Hi Bruce, thanks for the reply. I think you may have misunderstood my
question a bit. Basically, I need (or rather want, I could get away with
fewer, and may have to, as my calculations show if they are correct) 256
"gradients" of delay. Each individual section of delay should be ~162.5
instruction cycles. If I could set the prescaler to 256 different
values, I could do it your way. But I only have a few values that I can
set the prescaler to, so I think that should remain a constant.
Unfortunately, one of those values isn't 162, so I think I will have to
compromise. FWIW, the way to get 1:1 prescaling is to assign the
prescaler to the watchdog timer.

Thanks,

Josh

--
A common mistake that people make when trying to design something
completely foolproof is to underestimate the ingenuity of complete
fools.
        -Douglas Adams

"Gennette, Bruce" wrote:
>
> 1 TMR0 counts UP to 255, then to 0 then to 1 .... then to 255, ....
> 2 You can over write its current value at any time (2 instruction cycle
> delay in being set).
> 3 You can use the pre scaler to 'tick' TMR0 at a slower rate.
> 4 You can feed the pre scaler from the internal OSC/4 or from a separate,
> external source.
> 5 Each time it hits 0 it tries to set an interrupt flag.  The user must
> clear this flag manually.
> 6 You may choose not to use interrupts, just manually examine the current
> value to find when it passes 0.
> 7 You can manually count ticks until your desired delay has elapsed.
>
> For your delay use 256-162 = 94 as the value to start counting from.
> Set the prescaler to use by TMR0
> Set pre scaler to x1 <--- damn, not allowed, minimum is x2
> Recalculate your delay to 256-(162/2)= 175
> Set the timer interrupt AND global interrupts on
>
> In your interrupt routine over write TMR0 with 175, reset the tmr0 flag &
> global flag and return
>
> This should be enough to get you started ;)  OR maybe not ....
>
> To handle interrupts you either start your handler routine at address -org
> 4- or put a goto handler_address there (Note - this interrupt vector address
> varies with different PICs, check your data sheet).
> PICs turn off global interrupts when any valid, enabled interrupt occurs, it
> is up to the user to check and set (or not) any other interrupt flags.  The
> return from interrupt should not reset globals (giving the programer full
> control over how multiple interrupts are handled).
>
> Now this should be enough to get you started ;)
>
> Bye.
>
> -----Original Message-----
> From: Josh Koffman [mailto:listsjosh@3MTMP.COM]
> Sent: 13 November 2001 14:58
> To: PICLIST@MITVMA.MIT.EDU
> Subject: [PIC]: Beginner TMR0 Questions
>
> Hi all. It's time for Josh's nightly "I'm trying to use parts of the PIC
> I don't fully understand" question. Well, thanks to Bob, Tony, Wouter,
> and I think Vasile, I now understand the PCLATH. Well, enough to get the
> job done. Anyway, now I must calculate the values I will put in to that
> jump table. Here is the situation:
>
> I am taking an 8 bit value, and using it to come up with 256 values of
> delay. Just so you know, it's a 16f628. Each individual delay section
> will be 162.5 instruction cycles. I'll round it up or down for the ones
> that don't come out to whole values.
> <snip>

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads