[Coco] CoCo's TIMER and 60hz

[Dave Philipsen]

I'm guessing that he could tweak this to gain more accuracy. When 
generating the VGA signal, time spent in horiz or vert blanking can be 
changed ever so slightly to get it closer to where it's supposed to be. 
I ran into this exact same problem with an FPGA project I was working on 
with a VGA output. I used a 25 MHz dot clock as opposed to the standard 
25.175 MHz dot clock because it was easy to derive from the 50 MHz 
oscillator. This made anything clocked off of VSYNC run noticeably 
slower than it should have. I did a few calculations and determined how 
to tweak the blanking intervals ever so slightly to get back the 
accuracy. Every VGA monitor I've seen has a certain tolerance for these 
very slight changes.


Dave


On 5/29/2019 12:25 PM, Bill Nobel wrote:
>    The Matchbox VSYNC currently on the Matchbox is at 59.9 Hz so the Matchbox is a bit better than a real Coco3, but is still off.  He is makeing a special build for me that sends the Filed Sync clock to a pin as my Matchbox has extra headers attached that allow for easy testing with an analyzer.  I will let you know the results when I get the build.
>
> Bill Nobel
> b_nobel at hotmail.com<mailto:b_nobel at hotmail.com>
>
>
>
> On May 29, 2019, at 11:17 AM, Allen Huffman <alsplace at pobox.com<mailto:alsplace at pobox.com>> wrote:
>
> On May 29, 2019, at 12:14 PM, Bill Nobel <b_nobel at hotmail.com<mailto:b_nobel at hotmail.com>> wrote:
>
> Roger and I ran a test last night with my logic analyzer on a real Coco3.  VSYNC on the particular machine I tested reported 59.7 Hz.  The Coco3 was never accurate on software time due to fact the IRQ line is shared with other devices besides VSYNC.  Yes they are queued but arn’t serviced accurately based on time.
>
> Anything that masks interrupts can throw a timer using it off, of course — disk access, and I think even playing a tone masks IRQs (?) in BASIC.
>
> My question at this point is trying to determine if Matchbox CoCo is just off, or if there would be similar drift between two different CoCo 3s running the same BASIC program.
>
> — Allen
>
>
> --
> Coco mailing list
> Coco at maltedmedia.com<mailto:Coco at maltedmedia.com>
> https://pairlist5.pair.net/mailman/listinfo/coco
>
>