> Unfortunately, I can guess: as soon as your design includes PCI you're up
> the creek. Intel has you by the short ones. (And just try selling someone
> on a motherboard that doesn't support PCI.)
This might not be an issue anymore; I just heard today that IBM/Compaq and
some other company (name escapes me for the moment) have got together and
designed a new PCI specification [PCIx] that's supposedly runs at 133Mhz
and can transfer up to a gigabyte a second. They left Intel out of the
loop - which can only be a good thing - I know many people are fed up with
Intel designing system buses and chipsets that leaves people up in the
wind. [cf. http://www.slashdot.org for the news item] Anything that
extends the life of PCI can only be a good thing, and most importantly, if
this can be extended to other processor architectures, we could perhaps
have the ability to use PCI cards in these, especially if these have
flashable ROMs, that could mean that in order to use a x86 PCI card in an
Alpha, all we need to do is to flash the ROM with alpha code and you're up
and running.
I've been reading up on the AGP stuff. Personally I think it's a big
marketing con designed to lock people into the Intel monopoly. It's not
that much faster than a PCI card anyway [cf. Tom's Hardware Page - forget
the URL at the moment - I'll look it up if you're interested in reading
his expose of the AGP]
> I'd be half tempted to try running your design past the smart folks on
> the chip design side of Hamerschlag D level. (Who, me? I'm no chip
> designer; I compile software for 'em when I'm not installing systems
> for 'em. Naturally, they want Linux :-)
Heh, let me know the reaction. :o)
Cheers,
Alex
-- /\_/\ Legalise cannabis now! ( o.o ) Grow some cannabis today! > ^ < Peace, Love, Unity and Respect to all.http://www.tahallah.demon.co.uk
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/faq.html