Tag Archives: LAN

Update: The kits have landed, roll-out plans & Arduino vs Raspberry Pi performance

We posted another update a couple of days ago covering many of these points, but it seems to have evaporated (Maybe I forgot to hit the ‘Post’ button).

We know some of the kits have arrived locally in Ireland. However, one backer reported that the packaging was ‘compromised’  but that the components were OK. He sent us some photos & we will now update packaging on all future kit shipments. If any backer experiences a problem with the packaging, email us and we will sort it out. We plan to post the next batch of kits over the weekend or early next week.

On the Raspberry Pi front we have learned some differences compared to Arduino. For what we are doing with AnalysIR, Arduino appears superior, which was initially surprising considering the 16MHz vs 700Mhz CPU speed. But apparently, it’s because Arduino is dedicated exclusively to IO & RPi is a linux based OS doing lots of other things simultaneously. I believe RPi may perform better with kernel patches or over-clocking, but we don’t want to go there yet.

Yesterday we got the measurement of IR modulation frequency working on RPi using GPIO. This may be a world first for both Arduino & RPi. The only downside is that we cannot decode the signal & measure the modulation frequency simultaneously on the RPi, which we can on the Arduino. So for now, we will just provide a command line facility on RPi for measuring modulation Frequency. We have also written a script to take the raw output from LIRC on the RPi and convert it to AnalysIR format via LAN, which means we can also use LIRC as a real-time recording mechanism on RPi. (In theory, it should be easy to extend this to every other Linux platform with LIRC installed)
One bonus with the RPi is that we can already decode & analyse IR signals over the LAN, a feature which the Yún will also provide via WiFi. We also read the blog post on the upcoming Arduino Yún release and can’t wait to get AnalysIR running on it…… The Yún should work out of the box, via serialUSB as it also has a Leonardo on-board and we will have to write up some small script to enable the WiFi part (Oct+ time-frame for WiFi).

Regarding our planned roll-out (circa 9th Sep), we plan to stick with a 2 phase approach as outlined on the main campaign page. If there are additional slots available for the initial phase we will contact backers directly. The reason is to avoid wasting time for 100+ people with a new release.

I would like to finish off by thanking you all for supporting AnalysIR.

Update: AnalysIR network enabled, Kits, RPi, Yún

The first set of kits has shipped today and because we were able to get a discount on the components we also included some bonus components as a gift (2 x npn transistors, 4 x 100 ohm resistors & 4 x 4k7 ohm resistors). These extra components may be of use to those who want to build their own IR led Driver circuit with the TSAL6100. We didn’t realise that shipping components was such an ordeal (individual Customs & Air Safety declarations depending on destination). The shipping cost was more than budgeted, but was also offset by the discount achieved.

On the Raspberry Pi front, AnalysIR is now running well without any noticeable glitches. The approach we have adapted is to connect to the RPi over the network. The solution was to pipe the output over the LAN to a virtual serial port on the Windows PC. The virtual serial port utility is freely available for Windows and the network stuff on the RPi is just standard shell commands. I should point out that we haven’t yet tried to measure the modulation frequency on the RPi, just straight decoding though the IR Receiver. When idle we are using only 1.3% CPU on the RPi, when recording a single signal it goes to 1.6% CPU & when full out recording IR it goes up to ~ 4.9% CPU usage. This is pretty good as it doesn’t impact on other processes; considering our first attempt was using 99% CPU when idle 🙂

The potential to decode & analyze over a network may lead to some interesting features in future – ideas welcome. As a minimum, the work with the RPi will make the port to the Arduino Due easier (we are expecting a Due next week) and also when the Arduino Yún is released with WiFi