RE: [Playerstage-users] Sensoray

[Brian G. <ge...@ro...>]

On Thu, 29 Apr 2004, Kratochvil, Bradley Eugene wrote:

> I've been hacking away at it the past couple of days, and think I've made some good progress.  I have a much better idea of how the Player server works now.  I'd just like to say nice work to the architects out there :)
>  

Thanks; we do what we can...

> I added an aioproxy to the c++ clientlib.  It's pretty simple, but I can send you the file if you'd like to add it to the source.  I think I will also be extending the dio (and possibly the aio) interfaces to encompass output as well as the current input.  If you're interested, let me know.

That sounds good.  If you send a patch, I'll merge it into CVS.  

> I've been using the p20s driver as reference so far, and life has been going well.  I have run into a bit of a sticky situation though.  These Sensoray cards can control only 4 motors apiece.  We need 5.  So, I'm pondering how to add a second Sensoray card to the system.  The driver for the card supports it, so that's not a problem.  It's tricky to control multiple cards with a driver setup like the p2os one.  Do you think your new driver will make life any easier on this front?  If not, I'll probably have to ponder a little more.

Why is it hard to manage multiple cards with one driver?  Can't you
just multiplex the low-level device I/O (e.g., using poll()) into the
main driver thread?

And, no, I don't expect that the changes I have in mind for the driver
API will make this situation any easier to handle.

        brian.

-- 
Brian P. Gerkey                    ge...@ro...
Stanford AI Lab                    http://ai.stanford.edu/~gerkey