[Coco] CoCo's TIMER and 60hz

[Arthur Flexser]

I had occasion to investigate the accuracy of the CoCo's timing back in my
pre-retirement career as an experimental psychologist.  Here's a 1987
article I published about the CoCo in the experimental laboratory:
https://link.springer.com/article/10.3758%2FBF03205614  (Hit the "Download
PDF" link to view the full text of the article, for free.)

Art


On Wed, May 29, 2019 at 2:27 PM Allen Huffman <alsplace at pobox.com> wrote:

> > On May 29, 2019, at 1:20 PM, Arthur Flexser <flexser at fiu.edu> wrote:
> >
> > One thing to be aware of in testing TIMER's accuracy under Basic is that
> > the break key check done between Basic statements occasionally causes
> > interrupt processing to be omitted.  The clock "misses a tick" sometimes
> > due to the fact that there are a couple of reads from $FF02 in Basic's
> > POLCAT keyboard scanning routine that clear the IRQ flag.  If one of them
> > occurs just at the right (wrong) time, the flag gets cleared before the
> > interrupt can be processed and TIMER doesn't get incremented.  You should
> > enter POKE&HADEB,&H39 before running your testing program to knock out
> the
> > break key check.  You'll see that it makes a significant difference.
> >
> > I think different CoCo's can vary by a few tenths of a percent in the
> > timing rate.
>
> A few decades ago I would never have believed I’d get a response from “the
> ADOS guy”. Thanks, Art! Very good to know!
>
>                 — Allen
>
>
> --
> Coco mailing list
> Coco at maltedmedia.com
> https://pairlist5.pair.net/mailman/listinfo/coco
>