Mixing your beats to perfection
Today we feature Andrew Caminiti, a skilled audio engineering enthusiast. He and I are online friends, and I have always valued sending personal beat snippets over his way for mixing critique. Join us in our discussion and stream the beats below as you read!
Thanks for having me!
Well, you can't polish a turd. One of the most important factors to a clean mix is your source material. This means that developing good sound selection judgement is pivotal. It will often take time to train your ears for this, but utilizing tools like Splice can be useful. Try organizing by popularity to hear what others are considering ideal. This is not full-proof, though. Eventually you will learn to use audio that is both high quality and tailored to "your" sound. Beyond source material, things like over-use of delay and reverb and phasing issues come to mind.
My mixes are often way more dry than you would initially think. If I pause my track in-DAW there is hardly any tail. Turn down the feedback and decay on your delays and reverbs, and perhaps the release time on your drum samples. If you want a thick reverb sound, re-sample it to audio and cut it when other layers need to come forward in the mix.
What are phasing issues? Phase cancellation occurs when the waveform directed out of your left speaker is nearly the mirror image of the waveform directed out of your right speaker. Think of the peaks and valleys in a sine wave. If you have a standard sine wave coming out of the left speaker and the same standard sine wave, but flipped, coming out of your right speaker, the resulting waveform when added together is a flat line. A good way to detect these issues if you're having trouble hearing them is to set your master channel to mono. If a layer is losing energy or volume when doing so, it is likely the result of phase cancellation. You can also use a correlation meter to visualize phasing issues. If the meter dips below zero, the material is likely problematic. Between zero and one is your target.
Phasing issues can be tricky. This potentially comes back to good sound selection judgement. I try to avoid phasing material from the get-go, but sometimes the instrument I need has phasing issues that just have to be grappled with. Tools like Waves InPhase, iZotope Ozone's Imager and Voxengo MSED can do the trick. Often in that order. I start by trying to correct the phase cancellation with Wave InPhase, but this doesn't always work, or it doesn't fix the issue enough to feel like the job is complete. Often I'll slap the multi-band Imager from Ozone afterwards and drop the bottom band to mono. Drop the additional bands from left to right if need be. If you're still experience phase cancellation, the Voxengo MSED is the final icing on the cake. Decrease the side volume and increase the center volume until the correlation meter is in the target range.
Q. So Andrew, thanks for joining us! I've known you for roughly 7ish years, and we met through Facebook's online music community. I know that you've spent a lot of time perfecting the smallest details in your mixing. So let's start by you telling us what the most common mixing issues that you tend to notice among newer producers are; and what you would generally say to the upcoming producer community in regards to sound engineering/mixing.
Fantastic answers, man.
I agree with what you said about reverb and delay having too much tail. In my earlier production, I used to slam reverb on everything, thinking it made it sound "better," but all it did was muddy up the mix and made the other instruments undefined and hidden. I eventually discovered that less truly is more, and as you said, it's really all about control. If you have these effects sounding all at the same time, they'll surely create mud and a lack of clarity in your mixes. Knowing when to add to a song vs. when to subtract from it is highly important, and effects are indeed a huge part of this. Too many effects that don't serve an exact purpose can ruin a beat.
Q. Can you share with us some of your mindset behind sound engineering that helped you better understand HOW and WHEN to use things like EQ or compression properly? Why does this all matter to someone who produces lofi / Chillhop stuff?
Equalization can be tricky without a trained ear.
More than anything, it's about time and effort that rewires the auditory cortex in your brain, but there are a few pivotal methods I've learned that I think anyone can take advantage of at any point in this journey.
The first is "frequency sweeping." To do this, apply an EQ to the layer in question and create a point with a high Q-value. Gain it by 12-15db, then adjust the frequency up and down the spectrum to find troublesome (or particularly annoying) frequency points. Once found, dip the band back down to -5db or so (use your ears) and adjust the Q-value subtly. Then repeat. This is was I would call a surgical approach, and is one of a multitude of methods I tend to at least try on much of my project layers. However, I don't always feel the layer needs this done. Don't over-do it.
A second method is probably something you've heard you shouldn't do. That is to work with the limiter/maximizer on the master channel activated. I'm not suggesting you do this for your entire project session, but for EQ specifically, this is actually helpful to do. The reason is that the limiter will exacerbate the EQ issues and make the changes more noticeable when fixed. When approaching EQ in this way, I suggest less surgical Q-values and more dipping than boosting.
Because you asked about compression as well, I will recommend a third EQ method here that I find to be one of the most groundbreaking discoveries I made. Work with dynamic EQs. Dynamic EQ's are EQ and compressor combination plugins. This allows you to compress or expand specific frequency ranges without touching others. Think multi-band compression but more precise. My favorite VST for this is Ozone's Dynamic EQ, but a great alternative is the Gliss EQ from Voxengo.
Concerning compression: To be honest, I avoid compression as much as possible. I don't add compression for the sake of it. An area where I find compression extremely useful, though, is side-chaining drums to the instruments in the mix. This creates a ducking effect on the instruments whenever a kick or snare is triggered. There are many ways to do this, so google "How to Side-chain in..." and complete it with whatever DAW you're working in.
I would also use compression for scenarios where high dynamic range is too noticeable. This means that there is too large a gap between the softest and loudest point of an instrument. This often applies to things like vocals, drums, guitars, and bass. But again, do not add compression for the sake of adding it. Use it when you notice an issue in dynamic range. Some compressors are "colored." This means the compressor is actually adding to the sound rather than simply compressing it. I love to use these types of compressors on drums and very, very subtly, in my master chain. The Glue compressor is an example of this. This is modeled after the digital-analog SLL series. I actually use the SLL G-Series model from UAD. Which The Glue compressor is also modeling.
Great stuff. I feel like everyone is guilty of adding compression even though they aren't quite sure why they're doing it. Dynamic EQ is also a game-changer; I believe you were the one that taught me about that, and since then, I've been using it to fix high pitch leads/synths/ keys. Those tend to have high resonance, especially when you couple them with reverb or delay, which sounds extremely ugly. Having dynamic EQ in situations like these can make all the difference and help keep these problem frequencies in control.
One more question;
Another common thing I hear newer producers are very flat and lifeless sounding drums. They sound "Stock" and don't punch through the mix and sound as impactful/fitting as they should. Regarding drumlines and percussions, what are some tips/processes you would recommend for full-sounding drumlines? (Aside from sound selection)
Side-chain. More specifically, mid/side side-chain. If you send your kick and snare signals to a tool like FabFilter Pro-C (SC) you can compress the mid and side signals separately to give you the illusion of wider drums. You do this by compressing the side information more than the mid.
Something else to try is distortion/compression tools like the Fatso or Distressor from UAD or Slate Digital. The SSL Buss compressor can also do good work here. The point is to "glue" the different drum samples together. The last thing to mention is a multi-band expansion or "upwards compression" like the OTT from Xfer Records. If you're using Ableton Live, this module is included in the software under the "multiband dynamics" presets. This tool will increase the low volume portions of your sound without increasing the peak volumes.
I will also definitely say that multi-band expansion is one of my favorite things to work with. It can transform an entire atmosphere with just a few mouse clicks. Ozone's plugin has "exciters" that allows you to give an expansion to different frequency bands and can really brighten and tighten up the mix while increasing the overall presence and liveliness. Often when you feel like your beat is "dull" sounding, what it often needs is just some beefing up via multi-band expansion/compression/excitement.
I hope that this has been helpful to our readers! Audio engineering can be a daunting subject at first, but after years of just making music, you will surely pick it up naturally. Don't be afraid to tweak new knobs and press new buttons, some of the best musical creations have always been accidents!
You can find Andrew at the links below;