sound) disappears, I've either chosen the wrong sound or I should move it up an octave. Conflicting regions extend all the way up the frequency spectrum - I'll sometimes cut back a bass sound's high end if it's getting in the way of the lead sounds (even if it makes the bass sound less defined when solo'd - it's the mix that counts, not the individual sound). I'll sometimes pull back the high end of a particularly hissy hihat for similar reasons.

Then I apply a few boosts. I generally don't boost the low end, there's rarely a need. Only when I feel there's a particularly flat spot do I add a bump of a dB or so, with a medium to wide Q, either to a particular sound or to the kick and bass grouping. I use boosts much more to bring out definition - 1 or 2 dBs added to the kick, centered between 3 to 8 kHz. Mid/high bass sounds often benefit from a 2 to 4 dB boost centered around 1 kHz. A large, wide boost of 2.5 to 5 dBs to lead sounds, centered around 3 to 6 kHz really helps to bring these sounds through clearly. Drum sounds are sometimes treated to a boost around 1.5 to 2.5 kHz. Sometimes a high shelf boost is also applied to the drums to bring out high end above 8 kHz. However, often with appropriate levels and highpass filtering applied the high end will already be where it should be. All these frequencies and amounts are ballpark figures. Actual usage in a particular song may be quite different.

My send channels (delays and reverbs) get EQ'd, sometimes quite extremely. When I was starting out I didn't think to EQ effects, not realizing they could sound a lot better if I did. Highpass filters are applied to most send effects (except maybe the "small room" reverb), since low end content can quickly build up and it can sound odd to have very low frequencies spread out to the sides of the stereo image. This may be a holdover from the days of vinyl records, where the bass frequencies needed to be centered. The high end is also often pulled down a bit, if I want to blend the effect into the mix. Conversely, if I want an effect to stick out, I lift the high end up. In between the extremes, the same principles apply as with regular EQing - I pull down certain frequencies to make things sound better (sometimes 300 to 500 Hz can get a bit boxy/ boomy on my large reverb, so I'll pull that down for example). I lift certain frequency ranges to give effects a different character and make them stand out, especially on my delays, which I don't feel need to sound "natural". If I have any flat spots in a mix, I like to try boosting one of my send effects in that region rather than altering the sound of a particular instrument. This can work well, filling in the flat spot without changing the sound of any of the instruments.

Finally, there's the master channel EQ. Very rarely this won't require any EQ, which is satisfying in a way, but I don't lose any sleep when a few bands of EQ are required to make a song sound better overall. Sometimes I'll apply a steep highpass filter to bring down excessive content under 40 Hz. Often there will be a broad boost of a dB or two, at frequencies where all the sounds seem to benefit from the treatment. I used to go quite wild with my master EQ, using large cuts and boosts to try to squeeze my mixes into the shape of my reference songs! Nowadays, as noted above, things sound much better through choosing sounds which actually want to fit together and setting appropriate levels, meaning I EQ a lot less.

I hope this article has been helpful, keep making great music!

Fabian

Listening Techniques

Listening is the most important skill we use when working on music. Being able to hear what's happening to all of the instruments and sounds in our mixes, as well as being able to hear the mix in it's entirety, is essential in order to create songs which sound excellent. Good technology will help (high quality speakers, headphones, a high quality sound card), but the most important element is experience – with practice, your ears will develop and you'll be able to make many more fine distinctions than you were able to when you started out. Through experience, you'll sense when a bass or pad is boomy, when a hihat needs to have it's high end pulled back slightly with a shelving EQ, when the kick drum needs to be pulled down just a fraction to sit better in the mix.

There are good listening programs around. I've seen a few book/CD combinations, which take you through a variety of exercises, training you to recognize certain frequency bands, types of distortion and so on. These can be very helpful for some people. I went through one and perhaps I got something out of it – it's hard to separate all the different lessons I've learned through the years and quantify how much each contributed to my understanding. However, my feeling is that the many years of actually mixing audio is by far the biggest factor in my improved sound. I've mixed an enormous number of kick drums in my time, in a wide variety of contexts, and have made an enormous number of mistakes. By now I know a lot about what not to do! These days, experience saves me a lot of time, because I don't stumble down as many dark alleys as I used to.

