Page 1 of 1

Shaker loops

Posted: Tue Jul 15, 2014 6:51 am
by helloitsmeagain
I know all the arguments, but personally i never use sample packs for anything except one shots. I'm in euro, so you know i'm into taking the long way around and may never finish a song. ever.

anyway. ;)

I'm trying to roll my own shaker loops. I've gotten decent results by laying a very sparse pattern down, applying a small measure of random lfo to the pitch and varying the velocity values in my midi clip and additionally applying a touch of lfo to the overall velocity. Then I'm treating this with stereo delay, left and right with different ms values, numbers set to some timings just off of what is spot on for the bmp. i'll narrow the stereo field way down and then i'll resample the 100% wet signal and it'll be my pattern. i tend to also apply swing to the final output.

I'm interested to hear your techniques.

**this is post number 404 for me. something symbolic in that i'm sure...

Re: Shaker loops

Posted: Tue Jul 15, 2014 9:16 am
by msl
Well shakers are one of the few things I do use loops for, you know you need that human feel for them to sound right. I've rarely had success by programming them only with static samples. I do sometimes record my own short takes and loop them, its pretty easy, all you need is an egg and a cheap mic.

Re: Shaker loops

Posted: Tue Jul 15, 2014 12:41 pm
by Críoch
I dont really use shakers apart from doing a 16th pattern or on the offbeat, pitched down.

Was looking at that imashine app for the iphone earlier. Theres a good vid from Rob Jones (??) at Loopmasters showing the velocity page in it. Basically you can tap in the velocity as the loop is playing maxed out 127 vol samples & they're affected by your tapping. Though it looked cool. Very human sounding.

Re: Shaker loops

Posted: Tue Jul 15, 2014 2:22 pm
by oddmyth
Can you explain why you are adding these LFO's for pitch and velocity and why you chose random as the waveform for pitch?

Pitch in a shaker really isn't going to vary all that much, especially an egg shaker. If you were using maracas it would make more sense as usually one is high pitched and another is low pitched but that's about the extent of it. If anything the pitch of the shaker is directionally proportional to velocity using a random waveform is going to produce strange results (albeit that could be cool to use too).

Velocity I can understand being variable so after programming your accents and initial velocity values for the shaker it makes sense to add a little variation to it over time.

I'm not sure what the stereo delay is there to achieve, most shakers aren't recorded in stereo and I'm assuming that the delay is there mostly to play out the upbeat shakes, but you would gain more control and a better understanding of the instrument itself without the delay in there. With it in there you are relying on generative process to make the sound and that has it's own caveats.

I don't have a particular process for shakers in general, but the same rules of drum physics and motion still apply. I try to find three unique shaker samples in a shaker loop, one for the foward motion, one for backward and one that is an accented hit. I usually have a volume increase towards the downbeat accent of the shaker (shakers are usually played away from the mic until the downbeat), then volume comes back down afterwards. Usually I pan shakers towards mid-right in the overall image of my mix, opposite of my hihats, after that the only treatment is EQ and glue comp if I want it to sit with the rest of the Porc or I'll resample through a particular amp/room to try or reverb to give it a different space.

Re: Shaker loops

Posted: Tue Jul 15, 2014 3:05 pm
by helloitsmeagain
oddmyth wrote:Can you explain why you are adding these LFO's for pitch and velocity and why you chose random as the waveform for pitch?
The amounts of lfo are tiny. Around say 3%. Generally I do this to most samples, to add a touch of humanity. It's not always random, sometimes i'll use a sine. The results are barely perceptible, which is exactly what i want. just the slight illusion of humanity.
oddmyth wrote:I'm not sure what the stereo delay is there to achieve
I use it add bounce, because left and right are offset by slight amounts. It's also almost down to mono when i'm done. say 10% width... honestly it's not necessary to retain any stereo.
oddmyth wrote:but you would gain more control and a better understanding of the instrument itself without the delay in there.
I agree, hence my question 'cause now my patterns suck without it.
oddmyth wrote:I try to find three unique shaker samples in a shaker loop, one for the foward motion, one for backward and one that is an accented hit. I usually have a volume increase towards the downbeat accent of the shaker (shakers are usually played away from the mic until the downbeat), then volume comes back down afterwards.
cool, nice tips.
crazeehorse wrote:A seriously useful trick with offgrid placement that ive been keeping to myself lately (till now) is to program the pattern waaay slowed down, i find 80 bpm works well. then speed it back up to where you were. same effect as zooming right in for greater definition. this is only useful if you are actually using a computer to sequence it though. I havent a clue how one would go about this in a controlled manner on an analogue sequencer...
Yeah, I don't use my Cirklon for drums 'cause i can't get it sounding anything but robotic. I tend to sequence drums with my MPC, so i have done this sometimes. I'm sure you know how it is, you don't use something regularly, you forget it! I should give it a try here. Oddly enough hip-hop heads on the MPC forum speak of programming beats in at double the tempo for tighter patterns. sounds like a fuckin' urban legend to me....

Re: Shaker loops

Posted: Tue Jul 15, 2014 3:34 pm
by oddmyth
helloitsmeagain wrote:
oddmyth wrote:Can you explain why you are adding these LFO's for pitch and velocity and why you chose random as the waveform for pitch?
The amounts of lfo are tiny. Around say 3%. Generally I do this to most samples, to add a touch of humanity. It's not always random, sometimes i'll use a sine. The results are barely perceptible, which is exactly what i want. just the slight illusion of humanity.
I can understand the velocity LFO, makes sense, the pitch however just doesn't make sense (to me) unless you are synthesizing the entire hit. For example what happens when your LFO bottoms out decreasing the pitch on a accented hit (which should be pitched higher). I used to run things like this all the time, especially in the interest of 'generative' music, but in the end I wanted to understand what I was doing so I stopped and decided to figure out how this stuff should really work.
helloitsmeagain wrote: Yeah, I don't use my Cirklon for drums 'cause i can't get it sounding anything but robotic. I tend to sequence drums with my MPC, so i have done this sometimes. I'm sure you know how it is, you don't use something regularly, you forget it! I should give it a try here. Oddly enough hip-hop heads on the MPC forum speak of programming beats in at double the tempo for tighter patterns. sounds like a fuckin' urban legend to me....
Nah it's real, the point you are missing is that when you program on a machine with non-infinite quantization (say 1/32nd being the max quantization), then if you program at double time (say 240BPM for a track at 120BPM) then each bar at 120BPM is two bars at 240BPM. So a 4 bar loop for a track at 120BPM is 8 bars long at 240BPM etc. This means that the quantize resolution at 240BPM is twice as large IF you are bringing the track back down to 120BPM - thus the quantize resolution of 1/64th, which can in fact allow a tighter sequence, assuming the machine won't auto-quantize as you drop the BPM.

Re: Shaker loops

Posted: Tue Jul 15, 2014 3:41 pm
by innovine
Mpc timing is 960 ticks per quarter note, so doubling the bpm and halving your music tempo (so it sounds the same) gives you double the resolution.

Re: Shaker loops

Posted: Tue Jul 15, 2014 3:56 pm
by Lost to the Void
I use shakers a lot.

I`ve a massive library of shakers I have generated from sanding different things with sandpaper, blasting from air hoses and opening and closing the air hose with my hand to make rhythms, sharpening knives on a knife steel, shuffling feet on a dusty floor, wire wool on the side of a metal dustbin.

Tons of stuff.

I always program the patterns myself, or just adjust the patterns I have recorded.
When self programming I try to keep the pattern loose, I`ll go "off grid" and let the notes slip in to the beat a little, velocity changes are there to help emphasise the offbeat or whatever.
If I am using sample hits then I will also change the note decay and release over the phrase, effectively like having an open and closed hi hat. That works nicely.


sometimes I`ll add ghost notes an octave down, which can add a melodic call and response to the shaker pattern, which sort of mimics the close to mic, far from mic you can get when working with real shakers.

Never used delay though, it seems to add mess. Unless it is micro delay to add stereo.

Re: Shaker loops

Posted: Fri Jul 18, 2014 3:00 am
by helloitsmeagain
oddmyth wrote:Nah it's real, the point you are missing is that when you program on a machine with non-infinite quantization (say 1/32nd being the max quantization), then if you program at double time (say 240BPM for a track at 120BPM) then each bar at 120BPM is two bars at 240BPM. So a 4 bar loop for a track at 120BPM is 8 bars long at 240BPM etc. This means that the quantize resolution at 240BPM is twice as large IF you are bringing the track back down to 120BPM - thus the quantize resolution of 1/64th, which can in fact allow a tighter sequence, assuming the machine won't auto-quantize as you drop the BPM.

Thanks for the explanation. Not sure my crappy finger drumming is gonna do me any good at 240 BPM... :?

Re: Shaker loops

Posted: Fri Jul 18, 2014 1:03 pm
by oddmyth
It's not 240 BPM though .. that's the point.

1 bar at 120 BPM is the same as 2 bars at 240 BPM is the same a 3 bars at 360BPM or n bars at 120*n.

All you are doing is using a higher BPM to gain smaller note intervals in between each note.

Take the 16 step sequencer. If you look those 16 steps as one bar, then each step is 1/16th note. Now if I double the BPM, there's two ways of looking at this:

1. This is 1 bar at double the BPM. Each step still represents a 1/16th note.

or

2. This is 1/2 a bar at the original BPM. Each step now represents a 1/32nd note.

Re: Shaker loops

Posted: Fri Jul 18, 2014 4:13 pm
by helloitsmeagain
Ah, I'd imagined to only set it to 240 when laying down hats! I'm a dumbass! Brilliant.

Re: Shaker loops

Posted: Mon Jul 28, 2014 7:26 pm
by GLinet
In addition to the velocity LFO I would suggest automating the sample decay slightly. That can sound quite cool. Sustain automation might work as well.

I don't have an MPC but there ways to implement "choke groups". On my DAW I sidechaining different hats/shakers to each other so that one set ducks out the other. That can create an interesting sound as well.

Sidechaining the shakers to the kick slightly can also add movement. You may drag your ghost kick off-time slightly or even add a groove to it.

I get some really neat sounding patterns by applying Echobode's frequency shifter/delay... but I don't know what the substitute is for that in the world of VSTs.

Frankly, though, as you said yourself... you can get lost in that nonsense and not finish anything. So I'm saying "screw it" more and more often, finding nice jazzy brush or shaker loops so I can actually finish my tracks.

Re: Shaker loops

Posted: Thu Aug 07, 2014 7:20 pm
by qb748t394
excellent thread for sure... I gotta question for you all.. especially Lost to Void... do any of you tune your shakers to the song?.. or since they aren't tuned in real life, you dont...?

Re: Shaker loops

Posted: Thu Aug 07, 2014 7:58 pm
by surface
They're mostly made of noisy kinda tones so it never really bothers me.
I usually run them through a band pass filter, max the resonance and tune the resonance by ear, then roll of the resonance again til it's low enough not to be too noticeable.
Not sure if this is standard practice though ..

Re: Shaker loops

Posted: Thu Aug 07, 2014 8:00 pm
by Lost to the Void
qb748t394 wrote:excellent thread for sure... I gotta question for you all.. especially Lost to Void... do any of you tune your shakers to the song?.. or since they aren't tuned in real life, you dont...?

Yes, always.

You can get away with not tuning your shakers, but when they are in tune, your percussion and your tune will work, your eq will work, everything will fit much more effectively.
You don`t need to be ear trained, just grab the transpose nob and move it around until the shakers fit your tune.

I can do a little demo video to show you the difference, but just try it, it will make all the difference.

Re: Shaker loops

Posted: Thu Aug 07, 2014 8:05 pm
by oddmyth
qb748t394 wrote:excellent thread for sure... I gotta question for you all.. especially Lost to Void... do any of you tune your shakers to the song?.. or since they aren't tuned in real life, you dont...?
I don't think there's a single thing I wouldn't repitch, shakers included. Repitch is just one of those things that makes electronic music so amazing. However I don't think it's a necessity.

Re: Shaker loops

Posted: Fri Aug 08, 2014 3:34 am
by qb748t394
i have already been myself... like oddmyth says... I frickn tune everything for drums... it totally makes difference...

surface: thats an interesting technique... gonna add that to the toolbox fuh sho'