Most producers after a while come to realise the necessity of frequency stacking certain elements to achieve a higher standard of production. In short, this involves splitting a signal (usually a bass) into low, mid and high components and then processing each band differently. The benefit of this is that you have much greater control and you get a wall of sound feel to your production. So in this article i'm going to outline how to do this using EQ and various types of other processing such as compression and stereo widening. I wanted to write another article next week on how to do this using Phase cancellation, but it turns out that the delay compensation in Ableton isn't good enough to allow for complete phase cancellation after inserting an EQ to seperate out a band of frequencies. So we'll have to stick with using EQ I guess, as follows.
Simple Frequency Stacking using EQ Only:
1. Compose your bassline melody using your synth or other weapon of choice.
2. Create three new channels, one labelled low, one mid and one high.
3. Group these channels so they are all outputting to the same destination.
4. Rout the inputs of the three channels to that the signal is coming from your bass melody into them.
5. Make sure that the channel with the melody itself is no longer outputting to the master.
6. Now on each channel in your frequency stacking buss, create an instance of an EQ.
***Disclaimer*** the following steps require you to use your ears and judge what filter settings will work the best with whatever source material you are using. I am putting in frequencies as a suggestion or guide only.
7. On the low channel, create a low pass filter and set it to around 200Hz.
8. On the mid channel, create a bandpass filter with a low cut around 150Hz to 200Hz, and a high cut somewhere around 1000Hz. You'll need some overlap to prevent pockets of frequencies being cut out in between bands.
9. On the high channel, create a high pass filter with a cut around 500Hz to 1000Hz. Again, you'll need to create some overlap for the reason mentioned above.
Now you have a series of overlapping bands dividing the same signal into three regions.
10. The next step is to add processing. I would start by mono'ing your low band. Grab a stereo widening tool and reduce the width to 0% which will keep your sub bass nice and tight, and avoid any phase problems.
11. Add compression to each band and drive the compression as much as possible. Then do the same for the bus as a whole. Adjust the relative levels of each band if you need to. Remember the human ear perceives higher frequency sounds as louder, so you may want to reduce the level of the mid and high bands in relation to the low band as you are EQ'ing / compressing.
12. Add any other effects that you think you might want to include. Usually I go for a touch of short reverb on the mid and high bands, as well as some chorus or stereo widening and possibly some distortion to add bite. I leave the low band relatively untouched. As a rule of thumb, the higher in the frequency spectrum you go, the wider the stereo field can be, within reason. Think of it as an inverted triangle.
- You may want to use a sine wave instead of your low end. I find I can get good results both ways. If you are using a sine wave you shouldn't need to compress the low band almost at all because the function of compression is to reduce the dynamic range (difference between the soft and loud parts) of a signal, however a sine wave is a fairly constant tone and doesn't have much change in amplitude over time.
- Some people seem to like to split their signal up into four bands rather than three, these being low, mid, high, and treble. I find this technique gets incredibly fiddly and difficult to control if I have four bands. though it may be something that you want to try.
- There is a plugin called CrossOver which uses liner phase EQ to plit a signal up into four bands, and then output these bands to different channels. This plugin is free and if it is of interest to you, grab it here.