Creating Ableton Live 10 – The Story Behind The Update

Ableton Live 10, now available, promises to be one of the biggest releases ever for the popular DAW.

Highlights of Ableton Live 10 include four new devices; a completely redesigned sound library; major workflow improvements; Push enhancements; and full integration of Max for Live.

We recently had a chance to sit down the product owners for Live, and get a behind-the-scenes look at at the creation of Live 10. We talked with Amaury Groc, Product Owner for Live; Dylan Wood, Product Owner for Sound & Devices in Live; and Simon Hayes, Product Owner for Push.

They shared an inside look at what’s new in Live 10, and how they decided what to include in the update.

Elisabeth McLaury Lewin, Synthtopia: To start off, could each of you introduce yourselves, give us a little bit of your background and what your focus is for Live 10?

Simon Hayes

Simon Hayes: I’m the Product Owner for Push.

I’m actually an engineer by trade, but I’m now working in quite a different role, but I enjoy it quite a lot. I’ve been at Ableton for maybe five years. I started at the bottom and I’ve worked my way through sales support and went across different areas.

I did some user research, as well, so I got a good feeling for all of the customers, and I hopefully have taken that with me into the product development.

I work with a small team; we’ve been working on Push 1 and now Push 2.

Amaury Groc: I’m the Product Owner for Live. I’m here for soon 13 years.

Before that, I was freelancing as a sound engineer in theaters and radio and studios, a bit of everything, and also composing music for theater and then shows. I came across Live and used it, and used the forum.

So I was just simply a user and somehow there was a door open and I put my foot in it. I also started in support, technical support. Only three people at the time, because it was a while ago. And like Simon, slowly … in different tasks but it led to the product team, somehow.

Dylan Wood: I’m what we call the product owner for the sound and devices and Max For Live and the content stuff that we work on. I’ve been here for about four years.

Before that, I started out much the same way. Working, selling music equipment and playing in bands and producing and doing a bit of commercial music production. And then I worked at Serato in New Zealand for seven years.

I started in support as well, and then got into doing product development and that kind of stuff.

Synthtopia: How do you go about deciding what to change, and figuring out what new features to include in a major update of Ableton Live?

Amaury Groc

Amaury Groc: It’s kind of a process. Some changes are set, where we know we want to add or change something. But there’s also flexibility, especially in Live 10.

This time, there are a lot of what we used to call small improvements. Actually, those changes are not small to people, because they save them a lot of time.

Synthtopia: Do you visit Live users and analyze how they use Live in their own studios and performances to get ideas?

Amaury Groc: We decided to focus on dedicated music makers for Live 10, which means people who are passionate, who use Live regularly. It doesn’t mean they have to be professionals. They don’t have to be super proficient, necessarily, but they have a drive to use the products, use the software, a lot.

We work with the testers, but also we decided on improvements to make by visiting people. We increased the amount of field research over the years, more and more and more and more, and this pays off, because going to people, talking to them, they tell you what they want.

But also, observing them is enlightening.

Synthtopia: Does it ever surprise you how people do things?

Amaury Groc: Definitely. There are so many different ways of making music.

I had a trip to LA two years ago, and we went to, like, 14 studios. It was everything, from the Hollywood composer who’s at his desk and he’s doing some composing for a picture, to younger people jamming on a laptop, literally like stopping and starting and re-sampling, and all of these people are working with the same software.

So, yeah, that’s surprising.

What’s also good is to balance time spent talking to people and [time spent just] observing. There are some workflow improvements around navigation and zooming in Live 10, and they come from our direct observation of users.

We observed that some people spend 30% of their time zooming and scrolling through the arrangements. So we did a key to just zoom to the selection. Users didn’t complain about it (navigation) – they didn’t necessarily ask for a change.

I met one guy actually, and it was like painful, watching him struggle with zooming and scrolling. He doesn’t even mention it. I don’t want to point it out to him, because he doesn’t even see it. But coming back to work on the update of Live, it’s like, “We need to do something about that!”

