Autechre “We wanted to have something where we could change, at the last minute, any detail or any event.”

It’s hard to argue that Autechre haven’t changed since they first met in 1987; it’s even harder to argue the IDM duo from Manchester has not changed constantly with each album they create. With their new album Oversteps, they present a new approach to recording, as well as new takes on favorite methods of composition. Overall, though, the result remains the same for the listener: A unique, provocative experience that will be unheard of again.

Tiny Mix Tapes spoke via phone with Rob Brown in London concerning the development of Oversteps (TMT Review) and previous album Quaristice, as well as the use of software, algorithms, and the meaning of hip-hop.


Let’s talk about Oversteps. What can you say you have done different this time? Over the past 5-10 years, in fact?

Every album’s a different step in a major direction, I guess. There’s nothing new in that respect. Every time a new album is released, we tend to talk. So we usually come back off tour wanting a completely fresh outlook. You know, start from somewhere else. Oversteps is a bit like that: Characterized by a general pattern of going out, promoting an album, doing a tour, coming back and wanting not to do anything like the previous two events. We kind of just set about doing tracks without any premise of what the albums should be from the outset. We tend to just work on a few tracks. So there’s nothing different about each album in that respect. We tend to just come back and start fresh tracks, maybe dig out a bit of old equipment that hasn’t been used for a bit, and see if we can get something fresh going instead. Just something alternative to what we’ve been doing. Often, a few tracks will end up on a shelf. Then a year or so later, we’ll dig out all our favorites and compile them …

A major difference from Quaristice was that Quaristice was, in fact, loads of jams to tape. Just loads of ideas down onto the hard drive without much notation in a computer. We’d come back off the previous tour, and we wanted to archive our live set in the studio. That way, we could reset the gig, wipe the machines. We got totally carried away with Quaristice, and it ended up being that we’d done loads of tracks that were nothing like our live set. So we ended up with loads of jams, just stereo jams. We couldn’t manipulate them in post-production, for they were just big audio slabs. The best we could do was edit them down to get the best ideas, or juxtapose a few of the edits.

What we didn’t want to do this time was to be so fixed to these finished tracks. We wanted to have something where we could change, at the last minute, any detail or any event. It was sort of a dichotomy, knowing that the gear was safe, everything was saved, but at any time we could completely change the idea. This is unlike Quaristice, where we were pretty much fixed with what we have. We felt it was useful that, seeing a thread, we could this entertain this thread and develop it with that in mind, because if ever we thought it was a bad idea, we could back and redo the lot of it. Even at the mastering stage, we could change one small element in one track that perhaps wasn’t fitting well with others. In the past, if any track didn’t fit in compilation, it would get axed or shelved for some other project.

We managed to keep it really open, really solid, and versatile at the same time. We felt free to go a bit mad and do all sorts of things that you wouldn’t normally commit yourself to do, because we knew we could undo it, or we could add to it, or infuse more ideas in there. We still had the framework that was alive, it was still intact and active throughout its entire shelf life. Obviously, there are certain tracks that didn’t follow the common thread. They just served to bolster the narrative curve that we saw. At the same time, we weren’t committing certain tracks to certain positions in an album. We still kept the mentality of keeping it open until the compilation at the end. But we allowed the character of every single track to suggest new characters that were missing that we could induce into a new track. Therefore, we could fill in the blanks in a narrative scope. We could almost perceive how the album was going to be earlier on than in the past. It used to be that we had loads of tracks on a shelf, and were really happy with a lot of them. Some of them we couldn’t get into, or they were waiting for their special moment to shine in a certain context. But it was a superstitious thing in the past to count the number of tracks you had in case you didn’t have enough. The counting was a psychological blow to us. This time, it was a lot more positive as a result of having the freedom to know you could do anything you wanted, pursue more intense ideas without all the drawbacks of the past, if you will.

Indeed. Going back to the notion that Quaristice had a very fixed element to it, what would you say triggered that fixed element? Was it composing on a computer and using languages such as Max/MSP and Kyma, which you have used in the past, or was just a mindset at the time?

