Blog 2011
From WFFwiki
Time for a change?
Tuesday, October 18, 2011 at 11:30am
Let me start by stating a simple fact. PICs are by far the best microcontroller silicon out there. They are feature rich, there are loads to choose from, they consume very little power and require few external components to work. However, when I started with uCs the accepted method of coding was, pure and simply assembly, but times have changed…
Now, more and more, we see far more complex applications involving standards such as USB, Ethernet, FAT filing systems and the like. This means the reliance on the programming tool-chain and manufacture supplied application libraries grows with every passing day. With the advent of embedded engineering agnostic solutions such as the Arduino the uC hobbyist landscape has shifted away from the question of pure-silicon based advantages towards a model of open-source tool-chains providing the users with access to tools and libraries without restrictions and compilers which do not require payment of big bucks just for features like code-size optimization.
5 years ago I had never heard of Atmel or the AVR however they did something smart which Microchip is still failing to understand. Chip foundries make money from selling chips, not software. Atmel embraced the open-source community and in return generated a large following and code-base almost overnight in product life-cycle terms. Meanwhile Microchip has stubbornly clung to a close-source approach made even worse by a proliferation of compilers (2 of which they own themselves). Their application libraries are ‘free’ but only in a very limited sense and only work with the C18 compiler. To make matters worse you can’t alter the library and redistribute the code without breaking their license agreement.
Hobbyists may not be the big sales generator for chips however today’s hobbyist is tomorrow’s embedded engineer. It is arguably already too late for Microchip to reverse their policies; this is a moot-point however since all indications are they don’t care and don’t want to (take a look at their Arduino compatible PIC32MX board for a good example – a complete waste of time from start to finish).
Now I won’t go as so far as to say I will never use a PIC again, but for all future projects I will be focusing on the AVR uC. The vendor lock in caused by the collection of debugging, programming and analytical tools only reaches so far. If Microchip grows a brain in the near future I will consider moving back but why wait for what you can already have?
P.S. MPLABX now for Linux! Wait… Netbeans?!? Seriously, I really want some of what they are drinking at Microchip.
Autumn begins
Monday, September 12, 2011 at 10:00am
Recently I've been putting quite some effort into building control systems for the autonomous sailing boat I'm helping to build. Although this has generated a lot of electronics most of it has been experimental and I've not really got it to a state where it is useful enough to be published. Hopefully that will change as the days darken here in Sweden and I get more time to focus on electronics. Yesterday I published my RC Servo Processor which is part of the prototyping effort for the The Magellan Machine. We have now built a prototype hull and sail, just the rudder to go and we can drop it in the Baltic and see what happens. I'm also working on some more musical projects around the DubReq Stylophone which I hope to publish in the near future.
Sailing on an ocean breeze...
Thursday, May 19, 2011 at 04:00pm
Over the past few months I've been involved in an ambitious project to build a autonomous sailing boat with the aim of designing a boat capable of circumnavigating the globe. The Magellan Machine is a small team of like-minded engineers based in Stockholm, Sweden which I found through my new-found Hackerspace Abbenay. Although the end-goal of the project is a long way off, the aim is to have fun whilst educating ourselves and each other on a journey towards a common goal.
Right now I am working on the first generation of boat control systems building towards the complexity using a step by step approach. I hope there will be a number of spin-offs from this work which will prove useful to others (for boats and other robotics) and, of course, I plan on sharing the outcomes of the work here on my website alongside my other projects.
Both the Magellan Machine and (more generally) Abbenay are always open to new members; so if you happen to be in the Stockholm area follow the link to Abbenay, pop by and say 'hello'. It would be a pleasure to meet you :)