View Full Version : Question to Parallax and Stamp users
Question to Parallax (wish list)
Since the SX is now being utilized on the Basic Stamps, any future plans to use;
CMOS inputs Schmidtt trigger inputs Analog comparitor inputs Edge detection Interrupts
Post Edited (KenM) : 3/8/2005 3:22:08 AM GMT
I believe the basic stamps are meant to be simple, IMO if you want to do that simply get the sx. I use both the BS2sx and the SX28 individually now. They each have their place if you dont mind the assembler.
03-08-2005, 11:06 AM
Interrupts are not practical in a BASIC microcontroller like the BASIC Stamp -- even some controllers that claim interrupts actually disable them before high-level commands to prevent the command (e.g., SERIN, SEROUT, other time-critical functions) from getting trashed. The other ideas are interesting.
The reason we released the SX/B compiler was to give you the ability to have interrupts while still having a language very close to PBASIC.
Applications Engineer, Parallax
Dallas, TX· USA
03-08-2005, 09:28 PM
Unfortunately there's no "None of the above" selection. I think Parallax has milked the functionality of the 16C54 architecture as far as it can go. The BS2sx has used the SX chip for some time, actually. And note even the SX chip itself is an uber-powerful version of an old PIC design. If it didn't run at 50 MIPS I wouldn't mess with it.
"Basic Interrupts" might be nice (these are interrupts that only occur BETWEEN Pbasic tokens). Having seen another vendor's implementation of them, they're very clunky, hard to understand, slow down other processing, and not very useful (you really want 'real' interrupts, but in a single-tasking run-time that's really hard). And it would probably cost a LOT to implement, for VERY little benefit.
I think their current work improving the SX/B compiler for the SX chip is brilliant. Much more bang for the buck there, and you start out with 'native' interrupts.
Improvements are fine, just make sure you don't kill the goose that's laying the golden eggs in the process.
03-08-2005, 10:43 PM
The BS2p and the BS2pe do have the polled interrupt capability, although limited, already. I think the other capabilities are within reach and would be useful additions without messing with past success. One thing at the top of my list in addition to the above would be pin selectable weak pullups. To attach a NO pushputton to an input, you could use the internal pullup (which is about 20 kohms) instead of an external resistor, and you could turn the pullup off to save power when not needed.
Advanced features are great, but each new feature you use does introduce the possibility of a bug somewhere else in that same program.
About the configurable pull up resistors, I purposely left that out. And maybe you know why when you indicated its use with a NO switch.
I used the configurable pull ups connected to the collector of a 2n3904, results were unpredictable, the input would sometimes read low, even with the 2n3904 off. Since then I have shyed away from using the pull ups.
I should probably revist and investigate further·some day. Does anybody know what the impeadance of an open collector is for a standard 2n3904?
03-08-2005, 11:29 PM
Well, son of a gun, the "POLLxxx" commands on the BS2p and BS2pe do exactly what I was talking about in "Basic Interrupts" above. That capability already exists (in the really fancy BS2's, but what you gonna do?). Thanks!