Synthtopia: It’s interesting that some of the improvements or added features are things that you had to get from observation, because end users may not even realize that there could be an easier way.

Amaury Groc: It’s the same for all of us using a product. We just use it, and we work with the thing, and we don’t necessarily see what it could be.

Synthtopia: So how long ago did you begin to work on Live 10? Was it as soon as Live 9 came out?

Amaury Groc: Yes, in part. But as I said, I think we polled and did all the things, like Live 9.5 …

Simon Hayes: It was different for the Push team, because we also worked on a bunch of Push improvements right the way up through until [we released] 9.7.

We were working quite a long while in Live 9, and then we switched over.

Amaury Groc: And then Push 2 was … When was it released?

Simon Hayes: 9.5.

Amaury Groc: 9.5, so it was in the middle of the Live 9 cycles.

Simon Hayes: We had Push 2, which we were developing, so, yeah.

Amaury Groc: But after Live 9, part of the team started working on Live 10 stuff.

Synthtopia: When did you transition from working on 9.2 and 9.5 to saying the focus is now on Live 10? How long ago was that?

Dylan Wood

Dylan Wood: That depends, because the different teams transition at different times.

There’s not a point at which you go, “Right, turn all that stuff off and we’ll turn this on.” It’s more of like a cross-fade, in some ways.

Amaury Groc: Yeah, actually, it’s a cross-fade!

Dylan Wood: So….the majority of the changes have kind of actually started landing in the software in the last year, year and a half.

That’s when you actually start to see the change in the software.

We’ve been making music with Live 10 for maybe a year and a half?

Simon Hayes: It was prototype stuff in the beginning.

Amaury Groc: But the early phases are more research and field trips and prototyping. And ideating and debating and updating. That type of work.

Dylan Wood: Yeah. Like the research-development kind of stuff, where it’s like a lot of the stuff in the wavetables being researched for over many years.

Not actually in Live, but as part of some of the oscillator technology and stuff.

Amaury Groc: The Capture feature, as well, it landed in Live 10, but it’s something which we had the idea of a long time ago.

The researcher here could enable it, who also spends his life thinking of these types of problems.

Synthtopia: So what were the problems or challenges in Live that you wanted to address in developing new feature sets for Live 10, and the integrations with Push?

Amaury Groc: There’s never gonna be one [way to determine problems or challenges], because there are many layers and many users of the software.

I think one layer is through field trips, observation, talking to people, and developing the intuition of some changes that could help.

The general thing, I think, for Live 10, more than even all previous releases, is to make the music maker’s life better and easier. Not in the sense of dumbing [down] the process of music-making, but instead by removing barriers. I like to say that if there’s less software between the musician and the music, it’s better somehow. They can get faster to their results.

That idea has led to a lot of the things we have in Live 10.

We want to listen to people, and at the same time, do it the way we think is right for Live.

Synthtopia: Do you feel like more of the features and changes come from user requests, or from things that people don’t even know to ask for, like the time spent scrolling and navigating?

Amaury Groc: It’s a balance. I can’t give you a proportion. I’m tempted to say half/half, but I don’t know.

We consider every feature request and try to honor them as often as we can. My official title is “Wish Manager,” here, because I was “the guy for the wishes.”

Someone asks you for something, and then I love to ask questions back to see what exactly they mean. But then you talk to 20 people who wish for almost the same thing, and then we think back, “What do they really need? What for?” and et cetera, and then it leads to some improvement.

Also, wishes and needs blur.

There’s the wish that we directly address, and then there’s something that people don’t even know they need. And there’s everything in between – which is the reinterpretation of what people wish for, if that makes sense.

Dylan Wood: Yeah. Because for the device stuff, we did Wavetable, Drum Buss, Echo and Pedal. It’s not like we were sitting on a list of people who were saying, “We need an analog model guitar stomp box or we’re not using Live.”

It’s kind of emerged from, like you say, meeting with people and watching what people are doing, or looking at what they’re trying to achieve and figuring out where they’re getting stuck.

A theme that emerges. It’s maybe not specifically what the users said or asked for, but you start to see a gap or a hole.

And also, with the instruments and stuff, a little more and more of a place we could see people wanting to go, even if they’re not quite sure that they want to.

So Echo has been one feature that people have responded to really strongly. People are really happy about it, and I think in some cases, it does a bunch of stuff they maybe didn’t even know they wanted, but they found it inspiring. There’s a bit of that, as well

We look at what people are already doing, but we also have to do some work to try and imagine what they would like to do, or where they would like to go, and help them do that, as well.

Synthtopia: I would think anticipating needs that the user doesn’t even know they have yet would be really challenging.

Amaury Groc: I think that’s part of the part of the trade-off an instruments-maker – when you build some new instruments, it’s not something people have been asking for.

That’s part of the job, somehow.

Dylan Wood: When people ask for something, you have to work out what they actually need. That’s a big part of it. “I want this.” You have to work out what users actually need when they ask you for something. That’s the fundamental part.

But then also you have the other side of it, which is looking beyond what they’re asking for in a product.

Amaury Groc: Inventing something.

Synthtopia: With something like the circuit modeled stomp boxes, is that a matter of looking beyond your current users also?

Dylan Wood: It wasn’t really the intention, actually. We have a guy in house who’s really into analog modeling and he’s a guitar player, and so he had some of the initial ideas about that.

Then I did a trip to LA the year after [Amaury] did, and met a bunch of people. We were just seeing people using all of this non-linearity and distortion and stuff, all throughout their tracks, not just for guitar, but for vocals or drums or keyboards, synths, all this kind of stuff.

So it wasn’t really about thinking, “Oh, we need to go after guitar players.” We have a bunch of Live users who are guitar players. We would go to these studios and there’d be a couple of keyboards and a computer, and a pile of stomp boxes and stuff. So it was like, “Well, that’s a thing that people like to do. How can we make it easy to do that?”

So it was actually more focused around that kind of thing.

Amaury Groc: And we made sure to tune a pedal to sound good with synths and [other things], so it’s not only for guitars … the range is larger than that.

Simon Hayes: I think Dylan made an interesting point, though, saying that there was a person in-house who was really into those things.

A lot of the new features that we’ve built have come from a particular person who is really driven to push forward an idea. We build what are called hack sprints into our working schedule, which is basically time for developers to explore, completely freely, new ideas that are outside of our regular work plan. There’s been a lot of really great stuff and new features that have come from this, as well. Stuff that’s not even necessarily what users have asked for, but just a completely new idea that someone has just invented.

Synthtopia: When you have a hack sprint, how long do you have to just go wild?

Simon Hayes: It depends. In the last while, as we got closer to the release deadlines, we really had to get stuff ready, we shortened the hack sprints to one week. It was two weeks before that.

Synthtopia: How often do these hack sprints fit into your cycle?

Dylan Wood: We do it four, maybe five times a year, on and off.

But we usually try and build in a bit of time for that kind of work, as well. We do hacking and prototyping, [in the process of] the work we’re already doing.

Simon Hayes: For example, the new Push step sequencer that we did, we did a new melodic step sequencer, that was born in one of these hack sprints.

One of the developers just tried it and passed it around, and we all had a go of it and thought, “Wow, this thing is amazing. We should take it and develop it further and get one of the designers to work with them,” and we ended up building this feature.

Synthtopia:  So you have a sort of Agile development approach [a popular approach to software development], where you work on specific things for several two-week sprints, and then you have an open sprint where you do whatever?

Dylan Wood: Something like that, yeah. We do run sprints every two weeks. I think it’s every six sprints we do the hacks. It depends. The timeline is a little bit different for different things.

Agile is about not defining a solution about what the problem you’re trying to solve is….

For example, for Echo, the challenge was not, “Make it exactly like this.” The challenge was, “Make something that lets people make the modern and classic delay sounds they want, that they find really inspiring to use, instead of hardware.” And then people go off and work out what that means.

It makes our job better.

I think in traditional product management, it’s a little bit like, “You need to make it like this and it’s got these knobs and it’s gotta have four of those, and then we can ship it.” That’s not how it rolls here at all.

Simon Hayes: Yeah, I think we’re more on a vision level, defining exactly these kind of things. It should be able to do this, but how is up to the designers, developers.

Amaury Groc: And where there’s two week cycles or where there’s working, software working, steps that we can all take together and see if it’s the right thing.

Synthtopia: What happens in the lead up to the Live 10 release? Beta testing primarily?

Dylan Wood: Yeah. The way that our public beta testing works is that people sign up for it and they get put in a queue, and then we add people in, in batches. As the feedback drops off, we’ll let another group of people in.

Stability is really important to us, so we would rather take a bit longer in beta and fix all the problems than rush it out. So as Amaury said, we’ll ship it when it’s ready.

We’re really interested in making sure that it gets in the hands of a lot of people, but you need a bit of time, actually, with something as complicated as Live. It’s got so many corners and areas of functionality and things, [different] computers and configurations and things that can work with it.

You just need a lot of time, with a lot of people, to understand and make sure that you’ve really covered all the bases. People internally have been using it to make music and released albums for a year, at least. But at the same time, there’s something that you only learn when it’s [in the hands of] a lot of people, on a lot of different systems.

Amaury Groc: Yeah, it’s seeing the different configurations that may not work and the work flows.

There are so many details of how Live works in real work flows. It’s good to have people use it and then, when there’s a detail breaking for someone, you need to know it.

It’s hard to foresee all of that.

Simon Hayes: I think as Dylan said, stability is hugely important for us. The stakes are quite high.

You’ve got people performing with our software in front of thousands and thousands of people, and you don’t want that to crash.

Amaury Groc: Or to malfunction, simply.

Ableton Live 10 & Push 2

Dylan Wood: The other challenge is that people use it so many different ways. We’ve got all these features that were probably built with one thing in mind and then we find some of the stuff people are doing with it. It’s just like they push it in all sorts of directions, as well.

So a seemingly innocent change can have a huge impact on these people that have invested all this time in learning to do it a certain way.

Amaury Groc: Live is built in a quite flexible way. Everything we do, we try to open, so less concept and more workflows to approach it.

And we see people using stuff in a way that we never imagined.

Synthtopia: Does the fact that Live has gotten bigger and can do more, does that end up slowing things down when you come to doing a release?

Amaury Groc: Anything you want to add has a potential to bloat the whole thing, to make it even more complex to approach.

So we spend a fair amount of time thinking hard how to integrate and to reuse existing concepts, instead of just adding new concepts. That’s a thing that people can learn concepts to a certain limit, and if you impose learning more and more and more and more different things, it’s harder to use.

It’s not like it’s simple. There’s a steep curve of learning to use Live, but we observed some people using only a tiny part of Live for the music-making who do excellent music, without understanding the whole.

For devices, I guess it’s a bit different, but that’s a point to say. You don’t want to add like 300 effects. So, it happens in the past that we had two compressors. [For Live 10, w]e made it one.

Or trying to simplify on the way. There are things we remove, actually. It’s not only about adding, it’s about removing mindfully, or replacing.

Synthtopia: How do you decide what to remove, and do it in way that won’t break things for people?

Dylan Wood: Essentially, because people’s investment into Live is the documents and stuff they’ve created, there’s some things we can’t remove from it, because they have to be there or people’s stuff won’t work.

When we’re doing device development, it would be way easier if we didn’t care about that, actually. If we make a tiny change in the device, we usually offer a way to go. [For example,] the old one [might] still [be]in the code and can be used if you load it up.

You can choose to upgrade it to the new one. This kind of stuff.

That’s not the developers ‘most-favoritest’ of jobs. It’s kind of painful. But it means you can load a set from a really old version of Live and it’s still gonna sound the same.

Synthtopia: Is Live backwards-compatible really far back?

Amaury Groc: Yeah, a Live 1 set, I think, opens in Live 10.

Synthtopia: Oh, my gosh.

Amaury Groc: Yeah, that’s important to us.

Live 1, we didn’t check for a long time. I don’t know if anyone has a live set from Live 1, but …. it’s usual to see people … still on Live 6, maybe. Someone has it set for their laptop for on-stage and they don’t touch it, which I understand completely. If, someday, they want to upgrade or to change the system, it should be possible.

We don’t remove a lot. But, still, on the way, we try to clean up. It’s really tiny stuff on Workflow, but things we removed in Live 10, some preferences for Rewire, things we know people won’t really use, but we’re careful to make space, somehow, if possible.

We don’t remove much. We can’t.

Synthtopia: Because somebody uses it.

Amaury Groc: Yeah. You think “nobody’s using that,” and maybe it’s only 100 people who do, but for them it’s really important.

So we don’t want to [take things out], unless it’s really in the way of something else. Even then, often we try to propose a replacement workflow or something.

Synthtopia: How do you add all these new features and new instruments and devices without making the software just huge and slow and clunky?

Dylan Wood: The engineering practices here are pretty strict, so they’re really careful, is probably the best way to describe it.

Amaury Groc: We also work continuously on performance optimization.

Synthtopia: Does the progress of technology reduce the need for that? If you look at the computers available, five years ago, they probably wouldn’t have had SSDs and the RAM computers do now. Does that just free you up a little bit?

Amaury Groc: To some extent, but not so much, because we still support quite old computers.

The minimum requirements went up, that’s for sure, so that gives some relief. We don’t have to support mono core processors anymore, or Pentium 4, but I think up to Live version 9.5, we supported Pentium 4 officially.

Synthtopia: Really? Do you have to keep an old tower computer in the back room to try with the new version?

Amaury Groc: It’s also that….some newer computers have actually less computing power, for more battery life. That’s a trend you’ll see in maybe inexpensive, very portable Windows machines.

So we want Live to run on these as well.

Dylan Wood: Yeah. There’s an effort in efficiency.

If you’re inefficient in the way you engineer and program something, it doesn’t matter how fast the computer is, it’ll blow up, at some point.

So… even if computing speed’s not such a bottleneck, you still just have to be careful to make it efficient. So a lot of work goes into that.

Amaury Groc: And the lighter, the better, on CPU and the RAM. Because it’s faster to load and it’s more enjoyable….than waiting.


Live 10 Resources:

18 thoughts on “Creating Ableton Live 10 – The Story Behind The Update

      1. Glad you liked it! There are three more Ableton interviews coming that we hope you’ll find as interesting as this intro that we’ll be publishing throughout the rest of this week. Gotta get back to editing mode now ….

    1. As much as I like Live and my Push 2 the omission of MPE at this late stage in the game is just bullshit. It’s either a marketing strategy to hold off and introduce it with Push 3 or as others have suggested, Live MIDI would have to entirely reworked from the ground up in order to make it happen. Very disappointing

  1. Yep, great interview with a bunch of good people! Nice to get some insight behind the process, though nothing earth-shattering from a development perspective. Interesting that backwards compatibility is so important to them. I bought the upgrade this morning and can’t wait to play around with it later.

  2. Cool Interview, best DAW there is and Push is amazing to use.

    I like to be able to name inputs/outputs now but i don‘t get why they didn‘t enable this for Midi as well. So now i still have to memorize whats what on my midiinterface, bummer! 🙂

Leave a Reply