Roy Schestowitz wrote:
> __/ [ OK ] on Saturday 19 August 2006 05:20 \__
>
>> http://channel9.msdn.com/showpost.aspx?postid=227259
There's no question that SIPs is an impressive technology,
particularly since it reduces hardware dependence (address violation
protection, et al). I was especially taken with the fact that it
facilitates passing pointers, rather than their corresponding data,
but without compromising integrity; something that can potentially
increase system performance by an order of magnitude.
As for this technology making it into Vista; well as stated in the
video "don't hold your breath". It would have taken Ballmer's
infamously failed "rewrite" to accomplish *that* task, assuming that
SIPs was ever part of that erstwhile plan to begin with. SIP's is only
in a reasonably early research phase, after all, despite the existence
of a very basic "busybox" style OE.
I was also very interested in a particular comment made by Calen Hunt
of the MS Operating Systems Group. When replying to a comment about
the fact that many Windows hardware drivers are currently in the
transition of moving out of kernel mode; he said "in which case they
will pay the hardware performance penalty". As if Vista didn't have
enough performance issues with it's heavy handed Uber-gooyey
eye-candy; now the hardware drivers are going to run at a snails pace
too.
--
K.
http://slated.org - Slated, Rated & Blogged
This message has not been photoshopped in any way.
Fedora Core release 5 (Bordeaux) on sky, running kernel 2.6.16-1.2133_FC5
14:00:57 up 63 days, 14:17, 3 users, load average: 0.05, 0.01, 0.00
|
|
