Those little tricks in music production

Hi!

Just thought to make a thread with collections of those little tricks, that go a long way when you know them and empower you to produce better music, or to produce it faster. You know, we all know a trick or two.

I thought I make a start with some things against tired ears when it comes to bass, I just found enormously useful when trying to learn mixing.

  • If you have music with deep dull bass notes, and find it difficult to judge them perfectly in key or not, temporarily transpose your bassline up one or two octaves. Much easier to hear now if it is in key or not though it will sound like mickey mouse crap! When tuned to taste, transpose back down again.

  • when working on bass and especially subbass freqs, I have a butter n4 filter with cutoff at around 150-300 hz on the last pos of master channel. When my ears feel like loosing focus on the bass and everything becoming muddy, I toggle the highpass on, listen a short while to the thinner sounds, then turn it off and am surprised how well I can hear the subbass freqs now! Also to toggle it to lowpass allows one to hear the (sub)bass in isolation.

  • likewise you can isolate or toggle on/off the highs to reset the ears. or bandpass/notch the mids.

  • the same filter trick with filter on master also works for judging how you music will sound on systems with little or no bass. for example a butter n8 filter at 200-240 hz will make your tune sound a bit like on a smartphone, butter n4 at 100-200 Hz is like a small stereo with weak bass. Biquad hipass with some reso tuned to 60-110 hz can give impression like a hyped boombox with lack of subbass but boosted/resonating bass. Just take account the reso will boost some, so you will want to lower gain before it hits the speakers. Non-reso filters might also overshoot, but just a little.

2 Likes

nice idea for a thread man :slight_smile:

if anyones interested my poorly explained videos contain presets and little tricks i use for my sound design

  1. No thread like this is complete without mentioning the ‘mix in mono’ trick. Occasionally switching to mono is not only good for checking mono compatibility (if you care about that), but also for better revealing frequency masking and checking frequency balance.

I think this touches the same principle as your filter trick. It’s all about providing an alternative “lens” to view the mix thru. Sometimes I even have fun mixing thru a hard bitcrusher (5 or 6-bit) temporarily. Once you get a feel for it, it provides an alternative lens that can guide your mixing in sometimes useful ways :slight_smile:

  1. We naturally tend to do a lot of stuff backwards in a DAW/tracker. I think this is because so many details and possibilities are thrown in our face. Because of that, I think it’s a good idea to consciously turn the process around 180 degrees, if you haven’t already.

Specifically:

A ) Use a top down approach when composing a song. Start with the song structure. Then move on to optimizing the emotional ‘energy levels’ by laying out the arrangement per section. Later on, do things like fills and details. Don’t refine sounds or mixing until the end. Try to finish as many songs as possible without too much care for details or perfectionism.

B ) Mix ‘backwards’. Other than simple gain staging per track, start by mixing thru buses (or even on the master).

  1. Some psychology: I’ve fooled myself for too long thinking that ‘quality threshold’ is what separates a good musicians from a not so good. This isn’t true at all. Practice and habit of finishing tracks is much more important. A typical mistake related to this is to try to make the best part perfect before ‘finishing’ a song. The problem is that it will become much more difficult doing a verse from a blank slate when you sub-consciously compare it to a fully produced chorus. Get rid of perfectionism. At best, it’s a trait only useful for the finishing touches. Force yourself to finish tracks fast by using a top-down approach.
  1. Turn volume down instead of up in mix. If some tracks are too low, decrease the volumes of the other tracks. This way you’ll always have a lot of headroom.
  2. Use pink noise as reference and to balance your mix. I simply have one track playing pink noise while turning on one by one track and whatching the master curve while sliding the volume up to where it ‘touches’ the master curve.
  3. Start with low volume on tracks and rather put a temporary gainer in the master or just turn up the volume knob on your monitors/soundcard. Always nice with a lot of headroom.
  4. Don’t use the wrong reverbs and try to make it fit. Instead use a reverb that sounds good even without much tweaking. I’ve come to realize that finding the right reverbs is one of the most difficult part of mixing and if the reverb doesn’t fit well in the first place it’s nearly impossible to tweak it to where it fits well with the mix. It’s easy to think the reverb sounds better if you turn down its volume, and it does, but that’s only because it’s less of it, not because it fits better with the mix. Even though you know the reverb sounds good, it doesn’t mean it sounds good in any mix.
  5. Don’t use exiters/enhancers early in the mix, keep effecting to the minimal, save it for the later stages of the mixing. These kinds of effects can easily fool you into thinking things sounds better, though that is often not the case after you have balanced your track using pink noise. I speak from own experience, i have been quite surprised sometimes when using much of these kinds of effects, then when i turn all of them off it suddenly all sounds much better. Such a waste of time and energy.
  1. Here’s a thing to remember: kick dums have a pitch. Play them at a higher octave to hear what the pitch is and make sure that when you go back to the original octave that the kick is in the same key as the rest of the tune.

