[Coco] CoCo Ethernet Revisited?

[Mark Marlette]

At 02:02 PM 11/16/2003 -0500, you wrote:

Mike,

KA9Q was ported to the CoCo3 by Robert Brose, many years ago. It was broken 
into various different parts and it's implementation was limited. This was 
because of the address space of the CoCo. Now there are techniques to 
eliminate this, you have proven that. From what I remember Bob saying is 
that fully implemented sockets are larger than the CoCo's address space. I 
don't know if this is true or if I heard that correctly. I do know that 
there is a mess of protocols today and the horsepower behind the 6x09 is 
not great enough to do it all and still be effective. That is why the 
solution to this project was the iChip.

Mark
Cloud-9

>In a message dated 11/16/03 5:38:23 AM Eastern Standard Time,
>mmarlett at isd.net writes:
>
> > I stopped at the first page. This device provides an Ethernet interface.
> >  You would still have to have a device or the CoCo implement ALL of the
> >  protocols. A HUGE TASK!
>
>Ethernet protocol is probably a lot simpler than TCP/IP or SLIP or even KA9Q.
>One of my last jobs at Bell Labs required learning how Ethernet works (after
>having used it in my Sun workstation for years as "black magic").  I was
>surprised at how simple it is -- but then, it's really old, almost as old as
>RS232-related protocols.
>
>The only thing complex about it is the "collision" retry -- you have to wait
>a random amount of time (or an amount guaranteed to sometimes be different
>than any other member of the network is using) and try again, if the "line is
>busy."  And that's not really complex.  No, I'm not volunteering -- this 
>was over
>5 years ago, and I don't have ready access to all the info that I did back at
>the Labs.  --Mike K.
>
>--
>Coco mailing list
>Coco at maltedmedia.com
>http://five.pairlist.net/mailman/listinfo/coco