This is a discussion on Re: Modern disk system for our -11s - VMS ; > On the PC hardware front, I'd expect PCI to last a good long while yet > (after all ISA is only just disappearing) and legacy PCI printer port > cards probably will be generally available for a good while ...
> On the PC hardware front, I'd expect PCI to last a good long while yet
> (after all ISA is only just disappearing) and legacy PCI printer port
> cards probably will be generally available for a good while yet given the
> simplicity of parallel port -> PCI interfacing.
Guess again, PCI is sort of a dying technology. On one side it is being
replaced by PCI-X (which can sometimes use PCI cards, if they meet the
backwards compatibility requirements), and on the other side it is being
replaced by PCI-Express which I believe is even less backwards compatible.
> I'd ideally like to go for as simple a solution as possible for the
> design, so a few registers on the -11 side to read status and write
> requests plus the logic for handling the DMA. Then on the PC side
> capability to read/write to same registers plus stream data to/from the
> DMA logic.
I'm not sure you're going to have a simple solution, not if you want
something that you don't have to reimplement every few years. If you look
back 5 years ago then PCI, EIDE, Serial and Parallel were all the standards
in the PC world. These days it's PCI, SATA, USB, and Ethernet (as I mention
above, PCI is on the way out). Right now you can get USB-to-Serial and
USB-to-Parallel, but it's hard to say what affordable interfaces will be
around in another 5-10 years.
This is why I'd suggested a solution that talks to the PC via ethernet, that
at least looks to have a semi-decent chance of being around in 10+ years,
especially if you use some messy broadcast sort of protocol of our own
devising. Granted you might have to buy a widget or two to bridge cable
types, but other than that you should be good to go, and on the PC side
you'd be software. Of course on the PDP-11 or VAX side you'd be looking at
a significantly more complex interface.
To unsubscribe (or subscribe) from (to) this list, send a message to
firstname.lastname@example.org, with the first line of the message
body being "unsubscribe" or "subscribe", respectively (without the quotes).