[ This is a repost. I don't think the previous message got out. ]
I am thinking of buying a PC. It will initially be used for developing some
animation software as part of my research. I plan to run linux with Xfree86
and write the software using PEXlib. (Hopefully, then I can run it both on
PCs and the university Suns, and it seems as good a standard to use as any.)
Questions:
1) What are the important criteria in selecting a graphics card for running
X11 programs that use PEXlib? I would guess the criteria are different
from text-in-windows applications, but perhaps the differences are dwarfed
by the time spent in the pipeline before PEX even considers displaying
the pixels. Enlighten me, please.
2) What chipsets/cards are recommended? The Xfree86-benchmarks summary
posted recently that compared Xfree86 servers running on different
chipsets showed some of the "best" to be:
xstones chipset
120+ #9 GXE Level 12 (cost?)
123 Diamond Stealth Pro
102 Actix Ultra+ (cost?)
89 Diamond Stealth 24
76 Orchid Fahrenheit VA
Are these benchmark results relevant, ie. would these cards be among the
best to use for my needs? If not, which cards are recommended, and why?
3) The Western Digital WD90C33 chipset is supplied as standard with one
of the machines I am considering. How does it compare with those above?
How well (if at all) is it supported by Xfree86?
4) Does PCI have a future (cf. Betamax--VHS, OS/2--Windows)?
If I go for a PCI machine instead of VESA, are good cards are available
for it NOW? Should I just stick with a VESA machine? (I guess a lot of
my decision on this will depend on availability of the card if it turns
out the the WD90C33 is not so good.)
5) Given a particular card, what does extra video memory get me?
More colours? Faster screen updates? Higher resolution? ...?
Please email replies to me directly if possible. The news admin was
muttering that news might become (more) unreliable while they change
to INN(?) so I don't trust posted messages to reach here.
thank you,
-jonathan
--
Jonathan H N Chin, 5 kyu | Dept. of *netics, | "Respondeo, etsi mutabor"
| University of Reading |