Interview: Eben Upton
Eben Upton (the one in the blue shirt), founder of the Raspberry Pi Foundation, Royal Academy of Engineering Silver Medalist and all-round awesome dude.
A few months ago, armed with your responses from one of our podcast’s Open Ballot questions, we visited the Cambridge HQ of the Raspberry Pi Foundation to quiz its founder about hardware upgrades, education and what success has meant for the project. The result was an epic 7,000 word interview, the first half of which we published in Linux Format issue 173. But as we didn’t have enough space in the magazine, we thought we’d put the interview online in it’s entirety. And here it is!
There can be few people as busy as Eben Upton. As the person most credited with bringing the Raspberry Pi into existence, he’s been at the heart of its delivery; overseeing the minutiae of its design, its production and its incredible rise to mainstream recognition. At the same time, the Foundation has moved manufacturing to Wales, launched both the Model A and a camera module, and continues to push the Pi into newer territories.
LXF: How difficult has it been to adapt to the success of the Raspberry Pi?
EU: I think a lot of the heavy lifting in terms of the logistics and the scale has obviously been taken on by RS and Element 14 (the two manufacturers of the Raspberry Pi).
LXF: Had you always planned to use them?
EU: That was the idea, originally, because even before we launched we realised we’d probably have success beyond the scale we’d ever be able to cope with, and we had enough money to build the 10,000 - we had a quarter of a million dollars - which is the 10,000 units scale, and that was what we had. I think it was the point when we released the first SDCARD image for the device and we had 50,000 downloads, or the buggy alpha quality operating system from the team of people who didn’t know whether it was going to exist yet, that was the point where we realised we were probably in trouble (laughs), and that we needed to think again about our model.
LXF: This is at the end of 2011?
EU: That was the end of 2011. So if we’d stuck with our original model I think we we would have spent the whole of last year dealing with the first day’s demand. While it felt like we spent the whole of last year dealing with the first day’s demand, we actually only spent 3-4 months dealing with the first day’s demands. We certainly wouldn’t have got anywhere - we might have got to 100,000 in our first year if we were really lucky, instead of being able to get to a million scale. So from Raspberry Pi’s point of view, I think we ‘magicked’ away quite a lot of the scale by handing over to these organisations who were all about scale. I mean, those are just around the area of Amazon’s scale of logistics operations. RS say they ship 44,000 packages a day. So although this is significant volume for them, it’s not anywhere near doubling their volume.
LXF: Yet RS and Element 14 seemed surprised initially.
EU: I think they were more surprised than we were.
LXF: The launch was an event!
EU: Yes, absolutely. I think we’d been living it for maybe 9 months and they had only been living it for a month or two.
LXF: Liz (head of communications for the Raspberry Pi Foundation and Eben’s wife) also did a good job of building the anticipation.
EU: Absolutely, so I think it was a surprise for them. I think it turned into a very good surprise for them - it has become a good bit of business for both those companies and you can see they’ve both been growing. I think there are scaling things for us. It’s changed our view of what it is we’re trying to do, and that’s been the challenge for us.
LXF: And what about on a personal level?
EU: Certainly there is a sense - I wouldn’t say it’s less fun, but it’s more serious. Because there are people whose jobs depend on this. It was never the idea for there to be people whose jobs depended on this, and not people here but people who make them and the people who distribute them. There are significant numbers of people - probably a 100 people - who owe their livelihood to the Raspberry Pi at the moment. So that does focus the mind a little but when you think ‘Ooh, let’s change this component on the PCB’. If we were building 100 of them a week and we accidentally crashed a week’s production, then that’s a 100 units we’d have to scrap. At the volume’s we’re running at now, of course, a week’s production is 30, 40, 50 thousand units, depending what week, and you could damage the business and then people would lose their jobs. So it’s not so much that that’s not really fun, but it does focus the mind a little bit.
LXF: It’s interesting that you envisaged the Raspberry Pi to be both a foundation and a business, because that is a very specific feature you don’t see often enough.
EU: I think it’s really important that if you wanted to be sustainable, that’s the thing, I think this idea that charity doesn’t scale - we wanted to be a charity, we are a not-for-profit. But the money that gets made gets ploughed back in. I don’t take a salary - I’m lucky enough not to have to take a salary.
LXF: Would that ever change?
EU: I could at some point I suppose. I do work, basically, more than full time - more than what most people would consider full time for this organisation. But I’m still actually employed by Broadcom. Broadcom have been enormously generous. I think there’s always been this suspicious that it’s a Broadcom marketing, and I think that fuels the idea. I think this is a thing which we’ve just been lucky, and it took a while to persuade Broadcom that this was worth doing.
LXF: Is it like when Linus Torvalds worked at Transmeta?
EU: It’s becoming that kind of relationship. Possibly less involved because I understand he was quite heavily involved in contributing to Transmeta so possibly. I spend, I’d say, 90% of my time now on Raspberry Pi activities. So yeah.
Would I ever take a salary? Yes, at some point it might get to the stage where that’s not a sustainable thing for Broadcom anymore and that would probably be the right time to move across. In any case I might move across for a salary but I wouldn’t move across - it’s not like people are getting dividends out of this - so any retained profit in the organisation after we pay people just gets re-circulated back into the platform. And that’s nice because it means that our bit of it is sustainable and that we’ve got a bit of strategic depth so that we can afford to do interesting new things. We can afford to do things like finance the development of the camera. The pi was developed by Pete Lomas at Norcott (Technologies Limited) as a donation, basically, as one of the trustees of the foundation. The camera board has been developed as a commercial product, so you we paid people to design the camera board for us.
LXF: Does that affect whether the schematics for the hardware will be released?
EU: The schematics have been released but not the PCB, and that’s an interesting question. But that’s an interesting question - would we ever release the design of the PCB? The intention has always been to release the design of the PCB; it’s still to release the design of the PCB. The issue is alone is really that you can’t buy the chips. There’s actually another problem we need to solve.
LXF: Is that why you’re not releasing the design of the PCB?
EU: If we release the design of the PCB, the reason why you can’t build a Raspberry Pi is because you can’t buy the chips. And a certain amount of... opprobrium... there were a small number of people who, I think, that were very offended that we haven’t given away all of our IP. And right now, that opprobrium is heaped on us. Now if we then were to give that away, I know where the opprobrium would end up. And I believe that Broadcom have been enormously supportive of this and I don’t believe they deserve that opprobrium heaped on them. So for now I’m happy to sit there and absorb those brickbats.
I don’t have a great ideological commitment to giving away my IP, given that the money that’s generated by our ownership of that IP goes to make Pi sustainable and increase the probability. One really important thing about Pi is that it’s likely to be around in 5 years time. That’s a big selling point. That you can build an industrial product on Pi, or you can design a curriculum around Pi.
LXF: And you don’t want that diluted by third parties building their own....
EU: Absolutely. And I think to the extent that if we gave away the design and we enabled a lot of clone manufacturers, and we undermined the financial basis of what we’re doing here, that would be all well and good. But it’s not like we’re scalping 10 bucks out of every Pi, right? We’re making little enough off every Pi and I think we’ve reduced the cost of the Pi to everyone else, so I think the profits that would just go into other people’s pockets would be extracted in dividends, would be extracted in profits, by companies. All that would happen would be that we’d end up destabilising the foundation; the foundation’s ability to invest in education, the foundation’s ability to invest in funding open source projects for no actually tangible consumer benefit. There’d be a kind of ideological benefit, because we’d be able to tick a box that says, “open!!”, and that would be it. And then there would be this problem that I think opprobrium would come to be heaped on Broadcom if Broadcom didn’t then make the chips available.
LXF: When is there going to be an upgrade to the Pi? Have you considered making the Pi modular?
EU: We have no room for additional RAM on the device at the moment. We’ve established that that’s as much RAM as we’re ever going to have - we have no ability to expand the RAM because that’s all the SoC (system on a chip) can talk to. We have no ability to increase the amount of processing power because that’s the amount of processing power the SoC has got.
We don’t really have a hardware roadmap.
LXF: In the back of your mind, you haven’t got Raspberry Pi 2?
EU: I think it would be really sad, and probably fatal for us, if we were still shipping the same Raspberry Pi in 2016, say. I think we’ll have to do something but I don’t know what that something is. I think it’s actually very difficult to specify what that something would be. The real problem is that I can imagine boards I could build at any price between $25 and $85. I can imagine a different board that I could build every $10 as you go up. I can imagine a different board that I could build at each $10 increment. None of which are currently being built. But finding one that’s actually attractive, that’s got the same kind of attractiveness of the Pi - Pi’s attractive because it’s got a really interesting price/performance tradeoff. It really is extremely hard to do. I certainly don’t think that at the $25-$35 price point, I can’t think of any other board I could build. With any chip that’s available to me, I can’t think of any board that I could build at say $25 or even $35 that would be as good as Pi, let alone better.
But it’s a push even at $45. $55 I could imagine that you’d start to get to the point where you can start to get better but it’s interesting that there’s nothing out there right now.
LXF: How do you then maintain the momentum of success?
EU: By doing a lot of software work.
LXF: So that’s where all the effort is now?
EU: With 700Mhz, (the Pi) is an enormously powerful media accelerator. The chip is 97% media accelerator.
LXF: And a default installation of Raspbian, what would that use?
EU: Currently, it uses the video output path, the USB controller; stuff that was there before the ARM. The ARM made the chip 3% bigger. You’re beating up mostly on the ARM, a little bit of system infrastructure, an SDRAM controller and a few tiny peripherals. A lot of the system is asleep most of the time. But in any case, the 700Mhz on ARM it’s still a powerful processor, but it is by historical standards a pretty beefy device. Because we’re kind of pinned in one place in our hardware, we’re doing more work on the software side to get all of the juice out of that system. So we’ve spent a lot of time optimising system level components, hopefully upstreaming, optimised versions of system level components for Linux - so, pixmap. Optimised versions of things like memcopy and memset.
We had an interesting debate about X acceleration. We don’t have an X accelerator, we don’t have an acceleration driver. We have a lot of components of the chip, a lot of sub-systems on the chip, which can be used to influence the X server accelerator. And it’s actually ok. Software X is ok. I’m really actually always surprised by how ok it is. It’s surprisingly good. It’s ARM moving a pixel at a time, although now we’ve done this pixmap stuff it’s an ARM going ‘pixels - bang, bang, bang.’ Insofar as a fairly low performance ARM can move pixels fast, we do now move pixels fast.
But we’ve had this debate about should we do hardware X acceleration. And I think we’ve come down the side that we shouldn’t. What we should do is move to the kind of brave new world.
LXF: Is that because it’s easier?
EU: I think it is easier. I think we can afford to do one or the other, is the other. We can afford to do one of the two, and given the choice I’d rather spend my money on the brave new world - given that that’s where we want to go. People want to go to the brave new world so I think we’re going to try and go there quite quickly. We’ve done some work with a company here in Cambridge called Collabora. So, Pek (Pekka Paalanen), who works for them and is a big Wayland developer, has been doing a back-end for Weston, the reference Wayland compositor, which drives our hardware composition engine so we have a very powerful, what we call HVS, video scaler in the device which is effectively a big hardware sprite engine. It just gives you a vast number of big hardware sprites and what you do is you put a window in each hardware sprite and then you ‘CLAPS’ bang them together onto the display. You can stack up a lot of these before you run out of memory bandwidth. At the point where you run out of memory bandwidth, there’s a fallback which does offscreen composition. It’s really cool. And that gives you pure 60hz window drag. In a world where every window is a hardware sprite, dragging a window doesn’t move any pixels.
Given we’ve got this hardware video scaler, which is probably about the same size of the ARM - bigger than the ARM, because the HVS has got a lot of buffer memory to control the sprites. Given we’ve got this on there which is dedicated to building composited render hierarchies and we’ve got a compositing window manager. We should do the software work to really join those together. So it’s a really good example of somewhere we can put software in and we can get a much improved user experience without minting a new chip. And it also gives us access to all that lovely Wayland.
LXF: This sounds like first high-profile use of Wayland we’ve heard of.
EU: I believe we’re the first upstreamed non-GL backend for Wayland. Wayland is typically on top of GL, but GL is actually a really lousy thing to do composition on top of because the scaling filters in GL are really primitive. Even with downscaling you have to generate a bunch of mipmaps in order to do that, whereas what ours does is, suppose I scale something down by a factor of 3:1, the problem with scaling by 3:1 is that it’s part-way between 2:1 and 4:1. If you do it with GL, you end up with a 2:1 mipmap and then you may get some noise, or you use the 4:1 mipmap and then you get blender. Or you try ‘linearing’ them together and then you get something that is a little bit better but still not great. What ours does is if you do 3:1 it will build each pixel by averaging together a 3x3 grid of pixels from the original image, every frame, for you for free.
