[Ti] A real world comparison

[illovox at comcast.net]

Another person on the G-Book list had this to say:

> From: Bill Briggs <web at nbnet.nb.ca>
> Subject: Re: Is Apple really moving to x86 architechture?!
> 
> Well, there are some clues if you read enough of the fine print. I
> too (as a poorly paid professor of electrical engineering in a
> Canadian university and an Apple shareholder) would be rather annoyed
> if Apple actually sold x86 Macs. But there is a more interesting
> possibility.
> 
> Back when DEC folded the Alpha project (which was probably the best
> processor design around at the time), all 300 of the engineering
> staff from DEC went to work for Intel. They are currently working on
> a new chip at Intel that will, if the press is correct, be shipping
> sometime in 2007. It will not be an x86 family chip, and will have
> tossed all of the legacy support mechanisms that the x86 family
> carries. If it's something in the same class as the Alpha then it
> could really be good for the Mac.
> 
> I dislike x86 architecture too (having written assembler for it I
> know all too well how ugly it is), but the fact is that IBM wasn't
> willing to invest in the future of PowerPC for Apple. I suspect that
> someone at IBM just pissed Steve Jobs off one too many times, and
> that's the end of it. I'm not selling my Apple stock. I'm still going
> to buy a new 15" PowerBook this year, and I'm taking a wait and see
> attitude concerning what's going to happen two years out.
> 
> Jobs didn't do this for profit, he did it for long-term viability. If
> IBM had been able to provide the CPU development, he'd have stuck
> with the PowerPC platform.
> 
> So we can all wear a black arm band for a week and then get on with
> work. But I'm betting that the chip you see in the Mac in 2007 is not
> an x86 processor, but a new one, and one developed by those 300 folks
> they got from DEC.
> 
> - web

> Message: 7
> Date: Sun, 12 Jun 2005 09:55:57 -0500
> From: Chris Olson <chris.olson at astcomm.net>
> Subject: Re: [Ti] A real world comparison
> To: "A place to discuss Apple's Titanium computers."
> <titanium at listserver.themacintoshguy.com>
> Message-ID: <2adc80044ae436f12ea9ab20a7853df0 at astcomm.net>
> Content-Type: text/plain; charset=US-ASCII; format=flowed
> 
> On Jun 12, 2005, at 5:20 AM, Richard Meyeroff wrote:
> 
>> I would like to know if anyone has made an in depth evaluation of the
>> difference between SSE3 & the Altivec implimentations in the G4+ and
>> the 970?
>> 
>> This to me appears to be where Intel falls short.  They appear to do
>> better in Interger computation, the reason that Mathematica did so
>> well, but fall short by a wide margin in the vector, the strength of
>> the PPC platform.
> 
> That's exactly what I'm trying to determine.  I'm going to use some
> PowerPC vectorized benchmark code ported to x86 to determine the
> viability of the x86 platform for our own use.  I haven't paid too much
> attention to the x86 branch of Darwin because I always figured it was
> insignificant.  Now I have to play catchup just to get it current :-(
> 
> BTW - I successfully booted 8.0.1 (x86 Tiger core) on an AMD Athlon
> 2600 with an Asus A7N8X just to see if it would.  It did.
> --
> Chris