In the past, a lot of tracks were islands on their own: Items that were at risk of not fitting in with other tracks that we’ve done. With Oversteps, we knew that a lot of the tracks had less drawbacks with other tracks in a similar period of time. We knew there were certain pockets of time that exhibit a really good framework to follow. So it inspired us to do more fitting material next, as opposed to “well, we got all that covered, let’s do something that changes the pace or the game.” With this album, we were really trying to encourage or nurture a certain feeling.

“We felt free to go a bit mad and do all sorts of things that you wouldn’t normally commit yourself to do, because we knew we could undo it, or we could add to it, or infuse more ideas in there.”

In regards to using computers for composition, what equipment did you use primarily in creating this album?

Just the studio, really, a lot of computer stuff. We used a little of our favorite outboard gear, like a few FM synths, a couple of our modular analog synths. But they were all MIDI’ed up. We made sure that the computer had complete control over everything. We wanted to every item to be accounted for, to be recallable at any moment. If Sean (Booth) and I were sharing files a lot, we didn’t want the files to be huge. If we record a piece of audio, say with an analog modular synth, the best way to have Sean duplicate my setup would be to have the same synth settings as I got mine. But a lot of the parameters are hard to recall with all-analog gear. So we found ourselves drawn to more recallable parameters that you’d find on digital gear. It meant that we could transfer ideas and files, usually small Logic and Max files, really quickly. No recorded audio or such to worry about.

That transfer thing sped up completely, and had an exponential effect on our working method. If I was up at Sean’s studio, we could still work together at weird or normally unsociable times. We could do sort of real-time streams and synchronize our beats, and work on tracks in real time together even though we’re in separate studios at certain times. It was quite good to have a fluid approach to certain things without missing each other’s company. In the past, we’d do our tracks separately and individually. We’d remix each other’s ideas. We’d start something off and the other would finish it. We got this all-over-the-place work method. There is no fixed role for me and Sean to follow. It meant we could be open-ended about the process.

While it is clear that you are using software languages, are specifically still using Max/MSP and Kyma? Or have you gone on to other languages, such as C-Sound?

More Max really. Kyma’s cool, but we haven’t touched upon it much this time, mainly because we wanted to keep that snapshot approach I’ve been mentioning. We only have one Kyma device, so we didn’t want to spread ourselves too thinly in that department. We wanted to create duplicates both in his studio and my studio, so if ever I had an idea I could take a snapshot, forward it to Sean, and he’d have a duplicate to recall. He’d have my sound straight away. With Kyma, it wouldn’t have been as simple because we’d only have one device. So we kept it to devices we had more than two of. It was mostly Max/MSP, Logic, and some duplicate outboard hardware that we have.

It sounds like your compositional style takes a MIDI-like approach in this regard. Is there still relevance to MIDI, especially in your work?

Yeah, MIDI’s still relevant. It’s such a lightweight medium, in terms of data, so it’s easy to transfer information. It seems like Ableton Live! doesn’t cover the full scope of MIDI, whereas Logic would. Max/MSP you can develop the setup yourself, so there’s no limit to what you can do there. We managed to keep it amongst Logic and Max.

I think a lot of new people, they use internal systems where MIDI’s irrelevant. But with MIDI, it’s a digital interface, meaning more than one component is alive at any one time. With us, that is considered typical of our methods, so MIDI is essential, more or less.

“If ideas conflict, then it’s a matter of settling which idea is best for the Autechre world. It’s almost like there’s three members of the band: Me, Sean, and this Autechre thing.”

How much do algorithms figure into your compositions as of late?

Quite a lot. Algorithms are a great way of compressing your style. One of our strong points developed in the past is the concept of experimentation. A lot of people think we’re very improv-based, and it’s true to an extent. But if you can’t go back to that spark or moment where you’ve created something new and reverse-engineer it, it can be lost to that moment. Sometimes, we try to capitalize on acquiring that moment and re-using it again and again, or finding the essence of it and applying it in different directions. It has always been important to us to be able to reduce something that happened manually into something that is contained in an algorithm. Then the algorithm allows us to add a bit more flair or a bit more deviation that we would also do ourselves in a little script. Just a few slight tweaks can spin it out into all sorts of recreations. It’s a great way to spawn yourself if you like (laughs), and spawn your actions. It’s an addictive way to work. Programs like Max allow you to reduce these ideas to collections of numbers and scenarios that are recallable, cascade-able, even nest-able.

We’re happy with it. I think that’s what changed us from being strictly hip-hop kids. With hip-hop, there’s always that flair in hip-hop to be fresh and to be new, to be a game-changer, even though you’re well within the rigid rules of writing music for hip-hop or electro or house. Working algorithmically allows you to specify those rules, yet flick out into the nth degree.

Now, does specifying those rules give you more or less restraint?

Depends on the track. Some tracks are complete jams of one particular idea, taken as deep as it can go. Other tracks are allowed to be more vociferous, making their own mark to a certain extent, and then you’ll get the bits that you like that weren’t expected and go back and see how it occurred and see if you can develop that further. So, it’s definitely a freer thing, because if we found ourselves penned in by something, we wouldn’t last too long. You don’t want to get hemmed in by too many things. We’re always trying to re-approach what we do without too much cognitive reasoning.

I think it is only times like when we’re talking about our work that we’re really engaged with addressing these things. When we started out, there wasn’t much press, we weren’t really talking to other people. Me and Sean would share a really similar idea of what our musical language was without any prior musical knowledge or training. It was quite easy for the pair of us to just get on with it. But as soon as we start to speak to the outside world about it, it gets a little bit trickier, especially when you’re talking to academic people, or opposite academic people. You’re sort of in this weird limbo where you can’t really express yourself. So, trying to talk about the rules that we employ or the lack thereof is quite difficult because they don’t come into being as ideas until the tracks are done, the album is finished, and we’re about to go on tour.

Do you use any other generative practices, particularly with Oversteps?

No, not really. It’s mostly like … do you use any musical equipment yourself?

Kind of. I have a Korg keyboard and use Reason, but seldom work with it these days. I have played around with Max/MSP, but I really need to learn that to make use of it.

Max is really brilliant because there are so many tiny modules you can nest within each other. So that’s a great starting point. But even back when Logic Audio was eMagic/Creative Logic, they had environments where you can patch delayed objects and repetitive objects together and pitch-changing objects. Before you knew it, you could create these cascading effects that gain a life of their own. We’ve been really into doing that for a long time now. Any bit of software that’ll afford you a little bit of algorithmic behavior without too much scripting and still quite a visual language if you like Works. I am a really bad programmer; I’m just a straightforward person when it comes to that. I’m not that boffin when it comes to scripting in C or C+ (sic) or whatever. It’s just your basic tools in that regard, nothing too exotic.

With programming, it’s a mindset thing. I’ve talked with a couple programmers about it, and you have to have that mindset to really be into scripting and such. Otherwise, it just becomes tedious.

It just feels like work, it’s hard work. Sean is a little bit different. He had a home computer at a very early age. I wasn’t that tweaky, I suppose. When we’re together, we’re not that tweaky either. We won’t really employ scripts that much. Sean might dabble a little bit, when he does his own bit, but it doesn’t go too far.

How much does sampling matter to your work of late? There’s been a lot of talk that sampling has been a bit overplayed to certain respects.

A lot of good hip-hop at the moment is synthesized. I think it was only a matter of time before it became more synthesized than sampled. Yet, when you can’t see the difference, when you can’t tell whether it’s sampled or synthesized, then it is definitely on-point. We used to employ really good samplers like Ensoniq’s EPS, ASR-10, and Kurzweils because they could take sampling to that synthesis sort of level. So it’s never been a big question to us.

I’ve got an MPC sampling workstation. It’s got a really nice feel to it and makes you work a certain way, and there is software like Renoise, which is all samples. But it’s so easy to get more synthesis inside, especially with computers being the host. Synthesis is in a weird transitionary fusion stage right now with samples. It’s hard to say.

Hip-hop to us is about being fresh, doing something that quite new. It’s something that is competitive and more upfront than the existing material, to get props from thinking new things. The old mentality of being competitive and being fresh is still with us. Hip-hop, scratching, and sampling were very much a DIY culture once sampling became affordable to people with no money, as opposed to your Peter Gabriel-like mega Fairlight setups that cost thousands.

It’s not even a question we bring up much. We tend to follow our noses a lot. If it’s going to be a synth or a sampler situation, I’d go with a synth if I wanted flexibility. But that’s just the mindset of 10 years ago. Samples are so easily re-synthesized in software, it’s almost a moot point.

“Programs like Max allow you to reduce these ideas to collections of numbers and scenarios that are recallable, cascade-able, even nest-able.”

You’ve been with Warp Records from the beginning, and they just celebrated their 20th anniversary last year. How do you feel, being close to a label like that for so long?

Really lucky. They see things quite a lot like we do. It might be because of where they grew up. Sheffield was only 30 miles from Manchester, where we grew up. Though it was a different county, the same ideas applied. Also, I think it was inevitable. The first few tracks they put out by the Forgemasters and Sweet Exorcist and LFO and Nightmares On Wax: It was just the sound we were into at the time, or filled in a few blanks we already knew existed. We were really impressed to see that someone was exhibiting similar traits that we were doing in private bedroom studios. Hooking up with Warp was like saying, “We do music a little bit like all your guys, yet a little different from all your guys. It’s a little weird, but do you want to listen?” and they were like, “all right.” Me and Sean have been working together for 20 years, anyway.

For them to be going that long and for us to still exist is a testament for the trust we have for each other. They never really tried to manipulate us as artists for any great games. They allowed us to decide how things were going to go out, or the cover art, or the track listings for the album, and whether we were going to tour this region or that region. They were quite open about it. They offered us advice or we asked them if we needed any opinions on a wider scope. But they pretty much let us do our own thing. For a label to do that, that’s pretty unique for a start.

The next thing is that they’re independent, and they do a 50-50 split with all the proceeds. So at the same time you’re doing the tracks, they’re working hard to get their side of the share. It wouldn’t be the kind of thing you would get with the major label, where they pump you full of money to get this desired result, and if it doesn’t work after a couple of attempts, they’ll just scrap the whole project. I think that Warp has been noble in that regard. There have some low points and some high points, but they’ve never held us for ransom, ever. I have to respect them for that.

Another thing is, they have never stood still. They never stuck in one sphere, in one world. They’ve been quite keen as people running the label to open up to things, such as film, animation, and different kinds of music. It’s quite refreshing, and I feel very free as an artist and as a person to be involved with someone like that. Not much pressure in the wrong direction at all. So, it’s just perfect, I guess.

Again, you’ve been in the business for over 20 years. Do you feel more or less restrained from the sum of your work, as well as the various outside influences in the world?

It doesn’t often occur to us that we’ve been doing it this long. “Two decades” sounds like the most ridiculous pair of words to put together in context about me and Sean. I don’t know why, maybe it’s because we met when we were quite young. We still sort of feel that young. That partly reflects the approach we have. We don’t pressurize each other. We compete with each other, ferociously. But where our egos are concerned, even now we’ve developed a protocol that’s still fluid but its foundation is: If you got a good idea, then try it. If ideas conflict, then it’s a matter of settling which idea is best for the Autechre world. It’s almost like there’s three members of the band: Me, Sean, and this Autechre thing. [The thing] sometimes takes prevalence, for if you got two separate ideas and want to combine them, then you take the best of both.

That’s where we are strongest, for rather than shaking us apart and breaking up … I think the best analogy is that when we first met, we had a really similar record collection. But whereas Sean had something missing, I might have had that record he wanted. So we’d share it. It’s really good to know there’s someone that’s fairly compatible but totally different as a person. He’d go about achieving a similar goal with totally different means, totally different methods. That’s what we learned from each other a lot. We also manage to compete each other in different, sort of strange ways, but pursuing a common outcome. Nothing’s changed in that respect, we’re still like that. We know each other a lot better, so at the same time we can manage to avoid things that would normally catch a lot of people off.

Most Read