Thursday, September 30, 2010


Hi all,

Yes, I'm alive. I'll develop later a bit on what's happened for the last year, but first I'd like to apologize to all those disappointed by the sudden stop in A320/x760+ development. I was certainly forced by the circumstances, but that doesn't change the fact that some of you donated money and you probably feel it's not been put to good use. If you want your donation back, please email me.

Now for the blast:

  1. I have now sometime to work again on dingux.
  2. Someone from ChinaChip, from the team that developed the A320, contacted me.
  3. They want to support dingux development for the A330.
  4. They will provide hardware info on the A320.
  5. They will provide hardware info on the A330.
  6. They will provide couple of A330 machines.
  7. They already provided schematics and some source code for the x760+.
  8. They will try to provide a second x760+.


The A320 is legacy and is no longer being manufactured (plus there seem to be quite a bunch of knock-offs), so they want to focus on the A330. The A330 is based on an ARM SoC (CC1800) for which, as far as I know, there's not already a linux port (as opposed to the Ingenic SoCs)... so this might be a bit off my skills. We'll see. The x760+ is no longer produced so it might not be possible to get another one (remember, I already have one purchased with donations), which would be helpful because there are two x760+ LCD types.

Having the schematics and example source code for the x760+ should make it much easier to get dingux up and running on it.

Hate to say this, but don't hold your breath. Now I have some time and I'm very motivated (the ChinaChip support is a blast), but my personal situation might change at any time and the job that pays the bills will always be the topmost priority.

And on the personal side (feel free to stop reading here): what's happened in the last year?.

By the time I was forced to put dingux on hold my company was struggling to survive, and downsizing like mad, up to the point that the I was the entire development department. Soon I became also the administrative, customer relations, manufacturing, and installation/repair department. Really. My gmail inbox just collapsed and at a certain point I just gave up trying to reply messages. I'm really sorry if some people felt ignored.

I was working on the hardware and OS of an OMAP3530 based facial biometrics access control system which was itself a quite complex project and wasn't expected to yield some income in the short term, so future looked grim. Then, all of a sudden, two customers (the local government of two small villages) contacted us regarding an outdoor wireless mesh PA system we used to develop and sell. There was an special stimulus plan from the spanish government which would provide financing for those two installations. This was unexpected, since the local governments are literally struggling to pay wages and as a contractor, you're lucky if you get paid six months after invoicing, but the stimulus plan changed the whole thing. It was evident that this was a one-time chance to get good, immediate revenue with a nice profit margin (not really, but we would be using stock that would otherwise have zero value), so I sort of finished the biometrics thingie as fast and as good as I could and jumped on to it.

That was about april.

To make things more fun, there were two choices: either use the v1 system, which had some problems, or finish the design of the v2 system, which would work much better, be easier to maintain, and allow to use all the existing equipment stock. So, I don't fucking know why, I went with the second, and from april to june (remember, at that time I was pretty much the whole company):

  1. Designed a new 200W class D weathertight amplifier. S/PDIF over RS422 and streaming over ethernet inputs. It has two microcontrollers inside, a PIC24 and a LPC1768, and I had never worked before with the later, though had some exprience with the luminary cortex-M3 parts.
  2. Built 20 units of the above, together with another 20 units of a specially modified wireless mesh node. I mean I literally built them. Ordered all the parts, soldered each and every component on the PCBs, machined the 20 cases and assembled it all. Well, actually my wife helped a lot by soldering most of the 0603 caps and resistors.
  3. Loaded my car with all the equipment and went on to install it. Climbing on top of a crane to put it on the top of the 14m poles and on roofs. Fortunately I'm not afraid of heights.
  4. Bought the two computers for the control centers, installed the OS, control application and configured it.
In retrospective, I just don't know how I could do it all myself. One man band working 16 hours a day. I guess it was a combination of good engineering and luck, lots of luck (never ever had a first prototype errorless and straight to production). The installations were finished on time and are being used now to do all the public announcements. Now that I think about it, in between all that, I had also to do some repairs on another old installation and fix some issues in the biometrics thingie. What a madness.

I spent july doing some adjustments to the wireless mesh nets. The customers paid promptly (nice special requirement of the stimulus plan, they must pay you before they get paid by the central government, otherwise... you know... six months to one year), which means we would be able to close the year in the green.

So I needed some rest and time with my family and went on vacation for the whole month on august. The last month I've been working a bit more on the biometrics stuff and developing an asterisk based voice inbox for the PA system, which is a late special request by one of the customers, but makes a good addition to the feature list of the system (which I'm not sure matters at all because I think we're not selling any more for a long time, at least here in Spain). The outdoor IP amplifier is a good piece of engineering and there's a slight chance that could be sold to another company. We'll see.

As I said, we're on the green up to the end of 2010, but there's no income prospect for 2011. So, while now I have some time to work on dingux, and I intend to take it easy for the next months, I will be looking for a new job, new city, new country, not necessarily in that order :-)