english /* english */
A conversation with the Apertus Open Source Cinema team as they take on the might of the film industry.
Industries, as they go, don’t get much more closed than cinema:
Ultra-hierarchical teams working in the service of some of the world’s largest corporations, using expensive, closed equipment to capture proprietary formats, edited on more closed equipment using proprietary software and finally delivered in a locked-down format under a business model entirely based on maintaining a copyright monopoly.
So as an open-source-obsessed videographer with plenty of experience working in this very closed environment, it is a great relief and also hugely exciting to be joining what I hope will be somewhat of a revolution in the field. I’ve put my name (and my money) down to be one of the first to own an open source hardware professional cinema camera, the Apertus Axiom.
The project is hardly a get-rich-quick scheme for its founders, as they’re selling this first beta batch of cameras through a crowdfunding campaign at cost price (50% off retail).
The idea of going through the entire product development and running a stressful crowdfunding campaign without making a cent of profit might sound a bit backward, but I suppose allowing anyone to modify, improve and sell their design could be seen as crazy too.
But the strength of any open source project is its community – and while software projects are able to pick up users and developers at almost zero cost, growing a community where the barrier to entry includes buying a €6000 camera is a different matter.
In this context, it makes sense to focus the first batch entirely on getting a critical mass of users, testers and bug fixers around the project. This community can then be built upon in the coming years with new batches, new models, modules and improvements.
A few months ago I met Oscar Spierenberg, Herbert Poetzl, Sebastian Pichelhofer and their working prototype in Berlin to talk about the project, its development and their open source approach.
SM: Can you introduce yourselves and tell me a little about the Apertus project?
Sebastian: Oscar founded Apertus more than 8 years ago, hacking and playing around with the Elphel, which was the first open hardware / free software camera module with a high resolution image sensor.
Oscar: There were a lot of people coming and going, and I was just working with tools I could find rather than inventing them myself – tweaking them, and finding people who could deal with things that I couldn’t.
With that kind of project, a lot of contributors come and go.
Then after two years Sebastian came along, and he stayed forever… he promised, “I’m exactly the opposite of the other people – once I start, I never stop a project.”
And he’s kept his promise – he was the first one who really pushed the project forward, and still is today.
Sebastian: We found some problems with the Elphel camera which were just not easy to overcome with software or basic modifications, so we decided to start designing our own camera from scratch, expressly designed for filmmaking.
The Elphel camera is a general purpose tool. It’s good for a lot of things but not 100% perfect for all of those, and so for us the sensor size was always an issue. For scientific applications, for example, a small sensor with high resolution is preferred. A shallow depth of field, for scientists, is an area that has no value for them – because if something’s blurry, there’s no information in it – so only the crazy filmmakers want a big sensor.
So now we have the Axiom prototype with a big sensor – the standard format of film making, just like a film camera, super 35mm .
SM: Had either of you had experience building cameras before this project?
Sebastian: Not at all!
Oscar: No – only with tweaking everything I could find, and making a 35mm adaptor for a video camera…
Sebastian: but it’s been 8 years now, so we’ve gained a lot of experience, now we know a few things, and we’ve learned from the Elphel architecture.
Oscar: We also gained a lot of people working with us, like Herbert Poetzl, he joined a year ago, so we have quite an extensive community of people – still a lot of people coming and going but the core team is very strong, and a very nice group of people also.
Sebastian: It’s of course always difficult for a hardware project to grow because you always need people that have the hardware and that means they have to buy it, or they have to build it, so it’s a big barrier to entry.
But now that the prototype is finished, and not too complex to replicate, I think we will be growing even faster soon – also with the crowdfunding campaign, that will be a big breakthrough.
So far the most successful open source project regarding filmmaking would have to be Magic Lantern, the unofficial open source firmware layer available for Canon DSLR cameras. The Magic Lantern team built on the success of the Canon Hackers Development Kit and reverse engineered the electronics of the 5DmkII, a professional photo camera with prosumer video capabilities. Through some clever hacking and hardware/software engineering, they gave filmmakers the ability to monitor audio, create timelapses, shoot raw video, magnify focus during filming, show exposure warnings, shoot HDR video…. and have now extended these capabilities to the entire line of Canon DSLRs. Magic Lantern has massively improved what I can do with my 5DmkII and meant that I can still use it professionally, 6 years after buying it.
SM:Has Magic Lantern been an influence on you, or is there crossover between the projects?
Sebastian: Magic Lantern is an interesting example because for filmmakers who have no idea what open source really means, Magic Lantern is like a keyword that makes it really clear to them – “Ah, that’s the thing which enables you to do all those things you couldn’t do before!” and the number of times we have been asked “oh, you’re building a new camera, will it support Magic Lantern?”
And of course we were kind of confused, because people obviously didn’t understand what Magic Lantern was [an open hack for closed hardware – there's no need to reverse engineer an open hardware camera], but after some thinking it’s actually quite a compliment because they kind of understand the openness now, and they just mean “ah, Magic Lantern means openness, so that’s what you’re doing as well.”
[note: Magic Lantern have now teamed up with Apertus to help squeeze every last bit of information out of the Axiom's sensor and electronics]
SM: One issue I run into when trying to explain open source to people is the assumption that open source is just another word for DIY – but your project is a great example of the difference between the two. A camera is a lot more complicated than a model airplane, it’s unlikely that many filmmakers would ever consider building it themselves. So if it isn’t really a DIY camera, what else can open source offer the user?
Oscar: actually, our camera is like any other camera, you don’t have to know programming, you don’t have to build anything yourself, but you CAN if you want. You CAN, that’s the big difference.
Sebastian: often people see, you know, we have printed circuit boards and source code on the website, and they get really afraid – oh no! do I have to make all this myself?
The answer is no! you can just take it out of the box, turn it on and shoot with it. but if you want, you can also dig deeper and understand what it does, modify what it does, and even if you’re not a programmer, and have no idea what’s going on in the camera, just like with Magic Lantern, there will be other people who do so, and you benefit from their developments.
Oscar: and there’s also the idea that the camera you buy will continue to evolve after you’ve bought it, and that’s only due to open source – another camera you buy it, and that’s it. If some better part is implemented in the next model, you have to sell your camera and buy a new one and that’s completely different with our camera, so it will be upgradeable as soon as improvements are available.
Sebastian: other camera manufacturers, they also do firmware upgrades of course, but that’s just fixes or simple additions and it’s always only what the company wants to share, not what everybody wants to have, and that’s the big difference – because the community knows what it wants to have, and can create that for everybody.
SM: Another real sticking point for people in discussions about open source as a business is the idea that you can’t possibly make any money – most people are scared that the moment you open source something, you’ll be copied by your competitors, lose all your customers and waste all the time and money you put into development. What has your experience been so far, and what’s your answer to that concern?
Sebastian: We have submitted a Horizon 2020 EU grant application, and it requires that you find 2 other partners who will pursue the project with you – and for us it was very important that these partners would also open source their developments on the project.
We’ve been contacted by many companies since we first announced that – some of these companies had an idea of open source, some had never heard of it, and most were scared away when they understood what it actually meant for them!
Oscar: it’s almost always the exact same reaction when they find out what it is, but I think actually open source is our protection in a way, because you can copy the camera, but you cannot copy the whole community we represent.
Sebastian: It seems strange because the business model is new, of course – but it’s not not a business model… many people think there’s the business model, and then there’s open source, but actually open source is also a business model, it’s just different. But since it’s new it’s always something to be suspicious about.
Many businesses who already have their ongoing profits and established products, they want to keep it that way and do what they know works.
But I actually talked to another camera manufacturer recently and he said, well, our camera is copied too – it’s a closed camera, it’s a different way of copying, but it’s still copied.
With us it’s all out in the open, so it’s also a way of protecting things – we are very determined in this direction and very confident.
But also in terms of finding partners, it’s a good filter – because all the companies we talk to, those who understand it, we can be sure they are really committed to the cause, and that they really understand what we are doing.
When we come across companies that are afraid of open source, it’s really easy to go, ok, maybe not with these guys… you don’t have many ways to get to know another company, they have representatives of course, and you can look at what they are doing, look at the website, but what’s really behind it is always difficult to judge, and so that’s like, the test we put other companies through to see if they’re worth teaming up with.
Thanks for the chat guys, and best of luck with the rest of the campaign!
There’s so much more to discuss about this project and its future plans, such as the modular nature of the design, and the ability for deep integration with third-party manufacturers who are able to develop accessories, parts and modules with a full understanding of the camera and its interfaces. Then there’s the dual Apertus organisations – one a for-profit company, the other a for-benefit foundation.
At the time of writing they’re almost at their €100,000 goal, so if you or a filmmaking friend are interested in furthering the open source cinema revolution, or you want your next camera to last a decade and evolve with you, head over to their campaign, contribute, and spread the word: