<kristianpaul>
when i do a signed short int read from a memory maped devices, the cpu is supposed to wide/narrow the data bus to fit my requested data type right?
<kristianpaul>
ha, nice find indeed :)
<wpwrak>
yes, that's generally what happens. i.e., truncation on writes, zero-padding or sign expansion on reads (details depend on the architecture)
<Ayla>
but why is a board necessary to get USB with the UBB?
<Ayla>
wouldn't bit-banging be a solution?
<wpwrak>
yeah, but you still need to make the physical connection somehow. the Ben doesn't have a USB A receptacle.
<wpwrak>
UBB has already been "mass-produced", so it's something that's readily available (for experiments. once you have the thing working, you can make a neater board)
<blogic>
aloha
<larsc>
wpwrak: blogic wants to build a 802.15.4. maybe you could give him some hints.
<wpwrak>
ah. welcome to the joyful world of RF ! ;-)
<blogic>
wpwrak: i am off to bed, i'll ping you tomorrow if you dont mind
<qi-bot>
[commit] Werner Almesberger: ubb-usb/pin-change.fig: illustration of how to change multiple pins with a single write (master) http://qi-hw.com/p/ben-blinkenlights/f8185f4