Having established that the best way to get better at listening is through a lot of experience, how do I make this article at all useful to newcomers to mixing? Well, there are a number of ways you can catch imbalances or bad sound combinations in your songs via creative listening techniques.

A really good technique is listening very quietly, in mono. I use a stereo imager plugin on the master channel for this. It allows me to set a stereo width of zero, so the signal is straight up the centre, and also has a level control, which I turn way down. Most of the time this plugin is bypassed, but it's very easy to do a quick mix check by turning it on. This combination is helpful for two reasons – in well mixed songs, the sounds come through clearly even at very low levels. When you're working at high levels, it's easy to think the bass is prominent enough. But often the bass can disappear entirely at low levels, since it doesn't have enough midrange content. The same is true for other instruments – a mix that is well balanced at low levels will sound even better when you turn the volume up. So turn it right down, listen and take notes – do any instruments get lost in the mix? Then listen out for instruments which seem to be sticking out more than they should – maybe you'll notice a bright hihat that's cutting through rather harshly, that will still come through fine if you drop it a couple of dBs. The other reason the quiet mono combination is helpful is that by listening in mono you'll firstly be able to tell whether any instruments disappear or drop considerably in volume due to phase cancellation (because inappropriate stereo treatments have been applied to the instruments) and secondly you'll get a better feel for the relative volumes, without the distraction of having some sounds out at the sides, some in the centre, some narrow, some wide. A lot of venues play music in mono, so it's good to have some idea how your song will play under these conditions (unless you're adamant your music will never be played in venues).

Another technique which is along the lines of the “listening quietly” technique is to use high and low pass filters to cut out low and high frequencies, leaving just the midrange. The midrange is the key to a great sounding mix. Some systems won't reproduce a range as wide as you're working with. The reason this technique is similar to listening quietly is that both techniques bring the midrange to greater prominence. Human hearing isn't linear, certain frequencies are perceived as louder than others when played at the same volume (the human ear can hear very well at the frequencies associated with human speech). At low levels, the low and high end of the frequency spectrum drop off considerably. The louder the volume is pushed up, the more linear our response to sounds becomes.

A technique I've found useful is the “morning after” listening test. I usually combine this with listening quietly (though not necessarily in mono). This just involves playing the song after not having heard it for a good chunk of time, and noting which instruments are the most noticeable. I've often pulled a few faders back a touch the next morning after this listening test, or removed some frequency content from a sound which was sticking out where it shouldn't.

A technique I used to use more often a while back is listening to a song in a variety of spaces – in the car, on the hi-fi, at friend's houses, at venues when I got the chance. Each time I made mental notes of things that sounded odd and tackled these when I got back to my studio. In time, though, I got to know my listening environment in my studio and these days I rarely get surprised by anything when I hear one of my songs in an unfamiliar environment. For me, the “listening in a variety of places” technique served to teach me more about the sound of my studio – once I could hear the imbalance in my studio, I could hear it the next time it occurred, and fix it before I played the track elsewhere. Techniques such as stepping out of the studio and listening from various places outside the room are similar to this – again, I find myself doing this less and less because I don't seem to get surprised any more.

Having said that, I do check my songs both on my headphones and speakers. I do the vast majority of my mixing through headphones at low levels, but I do send the sound out through my speakers, primarily to check out the low end. Through headphones, it may sound like my low end is very decent, but sending it out to the speakers can sometimes disappoint! In trance, the bass has to hit me in the chest in a certain way. There's no way to hit my chest through headphones. I have a certain spot in the corner of my room where I stand and feel the chest-thumpiness of my song. I'll play a variety of reference songs interspersed with my song so I can tell if it's hitting my chest in a satisfying way.

Speaking of reference songs, these are absolutely vital. If you don't yet have a collection of great sounding songs in the style you're producing/ mixing, I can't recommend enough that you build one up. I have a whole bunch of short snippets from the middles of great sounding trance songs, so I can quickly flip from one to the next, then to my song, and compare like with like. This will quickly let you know if there are any areas that are lacking or overly prominent.

