On Thu, Aug 19, 1999 at 10:02:33AM +0100, Adrian Cox wrote:
>
> What the IBM reference design doesn't address is actually the hardest
> part: building a complete workstation with a modern graphics system.
Depends; I'm looking at this for a server environment. Graphics don't
really monitor. But I'm considering that too. I don't really see the
need for more than PCI graphics since most of my high-powered (but not
supported by XF86) cards are PCI, such as the FireGL 1000 Pro and my
FireGL 4000's.
> There are three problems:
> 1) If you use plug-in graphics cards, you get a lot more choice with
> AGP. The IBM chipset doesn't have AGP.
This can be rectified with some minor difficulty; would require changes
at the component level, possibly a dropin replacement with something
that VIA makes perhaps?
> 2) Most graphics boards require you to execute x86 BIOS code. This makes
> your boot firmware even more complicated, as it probably has to contain
> Gabriel Paubert's x86 emulator. I don't think IBM even supply any
> reference firmware, certainly not any that can execute x86 code.
Erm, let us worry about that? Heh. Wait, you're one of us. Well, the easy
solution is to use Apple-ok'd cards, which work on PowerPC assembly. But
almost all of those are unsupported Radius cards or crummycrappyickyshit
ATI cards that make me wanna puke.
> 2a) Or you need a deal with a graphics chip vendor under NDA. If you are
> not Apple, you are probably too small to keep their attention.
Depends who you talk with. I think perhaps we should get all the PPC people
together, and designate people to specifically look into this. I'm sure
there's either a video card manufacturer that'll let you have an older
BIOS/Firmware revision that you could look at. If not, perhaps we can
convince them to create a firmware revision that works with PPC. Although
this means you have to run the card in a PC first and flash the firmware,
at least it gives you a good functioning card.
> 3) If you integrate a PCI graphics chipset, you're probably stuck with a
> low performance C&T or similar chipset, which will not give you
> 1600x1200/32-bit/OpenGl/etc.
> 3a) 2a applies to 3a.
I dunno; I bet you could probably embed a Permedia2v with 8M of SGRAM on
that board. }:)
> Most of the VME/CPCI PowerPC systems use serial consoles for this
> reason. Even MCG have stuck with supporting one or two low-performance
> graphics chip.
> If I'm too pessimistic, tell me the graphics chip vendor who's actually
> useful.
Well, here's at least a wishlist...
3DLabs, Number Nine, S3, Matrox.
You may whap me upside the head. I'm a *BIG* #9 advocate; I'm running
#9 RevolutionIV's in all my x86 machines. Did we mention they have a
mention of a RevolutionIV flat-panel solution pack for PowerMac? Yes,
*THE* 17" SGI LCD display and 32M RevolutionIV which is supported by
XF86. They even have timings and XF86 info for it. So, *boom*, problem
solved. Put an AGP slot on there, or find a way to embed at least 16M
of SDRAM on the board (the T2R4 is 16M/32M only) and you've got your
video problem solved. Can't believe I didn't think of it sooner. :)
--
-Phillip R. Jaenke, Systems Administrator, Nexbell Communications
AIX/BSDI/DG-UX/HP-UX/Linux(Debian)/SCO/Xenix/Xenix286
"Unix is so much more than a way of life. It's a way of.. er..
it's a way of SOMETHING! I just don't know what yet!"
PRJ5 / prj@nexbell.com / prj@mokole.nexbell.com
Attachment:
mutt-mokole-32557-10
Description: PGP Key 0xA83D9C0D.