[Coco] PIAs destroyable by software on early CoCo models?

jdaggett at gate.net jdaggett at gate.net
Mon Dec 8 03:30:00 EST 2003


Torsten

That may have been on the 6820 device and not the 6821. I don't think many of the 
CoCo 1's shipped with 6820's. 

The 6821 on port A actually reads the port pin itself. So that PortA can be 
configured as an output or an input and still be read the state of the port pin. PortB 
is not that way. If a particular pin is configured as an output, you can not read the 
state of the output pin. Also PorttA is CMOS compatible. As an input it requires 
more current to drive the input. ALso port A has internal pullup resistors. PortB is 
not configure the same. To use with a keyboard PortB drives PortA. On power up 
Port A will go to a high state and Port B will float.  

The fact that PortB can not directly read the output pin when configured as an 
output may lead some to believe that the part may have been damaged. 

james


On 8 Dec 2003 at 8:26, Torsten Dittel wrote:

> Since I didn't get any comments to this one but I'm still curious
> about that, I repost:
> 
> One thing I remember far far away and very darkly about this stuff:
> 
> You have to be shure, that the PIA port you're POKEing to is in output
> mode, the one you're PEEKing is in input mode (you can set this with
> other PIA registers).
> 
> Another thing I remember (but this could be a rumour or I just had a
> bad dream as a child):
> 
> Of course you could this "vice versa": POKEing the rows and PEEKing
> the columns. Some games did it that way, and sometimes failed: On
> early CoCo-Boards (maybe using the 6820 instead of th 6821?), you
> could damage one of the PIA outputs by setting both PIAs to output,
> setting one as LOW and the other as HIGH and pressing a key. The
> damaged port was only working as an input after that. You didn't
> notice that from BASIC (because it always checks keys the way you
> described it), but those games doing it the other way failed. Later
> versions hadn't this problems, the ports were protected (maybe a
> feature of the 6821?) against this "short circuit".
> 
> Hmm... thinking about it I guess you could destroy any of that ports
> that way, depending where you set the LOW and HIGH states... maybe
> nothing more than a nightmare... Hardware gurus out there?
> 
> Regards,
> Torsten
> 
> 
> -- 
> Coco mailing list
> Coco at maltedmedia.com
> http://five.pairlist.net/mailman/listinfo/coco





More information about the Coco mailing list