Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!usc!snorkelwacker.mit.edu!ai-lab!wheat-chex!bson From: bson@wheat-chex.ai.mit.edu (Jan Brittenson) Newsgroups: comp.sys.handhelds Subject: Re: Dual tones for HP28S (and HP48) Message-ID: <12477@life.ai.mit.edu> Date: 20 Dec 90 17:09:02 GMT References: Sender: news@ai.mit.edu Organization: nil Lines: 25 In article TNAN0@CCVAX.IASTATE.EDU writes: > About the two-tone problem... Many micro computers of ~1982 had a speaker controlled by a single bit. Off meant the speaker element was retracted, on meant the element was protracted. Now, when the bit was toggled, the speaker change wasn't instantaneous. This delay, coupled with some basic timing, could be used quite creatively: turn the bit on, and before the speaker is completely protracted, turn it off. Then, before the speaker is retracted to where it was before the bit was previously turned on, turn it off. Keep juggling the speaker by keeping track of exactly where it is at any given instant. Anyone must have heard the Apple-II speak or played the infamous Castle Wolfenstein game. I believe this is how it was done - I never really did much programming at all on Apple-II, but I've used the trick on other - Z80 - microcomputers. It is not a perfect technique, and we're not talking about Hi-Fi, but it is *remarkably* good if carefully implemented. Now, back to the HP-48... Does anyone have any knowledge about how piezo elements behave? How instantaneous are they - is the delay less than, say, 10 instruction cycles? 1000 cycles? Is it potentially harmful to the crystal? What about battery drainage?