[Wolves] Apple moving to Intel?

[Jon Masters]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

chris procter wrote:

> It makes very good business sense, they get a cpu
> manufacturer that is dedicated to making fast desktop,

Intel are dedicated to outselling the competition and making a profit
for their shareholders - I've yet to be convinced they really truly care
about the tech on quite the same level, although some of their newer
stuf is fairly pwetty.

> and fast and low power laptop cpus rather then one
> supplier who concentrates on embedded and one that
> concentrates on server and games machine cpus.

Building embedded CPUs teaches the likes of IBM and Freescale how to
handle heat and power issues properly - maybe Intel have this magical
stuff up their sleeves, but I am forced to wonder whether this isn't
just Jobs being mollycoddled into thiking things will get better by
dropping their long term partners and going with Intel - who also won't
/really/ care about the insignificant numbers of Apple sales.

...but. One thing this whole thing did make me wonder was what would
happen if Intel turned around and *bought* Apple outright.

> Intel get a big name that has no backwards compatability
> issues

But they seem to still be jumping on the x86 bandwaggon rather than
using something that doesn't truly and utterly suck.

>> and now they've decided they want to start
>>producing cheap whitebox trash

> I doubt it will be whitebox trash

It will. I'll probably still buy it because I'm a whore, but that
doesn't mean I'll be happy about it - I don't recommend the ipod because
it looks pretty but sucks arse in many ways, but I bought one because
I'm a happy go lucky vanity freak who likes cute looking kit.

> it will probably be the most advanced x86 hardware
> anywhere with all the old crufty bits removed

It'll be shite. I can say that now - though they can freely attempt to
demonstrate to me that it's possible not to make shit x86 stuff. I hope
they keep OpenFirmware and those bits of the Apple stack that worked
well with the PowerPC - but I'm afraid there'll be a nasty "Apple BIOS
1.0" running on some unpleasantly hacked together piece of x86 crud they
dug up from some reference design somewhere.

> But I'll need convincing as well.

Lots and lots and lots and lots.

>>I've started moaning that IBM should somehow revive the PPC
>>Thinkpad they did.

> There is actually a gap in the market now for someone
> to produce non-x86 based laptops for the geek (how I
> hate that word) market. There is a minority of Apple
> customers for whom the ppc was part of the mystique
> (including me, and by the sounds of things you). Its a
> small market but all the software (linux) is in place,
> and the customers would evangelise it for you.

I'd go out and buy one. Maybe two - it's looking likely that I'm going
to have to buy another Powerbook next year just to keep me going until I
can find a vendor who haven't sold out.

> I cant really explain why the ppc is so important

I can. I like writing code on ppc. It's well designed and does all of
the good warm and cuddly theory bits really well - it doesn't suck.
Intel would freely admit that the x86 is too long in the tooth if they
could get away with doing so. They make some good alternatives, but I
don't see Apple jumping to go with those yet.

cue raising of the Itanic and accompanying music/jokes from geek sites.

> at the end of the day as a user you cant really tell the
> difference.

The average user can't, anyone who does anything remotely funky will - I
wonder what the audio folk will say, for example. Programmers lose most
of their useful registers and go back to the dark old stoneage days of
having a few braindead and useless x86 latches bolted on the end of a
moldy piece of toast.

New Apple computers! ding-da-da-da-ding! Who want's to bet Apple also
sell out and start playing that nasty and annoying Intel jingle?

> I guess the x86 is like windows, it works,
> it does what you want, but you always have this
> nagging feeling that its not as elegant and
> technically beutifull as it should be.

That's about it. Now, Xscale, anything else that doesn't suck, ok. But
x86? come on Apple. Why, oh why, oh why, oh why? :-)

> Actually the BSD layer probably makes no difference,

Darwin is more BSD than Mach. It doesn't run as a Microkernel - it's all
in the same vm address space for a start. It won't be a lot (relatively
speaking) of work to port things across, given that Darwin is known to
run just fine on x86 anyway - remember, we wondered whether there was
something more to the Darwin x86 release anyway.

> That does leave open the remote posibility that Apple
> could release a mips based machine sometime in the
> future...

Or that they could see the light and go back to something non-x86.

>>Let's face it, Apple is about as proprietary as they
>>come - despite their open appearance (they throw code over the
>>garden fence from time to time but remain a tight lipped
>>organisation)

> Harsh.

But fair. I rarely meet any Apple folk at Free Software/Open Source
events and managed to meet precisely zero of them at USENIX 2005.

> Apple have release some good stuff into the
> open source world, Darwin is so far the only kernel
> relased from captivity into the wild (at least until
> OpenSolaris appears).

True. Darwin is pretty tasty, but it wasn't their own creation. I met an
Apple guy a while back who was in charge of various OS stuff - I asked
if he was "Mr. Lovingly crafted" (read the fun comments in the darwin
source code) but he had no idea what I was refering to.

> changed the way they work with webkit to sort out the
> problems they had with the khtml team

I don't think that they *really* care, but yes, that was quite nice.

> I for one mourn the passing of our PowerPC overlords.

I was pretty gutted - I should have gotten out more that day.

Jon.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFCqbaPeTyyexZHHxERAn31AJ9IqCTdU642fY0pa/KU55zzqkShlQCfX5MH
/Mezs1+Vx9sRntizGBKSkbs=
=bcRi
-----END PGP SIGNATURE-----