Well, there are some brief thoughts about listening. You'll be amazed at how much more you're able to hear as you build mixing experience. You'll be able to tell much more quickly if a kick drum/ bass combination just isn't working, if a lead is too washed out, if percussive elements have particularly shrill high end frequencies. At this point I'll briefly mention frequency analyzers – as with some of the other techniques detailed above, analyzers can help you to train your ears to hear problem areas in your mix. By comparing a frequency chart of your song to frequency charts of reference songs you may see that your song has a big dip at 300 Hz compared to all the other songs. You may correct this area in your next few songs, based on the frequency analyzer, until you eventually start hearing and fixing it before you need to reach for the analyzer. Analyzers can be dangerous if used incorrectly (such as indiscriminately pushing up 300 Hz with an equalizer rather than looking first to correct the level of the bass), but as stated, can also be great learning aids.

Keep making great music!

Fabian

Giving and Getting Good Feedback

To varying degrees, many of us are holed up in our studios for significant amounts of time. We can achieve a lot by ourselves – developing our mixing and listening skills, maybe playing a few instruments, creating great sounding music.

At times, though, we're too close to our music and can lose objectivity. After we've set the room reverb for our drums and have listened to the mix a number of times while we're working on other areas, the reverb can simply become part of the way the song sounds – it's easy to not reflect on it further if it's not really pushing for attention. This is where a fresh set of ears can come in handy – another person may comment that the reverb is too bright, may provide a perspective we hadn't considered. Of course, they may not comment on the drum reverb, which could well mean it sounds fine (depending on who the feedback is coming from).

As with many things, clarity is important. Which aspects of your song(s) do you want input on? Who is going to provide this input? The ideal person to provide feedback on your mix is:

A very experienced producer/ mix engineer

Someone who is very familiar with your specific sub-genre

Someone who cares more about great sounding music than tending your ego

Someone who, in addition to criticizing, will suggest specific techniques to improve the mix

Friends and family, even if they're “easily accessible”, are often not the best sources for self-improvement, since many times they don't fulfill many of the above criteria. There's not much you can do with comments like “Wow, it sounds like real music” or “that's a cool guitar sound!”

Music Forums can be a good source of feedback, though there are drawbacks. Sometimes nobody may bother to provide feedback. Sometimes the only feedback you'll get is of the “friends and family” variety. Some forums are too broad – it's rare to find a forum dedicated solely to a single sub-genre. This means that you may be aiming for a particular type of, say, Death Metal, but get feedback from people who are into a different type of Death Metal. This can make it hard to work out whether their feedback is relevant to your song, or whether it reflects their personal biases. It would be great if every “feedbacker” provided examples of their own work so this could be taken into account. This would certainly help to find out which people are only just beginning to create and mix their own music, who have a lot of development ahead of them before they begin to make halfway listenable music. Frequently these people have no qualms about providing a lot of feedback about many aspects of your mix, as well as suggestions as to how things could be improved. Taken at face value, this feedback can be downright dangerous and actually set us back in our progress. In the end, it's up to each of us to decide what to do with the feedback we receive. If a consistent pattern of feedback emerges, it's more likely there's an issue there (though not necessarily, some people will only comment on something when someone else has already mentioned it). If only one person singles out an aspect, potentially it's not so serious (though again, not necessarily – maybe this one person just took more time to listen to your song in detail).

As important as the person providing us with feedback is the way we ask for this information. Sometimes it may be fine to say “Here's my new song – any comments?” People may reply along the lines of “Cool Hendrix wah on that guitar, dude” or “wtf is up with those lyrics, lolz”. Those comments may well be helpful (and may well stop us from writing lyrics for a while if we were already a bit insecure about them) but if we actually wanted to find out whether our bass sounds okay, since we're a bit unsure about it, then we should be more specific. We could either ask directly about the bass, or (if we feel that shining a spotlight on it before people actually hear the song may lead to skewed opinions) ask if there are any instruments which don't fit quite right and see if the bass gets mentioned. Maybe we'll find out our bass is fine, and it's just our dreadful kick drum which is making it sound like that.

