Seconding the video's recommendation for the book "Dilla Time", which is also apparently going to be a documentary soon. Highly, highly recommended for anyone into electronic music production, hip hop, neo soul, Dilla, contemporary jazz, swung beats, wonky time, fucking around with rhythm, etc.
In theory quantisation should snap everything into perfect time, and quantised swing will still be perfectly mechanical. But there are a number of reasons why the timing might drift.
This is trivially easy to test. Use the pads to record a groove with a standard collection of samples. Replace the samples with clicks. Patch each track to the separate outputs and record into a DAW. Any timing shifts will be very obvious, down to individual samples.
With respect to Mr Linn, he's quite wrong about randomisation being uninteresting. Real drummers play with a Gaussian distribution, even when they're playing to a click. The tiny variations add a surprising amount of interest and liveliness.
But this works best as a dynamic performance effect, and I've never seen it implemented in a drum machine. Baking a randomised uniform distribution into a short repeating pattern is a lot less interesting.
Agreed re: there being randomness with each hit in a live performance. I wonder, though, if the distribution of one hit is in fact independent of the previous -- perhaps the inner feedback loop of "am I falling behind / speeding up" in a musician causes the next distribution to be skewed differently or centered at a different instant.
In which case I can understand the thought process of it being uninteresting, in a way. And you're absolutely right that when the randomness is applied into a short loop and repeated it falls into a strange uncanny valley (at least to my ears).
Perhaps another aspect of this is that the randomness, added based on e.g. a seed to the whole track, would be hard to control as an artist. I know when I apply randomness to MIDI I tend to go back with a fine-toothed comb and re-distribute hits that don't go where my mind wants them to.
Ive been using the MPC1000 since around 2004, and owned and used an MPC60 for about 5 years. I recently had my 1000 serviced by a guy who does nothing but MPC refurbishments/customizations and he asked me how I felt about the whole "MPC Swing Magic" and what not.
My feeling is that the MPC60 really DOES feel better, but mostly because of its velocity sensitivity. You can really easily get a ton of dynamics without much effort at all. Much much more easily than with a 1000, or a midi keyboard, or any number of pads devices that I've tried. and yeah that for me includes machines which allow for variable velocity curves.
The thing that blew me away was that he told me most of the guys he's selling MPCs use them on FULL VELOCITY at all times. that making sure the pad sensitivity was in perfect condition was not a priority for them. I was shocked... but he was equally shocked I felt it was the velocity which makes the difference lol. Keep in mind these are pretty hardcore old school Chicago producers buying the best of the best vintage machines.
That's equally shocking to me but I guess it makes sense considering what the industry seems to prioritize. I got an Elektron Model:Cycles on which the only way the pads are even really usable is to set the velocity on them to full (or turn the velocity modulation depth so far down as to be mostly nonresponsive). Or, I guess, bash them so hard they'd probably crack the circuit board within a few months.
I have a Korg PadKontrol (now, disappointingly, discontinued) which has really nice velocity response and even lets you set different response curves to each pad. Not sure why they pulled them when there are very few alternatives.
I picked up my first MPC (the 2000) around 1999. Since then I've had the 2000XL twice, an MPC60 and I current have the MPC2500. I'm one of those guys that always had it on full velocity. Whenever I wanted a kick or different sound at a different velocity, I would just assign it to a different pad and add adjust the volume and maybe add some additional filtering to it.
For me, I found the 2000XL gave me the best workflow. It took a while to adjust to the 2500's way of doing things. Everything is hidden behind a menu, but it's easier to move stuff to and from the computer.
If you want to make bangers the priority is to get your groove down as fast as possible and add nuance at the mix stage. If you want to craft stuff with velocity sensitivity, layering etc, you're now more into composing than recording and the speed of your workflow is gonna be significantly slower, more half-written abandoned tracks etc.
I am in the latter class btw, because that's the kind of tweaking I enjoy. But the best way to finish tracks is to have a timer running and just bang shit out without thinking too much, pure instinct.
>But the best way to finish tracks is to have a timer running and just bang shit out without thinking too much, pure instinct.
This ^^
Especially in the jamming or "feeling" phase, it's about having your tools (in this case, a few pre-made sounds, or kits) available as quickly as you need them, so you can get into the flow of the track and not lose it. It's also why when I make beats or tracks, I oftentimes find myself composing the peak of the track almost by default and work backwards from there.
Also, USE TEMPLATES!!! Templates are considered a form of cheating by some, but I don't ascribe to that, and I love using templates because it can help me progress so much faster in each phase of production. My default Ableton setup for 4/4 house music literally has every section of the track already set, the default sounds (drums, samples, synths) I need to use to get a groove going, levels pre-adjusted, and everything is color-coded for ease of viewing....and I just go from there.
I saw a video put out by Ableton maybe 2 or 3 years ago that had some artist saying "just get something down in the DAW, and don't worry about what that something actually is - because you'll replace everything later when you're not in the ideation phase of production". The other thing is that if you mix the ideation and production phases, you'll get bored of your track very quickly because you aren't taking breaks, just hearing over and over what you've made so far. It's happened to me literally hundreds of times. I have probably 1k+ ideas (could be a track that's 90% done, or a few loops put together, or just a MIDI sequence that i was goofing around with) sitting on a variety of old computers and hard drives. None will ever be finished, partially due to my inability to set a goal, accomplish it, and close the DAW.
Some of the best production content I've seen has been "Against The Clock" videos where the artist (or artists) have 10 minutes to make something. They just flow for a bit, and then revisit it later if need be.
