From: christianlott1 on
On Apr 6, 10:39 am, BruceMcF <agil...(a)netscape.net> wrote:
> On Apr 6, 2:42 am, christianlott1 <christianlo...(a)yahoo.com> wrote:
>
> > HOOK IT UP BRUCE ! ! !
>
> I'm the one who says, "it's a hardware problem" ... eg, a software
> guy.

i always say that too. but don't you know those hardware guys are
there for you. one of them may actually want to help...


> Why do you think I focus on things where the hardware problem is
> getting the right wires to the right pins, and all of the complexity
> is on the software side? That I hope to be able to do.

maybe i can bribe someone to hook it up on the cart port for me.
(any takers?)


> However, I was definitely forgetting about the dot clock on the
> expansion port ... a lot of SPI chips are happy to go 8MHz, and indeed
> happy to go substantially faster than that, so a state machine driven
> by the dot clock is definitely a workable way to plug in SPI chips.

i'd be happy to see autoboot and a real DMA register implemented -
whether it was through a sequential state machine (ie <WRITE>,<SOURCE
ADDRESS>,<TARGET ADDRESS>,<NUMBER OF BYTES>) or separate control
registers (ie four latches and a sixteen bit counter).




> On the side of the SPI chip, I only want to solve the connector
> problem once. Since the simplest way to plug one in to experiment with
> it is into a joystick port (imagine being able to plug a fast modem
> into a C64DTV via a joystick port), I'm looking at putting SPI chips
> into joystick dongles, and then any other SPI ports will end with a
> DB9 male wired for the SPI chip in the same joystick dongle.

not interested in dtv but whatever's easiest for you to put together.
don't let my delusions of grandeur slow you down. they only get more
involved from here (even if i got the above correct :)