LXF: That sounds like it would be massive for all the other SoCs out there?
EU: I think it’s really important. The thing is, most SoCs have a piece of hardware which is a bit like this. I know of OMAP, for example, has a piece of hardware very much like this. I think that, given that what people often want to do is that they want to do GL and composition, if you use GL to do composition, a) you get low-quality composition, b) you take up a bunch of your limited, finite, GL resource just stitching the display together. If you’ve got a piece of hardware that does stitching efficiently and at high-quality and a GL pipeline, you wouldn’t do GL on a window - use GL to render what goes in the window and composition pipelined together to put it together on the display. So this is why we’ve been putting money in there.
LXF: Is there a timescale for results?
EU: We did some stuff at the end of last year, and there’s stuff upstream, so if you build top-of-tree Weston for Pi you can get a backend which is a kind of mixed-mode GL and hardware video scaler backend. What they’re just finishing up at the moment is the pure HVS backend which supports fewer transformations, so you have a little less flexibility in terms of how you draw windows - you can’t do that Doom thing that people do where they have corridors of windows.
LXF: Can you still do wobbly windows though?
EU: You can still do a lot of cool stuff. You can still do semi-transparent windows, which is kind of what people want - non-rectangular, semi-transparent windows, composited together very fast, being dragged around. That’s what people really want much more than they want perspective windows. The subset people care about it does extremely well. They’re just finishing it up, so by the time this interview is published, I would be staggered if this is upstream (Pekka Paalanen stated in his blog that the patches were merged upstream on 23rd May 2013).
We're not sure whether it was to keep visitors from outstaying their welcome, or to keep brains keen, but the Pi's meeting room had no chairs!
LXF: Is the Foundation making any progress in changing the ITC curriculum here in the UK?
EU: We got very distracted last year. And you asked (previously what were the challenges of scale? One of the challenges was finding enough time to think about the Foundation’s education ambition while we were in the middle of this blur around selling and supplies. I think that progress is being made. I don’t think we are making a lot of that progress.
LXF: Where are the channels to make that kind of change?
EU: What’s really good is that there are a lot of other organisations that are working well with this. The exams boards, and particularly the Cambridge assessment, OCR.
LXF: Is that because they’re close to you?
EU: No, I think it’s just because they’ve got the best GCSE already, and they foresee, I think, having a rigorous computing GCSE as a differentiator them, and they’re going to defend that bit of turf.
LXF: But making an open platform the core of their curriculum is such an important step.
EU: Yeah. We don’t want to mandate the Raspberry Pi into the curriculum. To the extent that we have been engaging, though, has been about is trying to avoid artificial dependencies on closed operating systems or on ridiculous amounts of computing power.
There’s no reason why you should require a 3Ghz core i7 and Windows 8 to learn to program computers, to learn to do computing. You should be able to do that on Linux and you should be able to do it on an old PC, or you should be able to do it on an ARM-based machine. You should be able to do it on a low performance platform. To some extent, we’ve restricted ourselves to this purely tactical stuff which keeps the door open to devices - to Pi and a million other devices like the Pi - and old PCs. Old PCs are still a great platform to learn to program on, right?
LXF: It seems schools pay a lot for the hardware plus support, and as a result, they treat their PCs more like holy relics.
EU: People might say, “Well everyone’s got a PC, right? Why do you need a Raspberry Pi.” Well, because it’s cheap enough not to be a holy relic, it’s cheap enough that you can afford to break it and then go and buy another one, and cheap enough for everyone to have it. As the fall back, you can say for the first time with absolute certainty, “Yes, we can provide everyone with a computer,” and you just need to think about it from a budgetary point of view.
LXF: But what can you say to the teachers who want to buy iPads, for example, because of their lack of complexity and the requirement for a support contract - isn’t that a huge hurdle?
EU: It is an enormous hurdle. I think the good thing is that there’s a lot of evidence that the ICT teachers and other teachers at school are keen to deliver some sort of computing - they’re professionals, you know, and they’ve done a good job. There is a massive motivation, I think, to try and learn the stuff you need to do a professional job. And we don’t help! Guys like Clive (Eben points in the office over his shoulder), we hired Clive out of teaching ICT in Ipswich. The temptation has always been whenever you meet a good ICT teacher to hire them! And there aren’t so many, there are a lot of good one, but there aren’t so many that we really oughtn’t be going around snaffling the best ones we meet.
LXF: Is there anything as a community we can do?
EU: I think it’s extremely important for technically literate people to get involved, volunteering, because, there’s is a view that the solution to this is we create a world where we train up all our ICT teachers to be professional quality programmers. But that’s just never going to happen. It’s a lovely idea, but it’s not going to happen. We’re always going to have ICT teachers who are not specialist engineers - apart from anything else, if they were specialist engineers we’d have to pay them like specialist engineers, right? I’m not sure there’s the political appetite to pay ICT teachers like senior software engineers! They’re very different salary brackets. So I don’t think we’re ever going to do that. We’re going to have to learn to live in a world, and thrive in a world, where those teachers are education specialists, but they’re not tech specialists, and one way we can do that is that technical people can volunteer to come and help - after school clubs. And it’s great here in Cambridge because ARM ( based in Cambridge) have an amazing outreach programme. You sometimes get that feeling that every school in Cambridgeshire is besieged by ARM engineers who want to go out there and run after school clubs! Elsewhere, a lot of other places in the country, if you’re here or London or Reading/M4 corridor, it’s kind of fine in one or two regional areas, you’ll find that there are a lot of engineers about, but outside of that there are whole tracts of the country where they just don’t have access to that pool. So making yourself available to organisations like STEMNET (http://www.stemnet.org.uk) to volunteer makes a lot of sense.
LXF: How should international users tackle shortcomings in their ICT curriculum?
EU: One thing we’re doing is going after the kids rather than their education, so we’re going around the side. I think we’re lucky in this country because we have guys like Ian Livingston, Alex Hope, popularizing the idea that we’ve got a problem. There’s actually much more awareness here than there is in most other countries that there is a problem. And therefore I think that the curriculum will actually get fixed, and really actually, if you read the draft curriculum it is fixed. I think it’s out for consultation, but you should go and read the draft curriculum. It has this amazing language in like, “Key stage 2 children should be able to understand the concept of a variable, a sequence, conditionals”. It has this wonderful academic language about computation and abstraction and things being concrete and things being abstract and you’re data structures and all the notation. And actually, it’s already fixed. All we have to do - and bizarrely there’s a lobby for old style ICT - who knew? There’s actually a lobby.
LXF: You mean there really is a lobby?
EU: There are people lobbying for... I think what it is, it’s large employers who want drones, for want of a better word. There is a lobby where people want that Powerpoint education because it’s what you need if you want to have, basically, unimaginative drones that sit in an office and hit keys all day. I think it’s fixed now (the curriculum) and all we have to do is if we can hold the line for a year and stop that lobby from changing the draft, all we’ve got to do is dig our heels in and we win.
But what we’re doing is going around the side. While that’s happening and in case it fails, at least until a few weeks ago when (Michael) Gove put computing in the EBacc - had it not been put into the EBacc we would have lost. That was a great opportunity to snatch defeat from jaws of victory, we would have lost. And there was still a good chance we were going to lose completely. So we went around the side and we’re like let’s just go and appeal to children, because that’s how it was in the 80s. I didn’t have computing lessons at school. I learnt to program on a BBC micro, and that was our golden age, right! And that can apply anywhere.
LXF: How much effort would be required to build a Pi laptop?
EU: Well people do build Pi laptops because you can use those Motorola (the Motorola Atrix 4G) laptops. I first saw one of those on the day we went to see Rory at the BBC (Cellan-Jones). He had two things that day; he had the Pi and he had this Motorola Atrix thing, and it was really really cool and of course it didn’t fly in the end but that was the day I first saw that was the day that everything changed for us so obviously they had a large stock of them. I think they’ve become basically worthless and now they’re value devices from the fact you can use them to build a Pi laptop. It would be lovely to do something more sophisticated than that.
LXF: Do you think you’ll ever sell the Pi in a case?
EU: I think we’re likely to sell a Pi in a case. Currently we’re a great platform if you are technically capable and you want to teach your child to program; we’re a great platform for that. If you’re a child who has no technical adult anywhere nearby, we’re not a really great platform. I wouldn’t encourage a parent, this month, who doesn’t know anything about computers, knows nobody who knows anything about computers, to buy one of these in the expectation it will help your child to program. I think you should wait six months while we get our ducks in the row. One of the ways we’re going to get our ducks in the row is we’ll put it in a nice box, with a very nice set of peripherals, a nice manual and a really nice software load-out, that really helps a child to learn. One thing I think we’d love to be able to do is I think a system which boots into a programming language.
LXF: That was one of our questions - why not boot into Python?
EU: We were thinking of booting into Python or booting into Scratch. For younger kids, boot into Scratch. Have an environment where it’s Linux underneath, boots into Scratch and hold down a key at a particular point during boot and it doesn’t boot into Scratch it just drops into the prompt. So you can play with Scratch for six months, once you’re happy with Scratch you turn over the page and ‘Hold down F1 during boot’, and it’s like ‘Oh look - it’s a PC!’. So I think that’s something we’d really like to do.
LXF: Scratch is really the 21st century equivalent to BASIC isn’t it?
EU: It really is, and you see the look on a kid’s face when they get a little cat moving across a screen, “I made a cat move! I made a cat move!” and it’s brilliant.
LXF: Is choosing a favourite project like choosing a favourite child?
EU: It’s the space balloon one (Eben then takes us into the office to show us the Dave Akerman’s Pi which he sent 30km up into the atmosphere).
LXF: What do you think are the biggest areas for growth?
EU: We’d like to do some stuff in the developing world - the BRICs. We’re reasonably strong in Russia already, so I’d like to do something in the BRICs, it would be nice to get into all of those markets. I think just more awareness of the developing world of what you can do with the Pi. One of the real surprises for us that people are using them as media centres. Actually end-users who are buying them to use as a product, like not as development tools, and so that’s been a real surprise and I think we’ve made more people aware that there is a platform you can use to build a really good IPTV box for $25, because actually the Model A makes a really good media centre of you want to have just one wireless dongle in the 1 (USB port), you only need 256MB RAM to run XBMC, one wireless dongle in the USB port, use an HDMI remote and it makes a great media centre for only $25. A lot of people here give them to their kids. So I think just making people more aware that these things are useful.
LXF: Did you envisage this at the beginning?
EU: No, we had no idea. We were thinking about kids. We were thinking about our really narrow... I mean, with hindsight it sounds ridiculous but we thought there’d be some moderate hobbyist interest, and we thought that we’d be making a few thousand of these to give to kids who we thought might come and study at Cambridge. That was the scale of the ambition. Pitiful really. But I think it’s better to be surprised in that direction.
We still have this thing - like when we launched the Model A this year; we didn’t build a lot of Model As when we launched Model A this year because the first time we think ‘This is going to be massive’ and we build half a million of them, that will be the point where it isn’t massive and those sit in a warehouse and bankrupt us all.
LXF: Are Model As doing alright?
EU: Model As are doing really well, yeah. It went on sale in America earlier on this week (March 2013), and it sold out - bang. There was very little inventory in the channel at that point but there’s more trickling in. We’ve sold through around 20,000 on the back of very limited availability; just European availability and this little bit of American availability, and they’re coming out of Wales - those are all built in Wales.
LXF: Have you got to build a Model A instead of a Model B?
EU: Last year, the reason we weren’t building Model A’s was simply because we’d have deleted Model Bs. Towards the end of last year, what we did was we got a new batch of component orders specifically for Model As - a new order with Broadcom for a chips, a new RAM order. Actually, what we did with the RAM interestingly for Model As was, remember we moved from 256 to 512 in Model B, what we did was we sped up the tail end of that move, the Foundation actually ate the last 50,000. We bought the last 50,000 RAMs off our distribution partners, because they procure the RAMs, we pulled those out of the chain in order that 512s could slide in and they could start building 512 size. We ended up withholding a few hundred thousand dollars of RAM. But that was fine because we had to build As. As soon as we had that RAM, that was really the spur for ‘We’ve got the RAM now, that’s one of the things that can be in shortage, let’s go and get 2845s (Broadcom’s BCM 2835 chips) , let’s go and get all the other components in the pipeline with Sony (Sony runs the factory in Wales) and commit to a build. And the last thing about the Model A; they’re all made in Wales, they’re 100% Welsh, where Model Bs are about 60-70% Welsh, but they should be 100% by July, maybe.
I’m from just up the road from where the factory is. I was born about 15-20 miles away and it’s kind of cool.
LXF: Did you just call them up one day?
EU: They got in touch with us! On launch day! I looked back in my mail and it was the 29th February - they got in touch with me on launch day and I actually answered the email, but they were bloody lucky to get through to me, I can tell you. I don’t know how they got through but it was an independent contractor who works in business development got in touch with me and and said, “We think this is feasible. How about it?” So I sent him a very very rough description of the Pi, this many components, these sorts of components, this much test time and this is the price target. And he came back and said “Yes, we think we can do that.” I sent him some more details and he said “Yes”. I sent him everything - you know, the whole board design package, and he came back and said “Yeah, come to the factory, we want to give you a quote.”
And it was then four months of spinning up the supply chain. Pete Lomas, who designed the board, had to do some work on the board to make manufacturable. If there will little inefficiencies in the production process, in China you just paper them over with more people. Because people are cheap you just throw people at the problem. Can’t do that in the West, so he had to do a certain amount of production engineering to make it fly. End of August, things started coming off the line and then, that’s it. They did around 3 weeks of building 256MB then switched over to 512MB. We did some filming there with CNN, or someone, they’d built their last 256MB Pi so all around were these big placards, because all around are these big placards saying what the line is manufacturing, all saying Raspberry Pi 4G for 4 gigabit (512MB), and we had to go around taking these things down to shoot these shots, then we went through the stills frame by frame just looking for anything that we’d missed. Because people were really eagle eyed.
The first thing they do (at the factory) is to screen print solder paste onto all of the pads. That’s the first step of the PCB population procedure; screen print the solder paste and then start dropping components on. And we printed a picture of this, and somebody posted a comment on the post saying “Yeah, nice job on pad D6!” And the guy from Sony says, “What!?” He goes and looks and pad D6 has a little corner, probably 20% of one corner of the pad is not covered in paste. He was like “This has never happened before!” People have that kind of level of attention to detail and we knew that if there was any pixels in there that betrayed - we rejected so many images on the basis they had a very blurry ‘4’ that you could just about see if you knew what you were looking for. So, yeah, they built 2G parts for about 3 weeks and then straight onto 512(MB) and has never stopped doing 35 - 40,000 a week at the moment. It’s brilliant. I know they’re much smaller, cheaper products, but it’s getting to the rate at which they used to build the Trinitron CRTs there - they used to build 10,000 a day. In terms of unit shipments, they’re actually getting up towards the unit shipments, albeit with a much smaller product, they they used to back in the 90s, so that is great.
LXF: What do you think about the Parallela low-cost supercomputer that’s been successfully backed through Kickstarter?
EU: These guys are absolutely amazing. It is amazing. I’ve met the guy, he’s either very convincing and a fantasist or he’s a stone cold genius because I think they’re ex-Analog Devices in Massachusetts and I met him and they are doing amazing stuff. A tiny team.
LXF: Do you think they’ll do it?
EU: I think they’ll do it! I would not bet against them. Had I not been too disorganised and perhaps lazy I would have subscribed to his Kickstarter. In general, possibly because we didn’t go down this route, I still take a slightly dim view of people using crowd funding platforms to build small computers because we didn’t do that. We did it the hard way. Partly the hard way out of cowardice, right? We were concerned that we would miss the price point. That we would just take a bunch of money from people and then it would turn out to be five bucks more expensive and then we’d be screwed. Because we’d have all this money and we’d have spent it all at that point and then we’d be in the shit. So we didn’t do it out of cowardice. We basically wanted to have made a bunch of them before we took anybody’s money; we wanted to have made a bunch of them at the right cost, know it’s all good, know it worked, and then we’ll take people’s money. And that was quite a principled thing. But these guys are trying to do something which is kind of what I think crowdfunding is for. I think crowdfunding, and if you read Kickstarter’s justification of why they exist, they’re whole schtick is that they’re here to fund riskier projects, and we’re not just here to fund a fairly conventional project in a way that doesn’t require you to go and raise capital, we’re here to let you do things that would be outside the range of normal. And I think this is a great example. I think this is brilliant. I love it. And to some extent that’s a weakness of the Raspberry Pi, right? We’re teaching kids to program a single core processor. We’re actually teaching kids about what computers used to be.
LXF: But this is where you’ve got to start...
EU: Yeah, this is where you’ve got to start, but this sort of thing is a great stepping stone because in the past computers used to have only one core now they have four cores and in the future they could be like the Parallela and why not start programming and get ahead of the curve? So I will buy one when they’re available and I really should have subscribed to their Kickstarter and I’m a sap for not having done so.