Post by thread:Timing Routine Question

All, Looking at the two routines below, can anyone tell me what the differences are regarding functionality? I am using the 2nd routine right now as a delay of 41 uS in an LCD write routine. All works well. But, if I substitute the first routine in place of the second, when the circuit is powered up, all I get on the LCD is gibberish. I don't quite understand this. I'm waiting the same amount of time in both of the timing routines. They both are entered and exited the same way. I'd appreciate any feedback on this question. I'd really like to know why the second routine works, but the first one doesn't. Thanks and Regards, Jim delay: movlw 0x0C movwf tmr5 decfsz tmr5 goto $-1 nop nop return delay: nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop nop return spam_OUTjimTakeThisOuTjpes.com

Look on the programm memory if your code is not on the middle of a page change. For example, if the line goto $-1 is on address 0x0101 when you execute $-1 it will go to address 0x000, it means that you have to configure the PCLATH register. Be careful, cause I'm not quite shure of this, this is only what i tink, but you can give it a try, don't you? Mauricio Jancic -----Mensaje original----- De: Quentin [SMTP:.....qscKILLspam@spam@ICON.CO.ZA] Enviado el: Jueves, 11 de Noviembre de 1999 03:51 p.m. Para: PICLISTKILLspamMITVMA.MIT.EDU Asunto: Re: Timing Routine Question James Paul wrote: {Quote hidden}> > > delay: > movlw 0x0C > movwf tmr5 > decfsz tmr5 > goto $-1 > nop > nop > return Hmmm, looks OK, quick adding in my head says it should be 41 microsecs. I use goto label, but from what I've read, the $-1 should be correct. Only other thing that is different from your second code is that W returns from the call with a value 0x0C. Quentin

Jim, I agree with Quentin. The routine looks fine and I use the $ notation all the time without problems. You are probably setting W with some value before the delay routine that you want to have after the delay routine returns. The other suggestion about the routine being on a page boundary is a possibility. Look at the .LST file generated by MPLAB or MPASM and see how big the application is. Adam On Thu, 11 Nov 1999 10:29:15 -0800 James Paul <.....jimKILLspam.....JPES.COM> writes: {Quote hidden}> All, > > Looking at the two routines below, can anyone tell me what the > differences are regarding functionality? > > I am using the 2nd routine right now as a delay of 41 uS in an > LCD write routine. All works well. But, if I substitute > the first routine in place of the second, when the circuit is > powered up, all I get on the LCD is gibberish. I don't quite > understand this. I'm waiting the same amount of time in both > of the timing routines. They both are entered and exited the > same way. I'd appreciate any feedback on this question. > I'd really like to know why the second routine works, but the > first one doesn't. > > > Thanks and Regards, > > Jim > > > > > > delay: > movlw 0x0C > movwf tmr5 > decfsz tmr5 > goto $-1 > nop > nop > return > > > > > > delay: > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > nop > return > > EraseMEjimspam_OUTTakeThisOuTjpes.com Adam Bryant (age 0x23) adamdbspam_OUTjuno.com Parker, CO, USA Robotics, RC Airplanes, anything using a PIC ___________________________________________________________________ Get the Internet just the way you want it. Free software, free e-mail, and free Internet access for a month! Try Juno Web: dl.http://www.juno.com/dynoget/tagj.

{Quote hidden}>James Paul wrote: >> >> >> delay: >> movlw 0x0C >> movwf tmr5 >> decfsz tmr5 >> goto $-1 >> nop >> nop >> return > >well, I don't see any obvious problems. this works, and I have coded >similar routines, although I usually use a label (not always, though), >and out of habit specify the destination as 'F'. As far as I recall, the >assembler defaults to 'F' if you leave it out, so that should be ok. > >However, I sometimes make the mistake of declaring two registers >at the same location. So whenever I get wierd results, that's one >of the first things I check. Perhaps that's what happened? It >would show up if you simulated the code. > >Phil Eisermann >H:(440) 284-3787 (@spam@mazerKILLspamix.netcom.com) >O:(440) 329-4680 (KILLspampeisermaKILLspamridgid.com) > >

Tony, tmr5 is defined as location 0x29 On Mon, 15 November 1999, Tony Nixon wrote: {Quote hidden}> > James Paul wrote: > > > > All, > > > > Looking at the two routines below, can anyone tell me what the > > differences are regarding functionality? > > > > delay: > > movlw 0x0C > > movwf tmr5 > > decfsz tmr5 > > goto $-1 > > nop > > nop > > return > > What RAM location is tmr5 defined as? > > -- > Best regards > > Tony > > http://www.picnpoke.com > Email spamBeGonesalesspamBeGonepicnpoke.com TakeThisOuTjimEraseMEspam_OUTjpes.com

James Paul wrote: > Looking at the two routines below, can anyone tell me what the > differences are regarding functionality? I assume you are using a 16F84? Register location $29 sounds fine. I can only suggest the blitheringly obvious. The loop routine alters tmr5 and most particularly, the W register. Are you absolutely *sure* the calling routine doesn't mind this? -- Cheers, Paul B.