On 03/04/11 15:04, Roberto Inzerillo wrote: > Good to know Jon, I'll be glad to talk about that with you. > I like your hardware approach, getting your hands "dirty" on a naked ATMega16 > should be fun too :-)
It's interesting, and results in *much* smaller code than you'd get using the arduino libraries, and I was reasonably happy with gcc anyway which reduced the learning curve. It's nice to be able to just type "make" and have the code compiled and blasted onto the chip via an AVRISP2. I sort of went about this in the opposite direction to most people - I started out with the bare AVR stuff, and got an arduino at a later stage just for hacking quick projects with, though it's currently stuck in a fairly long term project: http://leeds.hackspace.org.uk/wiki/index.php/Projects/Nav_Bot_I > I've settled down on sending the standby-mhz only cause I will send the swap > freq<-> button state too, that will swap the frequencies inside fgfs > property tree itself, without the need to send the active freq too. > Still I think it's better to make the external hardware send only the rotary > encoder's rotations from a design perspective, and not the frequencies at > all; I know that's debatable. I'm sending freq now for easy of development > only. > > I agree sending only changing values should be more desirable but serially > sent data should be sent all together, there's no way to send an input chunk > and leaving the others as-is; since my approach aims at integrating several > (not thousands but quite a few anyway!) physical input/output devices, I > think for simplicity everything should be sent at a constant frequency. And > no, I find 10hz is way too low for some other devices (maybe frequency > settings updates can tolerate a 10th of a second delay, but it's still > noticeable). > > I'm still wondering if you too have noticed that FGFS updates it's property > tree with not exact standby-mhz values when using such input devices. Can you > confirm? Did you solved this issue? > > I've also noticed you use<format>%03.3f</format> in an input chunk ... I did > try that too, I hoped that would get me a 3.3 digit format but nothing! It's > ignored. How's that at your side? Actually I wrote that with reference to the protocol documentation, because I didn't have the exact file handy - there may be slight differences between that and what I actually got working. I'll update it when I remember to grab a copy of the original file. I did get my prototype working though, so it's definitely possible to format the numbers properly. I need to sort out a git server too so other people can play with the code, currently the only access to my git repository is through ssh. Jon ------------------------------------------------------------------------------ Create and publish websites with WebMatrix Use the most popular FREE web apps or write code yourself; WebMatrix provides all the features you need to develop and publish your website. http://p.sf.net/sfu/ms-webmatrix-sf _______________________________________________ Flightgear-devel mailing list Flightgear-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/flightgear-devel
