Re: [PIC]: {handling keys} by Koen van Leeuwen .

On Friday 18 June 2004 19:59, Bob Axtell wrote: > While it would seem that testing that a key is pressed would seem > simple, it actually isn't because most applications need to know > when it was RELEASED as well. In fact, for me, a keypress occurs when it > is PRESSED THEN RELEASED. > > About 10 years ago I began using a bulletproof way to process keys: > > 1. I read the keys- all of them- every 10ms or so. I can either read 'em > all at one tome, or rotate so that I read each one on a rotating basis. > Doesn't matter how its done, it works perfectly regardless. > > 2. Every 10ms I read the key and shift its value into a register > decicated just for that application. > > 3. If the key is ZERO when it is pressed, I look for the pattern > b'11110000' on that key's register. When that is seen, that key was > PRESSED AND RELEASED, so set a flag. To prevent more action, jam the > key's register with '11111111' so it can't be used again. What happens with key bounce? Isn't it possible that bouncing during release of the key would produce the pattern b'11101000'? > This works great. If anyone is interested, I'll publish some code. I'm interested :o) Koen -- http://www.piclist.com hint: PICList Posts must start with ONE topic: [PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads