Re: PC Vector-generator card

From: Clay Cowgill <clayc_at_diamondmm.com>
Date: Tue Dec 09 1997 - 14:46:43 EST

>I know I sound like a broken record about this, but....
>
>What about using an IDE interface to the vector generator?
>
>If you can keep the number of registers down to 8, you
>can trivially interface it to an ISA card, and us Mac
>weenies (ok, probably just me..) can talk to it using
>a cheap PCI IDE card. There may be an integrated bus
>master ISA DMA part (I know the PCI parts have it)
>and you could have a DMAed list in PC memory for the
>display list.

I must have been dozing off... I hadn't heard the IDE interface idea
before, just the PCI one. ;-)

Actually... IDE sounds pretty good. I'll have to look into what
requirements that places on the PC (like adding a second $8 IDE card-- I'm
a little nervous to have my "IDE" device attach to the same interface and
card that someone's Hard Drive is on. "Ooops! Sorry, just scragged your
hard disk." ;-)

The DMA part could be nice. I'll have to think how that'd work into the
design. (I'm just making little pieces of it right now, line draw hardware
first, then the "engine".)

I was thinking of making Vector RAM map into some unused BIOS space on the
PC, then have an I/O port bit that basically says "GO". (This tells the VG
hardware that the display list is complete and it should draw a frame, and
can also serve to isolate Vector RAM from the PC bus and let the vector
generator have it for a while...) Frame rate is controlled by the PC that
way, and the interface is really simple. I need to go dig up the old
"Firmware Furnace" articles in Circuit Cellar Ink about using a 386
motherboard for embedded control...

Hmmmm. Well, I can put off the decision for a while. I want to just be
able to generate vectors on my o'scope first. ;-)

-Clay

Clayton N. Cowgill Engineering Manager
_______________________________________________________________________
/\ Diamond Multimedia Systems, Inc. clay@supra.com
\/ Communications Division http://www.supra.com/
Received on Tue Dec 9 11:47:42 1997

This archive was generated by hypermail 2.1.8 : Thu Jul 31 2003 - 23:01:04 EDT