This might create another problem though, if you play a kick at a different pitch from the one it was sampled in, it might sound either too flabby (when played lower) or too snappy (when played higher). If you have to tune it more than one note away from its original pitch, this might cause it to sound bad. Usually, the transient at the beginning of a kick is practically pitch-less, while the tail is what creates an audible note. Try experiementing with stacking the snap of one drum with the tail of another if you have problems with the kicks (or use the transient and the tail from the same sample in different tracks). Use a signal follower or manual volume fades if the different kicks clash with one another.

  1. Pitch bending livens things up.

  2. Think about what you can remove. Do you really need all those instruments playing at the same time? Think in terms of frequencies. Is the upper range packed with hats and bright synths? Do they all need to be there at the same time?

  3. If you must have many instruments and tracks, try to make gaps between the notes such that one instrument or sound plays when another doesn’t. This makes things both funkier and easier to mix.

  4. Look out for notes and fx sustaining into places where they mess up the harmonics (like key changes and other transitions). Long reverb tails can cause problems, especially when they’re in a send track you forgot you were using.

  5. Whilst listening to your complete tune, pretend that you are another, more sceptical person, and ask questions like: “is the lead sound supposed to be that quiet?” - “isn’t that intro a bit long?” - “why does that bit repeat so many times?” - “is it meant to be, like, really monotonous?” and so on. Taking breaks makes this easier to do.

  6. Always take a break between completing a song and completing the mix. Listen to professionally mixed and mastered tracks in a similar genre for reference.

  7. But, if it sounds good to you, that is important too.

  8. There is such a thing as too much treble.

Turn off your monitor and listen.

Go for a walk and listen to it on headphones with your phone/mp3 player.

if u get stuck on a track somehow; hit that #random button on ur vst & move on from there.

Turn off your monitor and listen.

Go for a walk and listen to it on headphones with your phone/mp3 player.

not related, but dawg!! that track of yours [high horse] - bomb!

not related, but dawg!! that track of yours [high horse] - bomb!

heh, when he said monitor i thought monitor/speaker … was like… woah… deep. then i thought, monitor with no s? he mix in mono, too… then i got undumb for a min.

oh, u right, tho! meef makes a good song there!

heh, when he said monitor i thought monitor/speaker … was like… woah… deep. then i thought, monitor with no s? he mix in mono, too… then i got undumb for a min.

oh, u right, tho! meef makes a good song there!

Haha that would be some zen stuff right there!

Thanks :w00t:

I don’t turn of the pc (visual) monitor, but it is a good idea I think. I know the “effect”, it seems to be a psychological thing that will let you listen in a different way than if you got all the visualisers eqs mixers and stuff in front of you. I render in 192k (yeh those much better highs when you use distortions…) and bake a hq mp3, and listen in a media player, along with some random other music, with renoise off or sitting in the background and after a silent cup of coffee - to get this effect work the strongest way for me. It is like then is the first time when I am able to really hear how the mixing space is working, if at all. As if being in control makes you percieve things in a different, more complex but also difficult and wooly space somehow, that is not as musical and enjoyable compared to pure playback without interaction.

Before mixing and making stereo panning descisions (and partially width also), listen for a short while to sounds in mono. to calibrate your ears for center perception. Will lessen the effect of panning things tilted by ear. Use full spectrum sounds, l/r balance might be slightly different dependent on frequency. Especially with headphones - with such in over-ear fashion, you can also try to adjust their position on the ears to shift the center to what you consider the middle while listening to mono sounds.

Great thread!

Set me off on a bit of a like button marathon :yeah:

Got a bit of a request for any tips on groove and mid-range energy? Specific challenge: without using loops… (though they have their place)

try to recreate drumloops yourself with drum, hihat, kick etc. Try do disconnect from the lazy world and work in those notes :smiley:

These are tips, not rules:

  1. Gain stage at -18 rms between plugins for maximum fidelity.
  2. Put a Inter-Sample Peak meter on Master, or keep peak level below -3db.
  3. Cut all frequencies below 20hz on Master.
  4. Don’t mix into a limiter.
  5. Mono all sound below 100hz for maximum sub impact.

The two “best friend ever” effects, native in Renoise, are Gainer and Stereo Expander, often used together.

Why? Total control of stereo image, e.g…

  1. Gainer - because it has a Panner, which means you can “pan into” FX that come after it in the chain. A regular approach of mine in the insert chain is Gainer > Reverb, so the reverb receives a panned signal, but itself is stereo. This lets the source sound be panned where you want it, but the reverb image is stereo. I find this a very useful way to balance the stereo-image / feel of the track, where the dry signal is mono and panned, but the wet (verb) is stereo. I never use the panner on the channel mixer, because I can’t achieve this kind of routing with it!

  2. Stereo Expander - because it allows you to create Mono signals from stereo signals, including (and most importantly for me!) using just the Left (L) or just the right (R) instead of choosing L+R, the summed mono signal. Why is this important? Simple: Summed mono signals often sound weird!

To explain in more detail: When I want a true Mono signal for a track that was originally stereo, I usually will choose to drop a whole side by choosing L or R in the Expander, thereby creating a Mono signal from just one side of the original image. I think people don’t try this enough! The problem with the mono sum (L+R) approach is that ‘mashing together’ both stereo sides like this often results in a weird sound. Yet I’ve seen countless tutorials wherein people just assume this is the best way to get a Mono signal from stereo. It may be the most ‘accurate,’ as it includes a sum of all the content from both sides, but the reality is that the original sound was designed to be stereo, not mono, and therefore the L and R content wasn’t designed to be summed together, and often won’t sound good in that context. Ever tried summing a stereo snare and getting a weird, phasey sound? This can be avoided easily by using just the L or R instead of both. You can check both, see which you like better, then choose, and have a true Mono signal for the channel, and pan it accordingly.

Sorry I ramble so much for what could probably be easily expressed in two sentences, but there you go, my two tips re: Gainer and Stereo Expander!

cheers,

-Michael

These are tips, not rules:

  1. Don’t mix into a limiter.

Not to counter-act these tips, as they are all good, but I think this ‘tip’ can lead to some confusion. If the limiter is only ‘catching peaks’ and not adding any actual gain, I can’t imagine why this tip would be useful.

Meaning, I think it’s always a good idea to mix into a limiter, though not necessarily one giving a gain boost. If nothing else, you will save your ears from unexpected loud noises that can result from sound design! The limiter is your friend, making sure nasty peaks / unexpected boosts don’t make it through.

cheers, -M

The problem with the mono sum (L+R) approach is that ‘mashing together’ both stereo sides like this often results in a weird sound…

Agree about choosing L or R for mono,

Ive had better results with Fluxs Stereo Tool when doing summing though. It doesnt seem to mess with the frequency spectrum so much. You do need to watch when pulling the Width control on it down though as you get a volume boost. I should check its manual to see why…

https://fluxhome.com/project/stereo-tool-v3/

edit: Here`s a doofer that compensates for it though. The rotary goes from 0% to 100% width, so no extra widening is done. You need to install Flux stereo tool (freebie) first of course :slight_smile:

7836 Flux Stereo Tool Compensated.xrdp

Not to counter-act these tips, as they are all good, but I think this ‘tip’ can lead to some confusion. If the limiter is only ‘catching peaks’ and not adding any actual gain, I can’t imagine why this tip would be useful.

Meaning, I think it’s always a good idea to mix into a limiter, though not necessarily one giving a gain boost. If nothing else, you will save your ears from unexpected loud noises that can result from sound design! The limiter is your friend, making sure nasty peaks / unexpected boosts don’t make it through.

cheers, -M

Totally agree. Nice clarification.

I’ve always got a limiter on my Master to catch unexpected peaks.

But, during the mixing stage (after design/composition) it’s tempting to push the limiter for gain, and that’s what I was referring to.

The result of mixing that way is not so good. Best to save the limiter for the Mastering stage.

However, it’s not a rule, just a tip :slight_smile:

When initial mixing (and also designing sounds), it can be of great help for me to have some very subtle very even and a bit brightish with clear tail reverb on the master channel. Just so ever subtle that you only hear the difference when switching on/off with the same sounds, but not really notice the reverb otherwise - then a slight step up. Use good 3d type reverb, for me the convolver with some high quality plate type IR works good, other reverb types might also be ok. Essence is, you must have some good sense of space position of the sound in the reverb.

This can help enourmously to localize sounds, and make (good hi res) headphones much more bearable for mixing, and enhance localisation in the 3d mix on speakers also. I decided to mostly work thru such reverb setups now. Also listening to very direct sounds while sound design is greatly eased by this, especially with headphones - you just need a reverb that sounds very neutral, so you don’t make descisions around its coloration. “Subtle” is the essence.