Re: How many times must Apple make such DRASTIC architecture changes?

[Paul Schlyter <pausch@NOSPAMsaaf.se>]

In article <dtz*kBQQq@news.chiark.greenend.org.uk>, 
timc@chiark.greenend.org.uk says...
>In article <1gxykbw.1h7o2pe1jygz5cN%neillmassello@earthlink.net>,
>Neill Massello <neillmassello@earthlink.net> wrote:
>>Tim Cutts <timc@chiark.greenend.org.uk> wrote:
>>
>>> one reason for Apple's
>>> switch, which Jobs alluded to, is that x86's integer performance is
>>> blistering (in both Intel and AMD's processors).  Way faster than
>>> anything else,
>>
>>I noticed that the power versus performance bar chart Jobs showed during
>>the keynote was for integer operations. 
>>
>
>Absolutely.  Integer performance is all that matters for the vast
>majority of desktop consumer applications (games might be an exception,
>these days).

ANY simulation of a real-world simulation is an exception.  And then
it doesn't matter if the simulation is for entertainment purposes
(games, movies), for engineering purposes (e.g. wind tunnel simulations
in aircraft construction), or for scientific purposes (e.g. numerical
weather prediction).

True, very few computers in the world are used for e.g. numerical
weather prediction -- but the outputs from these programs are used by
a vast number of people!

Btw more and more desktops are used for scientific computations too.  
Remember seti@home ?  That was only the first one -- similar desktop 
background applications now exist for such things as climate modelling and 
trying to find cures for cancer.  They are all floating-point intensive.

>Tim


-- 
Paul Schlyter,  Grev Turegatan 40,  SE-114 38 Stockholm,  SWEDEN
e-mail:  pausch at saaf dot se
WWW:     http://stjarnhimlen.se/