In this tutorial I will try to highlight some techniques you can use to shape the sound of your mixes according to your preference.
Imagine that you have a killer production, one that fills up the frequency range in a very good way and that you know you can make really great sounding during mixing. Imagine that somebody goes to the mixing console and starts deleting some tracks from the production at random, so that there are only a few tracks left, let's say 5 tracks left, 3 of those are virtual instruments. Imagine that your task is now to make this production sound awesome at a commercial loudness level, simply put you have been assigned the task to make this production hit, but without a number of key qualities about the recording.
Actually, this is what it is like to try to make a B class recording sound like an A class mix, engineers face this situation every day. The solution to the problem is to send the recording back to recording. But let's say your boss will not, for whatever reason, allow that this time, now you are stuck in a very difficult position. It is at this point that your sound shaping abilities becomes especially important.
By default, what usually happens in a case like this is that when you gain stage the mix, you end up in this never ending cycle where one sound source is always too loud. No matter how you balance, one or many sound sources constantly turns out too loud. It is like having 4 keys pressed and as soon as you press the last button one of the four buttons becomes unpressed, so you can never have all 5 buttons pressed at the same time. This happens because these 5 sound sources are very challenging to combine in order to create a balanced mix frequency response. This is one of the reasons why the EQ and compressor effects are so important when mixing and mastering - you must find ways to color the tracks you work on in such a way that you end up with a balanced frequency response that works.
To do this you need to be aware about both the rms and the peak information you get from the elements in the mix. By default the lead vocals and the snare might compete a lot, which means that to make both loud enough in the mix, you might end up with a too loud mid range, in this case the upper mid range and highs can become especially problematic, you might have certain hits where both the snare and the lead vocals are very loud in their upper mids and highs at the same time, causing pretty nasty mix transients on those hits that are then snapped off too fast with brickwall peak limiters further down the chain. But you might not want to rely on side chaining in this case, because it de-stabilizes the dynamics on the other sound sources when the drums in turn are side chained to other elements. To hence deal with the issue in a different way you can create peak leveling towards this region, by applying an EQ effect before the compressor that boosts the frequencies, say around 6 kHz. This makes it so that the 6 kHz frequency range contributes more towards the signal exceeding the threshold, hence also acts to level out the peaks around that frequency. Once you then have the peaks in that frequency range numb enough, then you can apply for instance tape saturation or an EQ boost on the other side of the compressor to for instance recover a little of those peaks at more even levels and put the vocals a bit more up front relative to the snare, hence provide some low and mid frequency room to the snare. In order not to have to boost so much high frequencies on the vocals - which can force the need for de-essing - you can slightly reduce some high frequencies on the snare after the compressor, this adjusts the perception of the snare from the high frequencies to the low frequencies, it gets some additional body - the punch is still there because of the high frequency boost before the compressor.
Imagine that you have a killer production, one that fills up the frequency range in a very good way and that you know you can make really great sounding during mixing. Imagine that somebody goes to the mixing console and starts deleting some tracks from the production at random, so that there are only a few tracks left, let's say 5 tracks left, 3 of those are virtual instruments. Imagine that your task is now to make this production sound awesome at a commercial loudness level, simply put you have been assigned the task to make this production hit, but without a number of key qualities about the recording.
Actually, this is what it is like to try to make a B class recording sound like an A class mix, engineers face this situation every day. The solution to the problem is to send the recording back to recording. But let's say your boss will not, for whatever reason, allow that this time, now you are stuck in a very difficult position. It is at this point that your sound shaping abilities becomes especially important.
By default, what usually happens in a case like this is that when you gain stage the mix, you end up in this never ending cycle where one sound source is always too loud. No matter how you balance, one or many sound sources constantly turns out too loud. It is like having 4 keys pressed and as soon as you press the last button one of the four buttons becomes unpressed, so you can never have all 5 buttons pressed at the same time. This happens because these 5 sound sources are very challenging to combine in order to create a balanced mix frequency response. This is one of the reasons why the EQ and compressor effects are so important when mixing and mastering - you must find ways to color the tracks you work on in such a way that you end up with a balanced frequency response that works.
To do this you need to be aware about both the rms and the peak information you get from the elements in the mix. By default the lead vocals and the snare might compete a lot, which means that to make both loud enough in the mix, you might end up with a too loud mid range, in this case the upper mid range and highs can become especially problematic, you might have certain hits where both the snare and the lead vocals are very loud in their upper mids and highs at the same time, causing pretty nasty mix transients on those hits that are then snapped off too fast with brickwall peak limiters further down the chain. But you might not want to rely on side chaining in this case, because it de-stabilizes the dynamics on the other sound sources when the drums in turn are side chained to other elements. To hence deal with the issue in a different way you can create peak leveling towards this region, by applying an EQ effect before the compressor that boosts the frequencies, say around 6 kHz. This makes it so that the 6 kHz frequency range contributes more towards the signal exceeding the threshold, hence also acts to level out the peaks around that frequency. Once you then have the peaks in that frequency range numb enough, then you can apply for instance tape saturation or an EQ boost on the other side of the compressor to for instance recover a little of those peaks at more even levels and put the vocals a bit more up front relative to the snare, hence provide some low and mid frequency room to the snare. In order not to have to boost so much high frequencies on the vocals - which can force the need for de-essing - you can slightly reduce some high frequencies on the snare after the compressor, this adjusts the perception of the snare from the high frequencies to the low frequencies, it gets some additional body - the punch is still there because of the high frequency boost before the compressor.
Last edited:
