On 04/09/2012 04:30 PM, Wolfgang Spraul wrote:
otherwise we can improve Milkymist first, which makes more sense - see above - since Milkymist innovates in 2 areas where the NanoNote was weak: mechanical and SoC.
And here's a third area where it will innovate: user interfaces. It's weak in all current Qi/Milkymist products, and probably contributes to the drawerware effect.
As I mentioned on IRC, I'm thinking about something like the Reactable, but using near field RF technology similar to that of Wacom graphics tablets. This would make it possible to do a slim device instead of the bulky videoprojector/webcam setup of the Reactable, and embed some intelligence in the objects (and add buttons, gyros, LEDs, etc.)
After I'm done with the core of the NG (Migen based) arch (mostly getting the new memory controller to work and some dataflow synthesis support), I'll start seriously experimenting in this direction.
By the way, if you want to help and learn some FPGA design at the same time, you could port some of the existing cores (Ethernet, AC97, USB, GPIO, NOR flash write support) to Migen. It should be a tractable task for a FPGA beginner with intermediate/good computer programming experience, and a good way to jump into SoC design. Join #milkymist on Freenode and ask your questions...
Sébastien _______________________________________________ Qi Hardware Discussion List Mail to list (members only): [email protected] Subscribe or Unsubscribe: http://lists.en.qi-hardware.com/mailman/listinfo/discussion