In terms of getting specific, there are a few main areas we could be seeking feedback on:

Engineering (how the mix sounds – relative instrument levels, panning, EQ)

Arrangement (whether the song is too cluttered, or too bare)

Structure (the flow of the song, how it transitions from one section to the next)

Production (Instrument selection/ sound design, automation, special effects)

Composition (Melodies, chord progressions)

Playing technique (where relevant)

Maybe we want feedback on all of these, in which case a broad request for feedback may well suffice. But it pays to think about which aspects of our song we're most unsure about and direct our feedbacker's limited attention there.

Hopefully the above information is also helpful when it comes to providing feedback. All the same concepts apply, just from the other side of the equation. We should be aware of our own biases, the kinds of sounds we prefer. We should refrain from providing extensive feedback in sub-genres with which we're unfamiliar. When we are unfamiliar, we should admit this, to give our feedback some perspective. The same goes for when we have limited experience in mixing a particular genre, or we're aware that our own music has serious limitations. When we criticize a song, we should be able to offer suggestions as to how the song could be improved. This is much more helpful than simply criticizing. It should rarely happen that we say something like “those background vocals just sound odd, but I can't really think how to improve them, or what they could be replaced with”.

I hope that in time a number of skilled producers and engineers offer feedback services on www.InsideMixes.com. This should alleviate some of the drawbacks of anonymous, free forum feedback. We will be able to specify exactly who we want our feedback from (after listening to examples of their work), within what time frame, what they should focus on, and to what level of detail. In short, ideal feedback!

Keep making great music!

Fabian

Striving for Perfection

Those of us who are suitably obsessive aim high. We try to create songs where both the music and the mix will knock our listener's socks off. Songs where we've held nothing back, where every decision we made improved the song and led to perfection.

As long as we have a good relationship with “perfection”, this mindset is very desirable and helpful. We want our listeners (as well as ourselves) to enjoy our music as much as possible.

However, when our relationship with perfection causes us to never share our music, to feel bad when we fail to meet our high expectations, it stops being useful and can ultimately lead to us withdrawing from creating music.

The thing is, perfection can't be measured. We can listen to a song and appreciate how beautiful the music is, and recognize how great the production is. But there's nothing in the song which we can point to and say “that's perfect”. Or, to look at it another way, there are many things we can point to as examples of perfection, but none of them couldn't be replaced by something else which is just as, or even more, “perfect”.

Perfection is a direction, not an endpoint. When I create music, I do it to the best of my ability, then I let it go. Could I come back a year later, with more experience under my belt, and adjust the reverb on my lead, or nudge the kick drum up slightly and end up with something which sounds better? Absolutely. I doubt there's a song in the world which can't be improved upon in some way. We can listen to a song with a great mix of sounds and see it as something set in stone – which it is, by the time it reaches us as listeners. However, the stone could have been set quite differently and we'd be none the wiser. Perfection has many forms (or none!)

What's true for the song is true for individual instruments or sounds. When I started mixing I obsessed about finding “perfect sounds” - I wanted to have the “perfect kick”, the “perfect bass”, and so on. Maybe I thought that once I'd found all these sounds, I'd simply slot them together and have “the perfect song”. I never ended up finding them all, and in any case, this line of thinking is nonsense. Finding “the best” sounds in isolation practically guarantees an impressive mess when we try to somehow mash them together into a song.

Sounds in a mix only have meaning in relation to each other. An instrument is loud because other instruments are quiet. A sound is hard because other sounds are soft. An arbitrarily unexciting kick drum can sound fantastic in the right context. A snare which stands out can sound great, as can a snare which blends in. Basses with clear, defined high end can work well, as can basses which occupy only the bottom end. I've sometimes been surprised to discover that two very different sounding kicks in two of my songs are in fact the same sample. Because of the other sounds, one comes across in a more defined, hard-hitting way, the other in a smoother way which blends into the mix more. Sounds only have meaning in relation to each other.

Perfection is a great motivating force. It feels exhilarating to put in the effort and end up with a great sounding song. As long as we allow ourselves to do our best, as long as we don't second guess our decisions and always hold our options open because we feel “the perfect lead sound is just around the corner”, striving for perfection will keep us improving every day.

Keep making great music!

Fabian