[rescue] Apple to ditch IBM, switch to Intel chips

Skeezics Boondoggle skeezics at q7.com
Sat Jun 4 12:00:00 CDT 2005

On Sat, 4 Jun 2005, Patrick Giagnocavo wrote:

> Actually NeXT never made money making the hardware, so the situation
> doesn't exactly map to the current profitable-Apple scenario.

Exactly!  Apple is making money on their hardware, and as I understand it, 
running pretty good margins too.  So WHY would anyone in that position 
want to leap down into the muck of competing with Dell and HP and all the 
rest as Yet Another Intel Box Shifter?  It's _insane_.

The fact that Apple is profitable despite the fact that IBM/Moto can't get
the PPC scaled to 3+ Ghz - YET - is quite remarkable.  Given time, I would
trust that IBM can and will catch up with the x86 world in terms of clock
rates (despite what bunk we here know that is, and how much Apple spent on
marketing to counter the "Mhz Myth").

Or is it just that The Steve is ticked about sheer production volume
holding back his ability to ship machines?  In that case, finding another
fab to stamp out more chips would be sensible - even with IBM as your sole
supplier, it's never a good position to be in as a manufacturer, relying
on single-source components...

And whatever happened to the Motorola 8500-series chips?

> NeXT was also in the process of doing an IPO based on the strengths of
> their WebObjects and YellowBox (NeXT developer tools and runtime
> running on top of NT) software, then Gil Amelio and Steve sat down and
> the reverse takeover of Apple by NeXT occurred.

Wow, that's new information - I hadn't heard they were going to do an IPO.  
Huh.  It's still too damned bad they didn't get to ship the NRW, the
"brick".  It would have been one of the computer industry's all-time
juiciest ironies if Apple/IBM/Moto had spent all those $Bs to develop the
PowerPC and - by virtue of having an 88K box already designed, then
converted to handle the PPC601 instead - if NeXT had actually shipped the
_first_ PowerPC-based machine.  Ah, well.

> Well, for a counter-point, Sun is actually selling their V20z and V40z
> Opteron boxes pretty well.

Yup.  And has it really helped Sun's bottom line?  I guess it's positive
in that maybe shops that would never buy a SPARC (or, did in the past, and
are disappointed that the "S" in "SPARC" hasn't meant "scalable" beyond
fits and starts) at least still have a Sun-logo box in the rack, and it's
likely that that box is running Solaris instead of something else... but
I'm betting the margins on the V20z/V40z are a LOT slimmer than on SPARC
hardware, and aren't quite a few of those Sun Opteron clusters actually
running Linux?

You can just tell that Sun wants to be selling SPARCs and Solaris, and the
dabbling with Linux and Intel/AMD is still sort of half-hearted... a
stopgap to staunch the hemmoraging... if the US-IV+ were more competetive
SPECwise with the POWER5 and Itanic (and whatever's left of the Alpha,
HPPA and MIPS offerings) then Sun would probably be pushing their own
hardware a lot harder...

So why not ask why Sun hasn't approached Intel (or AMD?) to build faster
SPARCs? :-)

> My guess is that Apple would have some way of locking down the OS to
> only work on x86 hardware they made, or they would use the Palladium
> stuff, some of which is already shipping into the market, to tie a
> copy of the OS to the hardware.

Bleah.  Leave that level of paranoia to MS and their registration schemes.  
Like Sun, Apple should be vigilant but not absurd about their licensing 
policies - they're still such niche players that *every* copy of MacOS or 
Solaris out there, legal or not, is still one less Windoze seat... 

Apple COULD, at any point, issue a MacOSX for x86 as a software product 
while still maintaining their own hardware line - hell, NeXTSTEP already 
had been ported to SPARC and HPPA as well, and all that "fat binaries" 
support is likely still in there, right?  Darwin runs on x86.  So why not 
just port the remaining proprietary bits and release the full MacOSX for 
Intel?  Afraid it'll scavenge hardware sales!?

Ugh.  Every time this issue comes up I just react to the idea that more
architectures are going to fall to the Intel onslaught, that the Wintel
monoculture will continue to devour everything in its path, and that
eventually there won't be anything out there interesting or innovative or
unique or curious or fun about the computer industry anymore - it'll all
just be the same bland, boring, bloated, locked-down DRMized crap dictated
by Redmond & Santa Clara, and all the rest will be also-rans.  I'll just
be one of those weird guys with a basement full of evidence that it wasn't
always like that, despite the vast sucking memory hole of the "consumer"  

Bleah.  Screw computers.  Time to go learn organic farming.  And Chinese.

-- Chris

More information about the rescue mailing list