Date: Sat, 24 Oct 1998 01:30:33 +0100 (BST)
Reply-to: classiccmp(a)u.washington.edu
From: ard(a)p850ug1.demon.co.uk (Tony Duell)
To: "Discussion re-collecting of classic computers"
<classiccmp(a)u.washington.edu>
Subject: Re: More thoughts on building a Z-80 (64bit!!!)
Wire
wrap?...hmmm I have problems with that in ohmic losses even the
connections is good, soldering them helped a bit but no dice.
Well, I didn't use wire-wrap for the power or (especially) ground
connections. I also soldered the decoupling caps between the power pins
on the socket. I have _never_ seen resistive losses on a signal
connection. And I've never had trouble with resistive losses in wire-wrap.
Done that bypass caps and few fitler caps scattered across 4 seperate
board strung with that blue wire wrap from rat shock. I can't
believe how much resistance it has and it's magnetic material not
non-ferrous stuff like copper and such!
I laid out standard layout in a way these boards can be stacked on
each other. The wire wrap was just a trial run.
Allison, you are asolutely correct on that wire wrap poor for power
transfer, hey, I was learning, right??? :-)
Have data transfers done with DMA, all memory
mapped and irq driven
to knock subCPU as needed to grab data then place it in CPU's lap.
Leave CPU alone for processing. How about that?
Fine for data transfer, but having some I/O ports to send commands to
peripherals, set the base address for DMA transfers, etc is surely
sensible for most machines.
Oh yeah the seperate I/O channel, better make it 2 busses?
Jason D.
-tony
email: jpero(a)cgocable.net
Pero, Jason D.