I forgot to include it in the original and the edit window has closed, but I meant to add that one of my favorite things I've made got done because an unfriendly roommate went out for a few hours - finally giving me the time to be in the comfy room and hear what I was doing on speakers instead of headphones. I had a not-terrible hardware sequencing idea, it kinda worked, did 2 or 3 passes until I got something interesting, added a couple of knob-tweaking synth overdubs, some very light effects, and got it onto tape and then back into a computer.
I only use the MPC 4000 for sequencing I need to talk to this shop you are using these things are getting harder to get parts for. I wish someone would make a replica with a copy cat OS
unless I'm turning the pads on full velocity, every single sampler, sequencer, midi controller, etc is going to destroy my hands with pain within a few minutes of tapping around.
I find its much easier to just edit velocity manually after the fact.
Can you talk a little bit about the origin of the sequencer in the Tempest? As in, did you start from scratch, were you looking to emulate a particular preexisting model? I recognize that the Tempest was built on different hardware than anything that I presume Roger had ever worked on, but I imagine he contributed to some conceptualization that you then had to translate into a functioning OS.
A long time ago now, I wrote a crude Javascript emulation of a Boss DR-110, and when I got to the sequencer part, I threw in the towel because it seemed to be a step beyond the "hobby" level of effort I was willing to commit, lol.
I did the whole thing from scratch. I was not trying to emulate anything in particular, I was working with what we had in the hardware to optimize. We basically pushed the chip to its limits without going heavily into assembly, although if I were to do it again (WHEN I do it again...) it'll be way better =)
I spent a majority of my life during that time doing everything I could to update it on a regular basis and burnt out. Sometimes you just have to let it go.
Is there anything that could in theory happen to allow the firmware to be open source, so to allow the community to continue with it? I figure in some time period, people will figure out many aspects (as they've done with the Machinedrum, and now have community driven firmwares).
(I have a Tempest, and I'm decent software dev. Wouldn't mind helping maintain a project around it)
Sadly no, for a number of reasons not least of which is multiple CPUs and too much necessary documenting to do that and even if… I don’t think Focusrite would be too keen on it
I know the answer is "no, not that either" but I'm still going to ask: What about sharing the spec of the Tempest's sysex format? Sure would be nice to have the opportunity to build a patch editor/librarian - but my assumption is, if it was reasonably doable to share information that could enable that kind of thing, that choice would have been made by now.
I treat the Tempest as a standalone instrument that happens to output MIDI clock. Nothing's gotten in the way of it being my favorite instrument (at least, once free-running LFOs were added - absolute game changer for claps and snares), and if I had a magic wand, the only thing I'd change is to make the 'noise' samples have a random start offset on the waveform playback (something I used to do with Impulse Tracker), which I think would eliminate the phasing that sometimes happens when two voices use the same noise sample. Maybe next time...
I definitely have given out the sysex info to people. Post on the forum and see if someone has it accessible, if not I’ll look through my emails to find it when I have a moment
Roger made sure it sounded 'right' but beyond that I had a lot of control over the specifics. I shot for a much higher degree of precision than other instruments (even lots of digital) and more or less hit the target I was going for. Low jitter, under 3ms in all cases I could push it into. Was quite proud of that
Oh tons and tons of design ideas... I plan on blogging about them at some point once I am more steadily designing my own instruments, right now it would just help the competition. Been writing out articles. I'll post some here when I do
It depends on the bpm and the intention. A swing for jazz is much different than for house or hip hop.
If some users don't get angry then it isn't innovative enough. Innovation naturally leads to people forced to learn things out of their comfort zone and everybody responds to that differently. The best I can do is guide them through the process and hope they don't resent me on the other side ;)
> I can’t help but imagine a skilled drummer quietly chuckling inside when they see someone going to so much trouble in order to avoid learning to play the instrument skillfully.
I got my start on tracker software in the late 90s. I got meticulous about drum fills, and custom patterns that had little nuances so that it wasn't just the same loop playing for three to five minutes.
In the early 2000s, I joined a punk band at their drummer, and over a couple of years, got to be pretty okay at it.
When, a few years later, I found myself in front of a drum machine again (the Tempest mentioned in this article) I suddenly found the idea of "writing beats" to be absurd. I still recorded patterns, but most of the time, I was tapping them in, in real time. I reflected on how silly it was doing things how I used to, now that I could just... Play the drum beat
I once built myself a parametric real-time rhythm sequencer - a little black box with knobs and buttons and lights and an LCD screen, which spat out MIDI notes to drive a drum machine - because that seemed easier than either learning to play the drums or learning to program drum patterns.
A neat thing about the ubiquity of cheap microcontrollers like the Arduino family is you can easily try this sort of stuff out at home.
Given a MIDI interface and something to trigger, it's not hard to see how counting six MIDI clocks for every trigger will give you straight-ahead semiquavers (16ths) because MIDI clock runs at 24 pulses per crotchet note. With a bit of thought then you can see how pairs of semiquavers will be 12 pulses long, which you can split up any way you like. "50%" is obviously six and six, 58% shuffle would be seven then five, and so on.
But you don't need to stop there. Imagine making a list with a velocity for each hit, and a "time to next hit" in clocks. As long as the total number of "time to next hit" added up to the right number of clocks per bar, the actual notes could be anywhere ahead of or behind the beat. If you deliberately used the wrong number of clocks, you'd end up with your pattern drifting out of step with the rest of the sequence, and if you do that right you've just (re-)invented polymetric rhythm.
https://youtu.be/B_fZVUn44h4