[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

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



ZnU wrote:
In article <Cp1qe.26441$J12.21905@newssvr14.news.prodigy.com>,
 Kelly Hall <khall@acm.org> wrote:


Sean Fahey wrote:

The PPC wasn't going to go where Apple needed , fast enough. The Pentium 4
is actually a pretty good processor, with better branching and latency then
the PPC. Now they've got dual cores, hyper-threading - it was a good time to
make the change. Get over it already.

Apple doesn't want Pentium 4, although perhaps the dual core versions will sell to a segment of their market. What Apple wants is the Pentium M for the notebooks.


I'm quite certain Apple also wants the later desktop chips that will be based on it. Processors that evolve from the current Pentium-M architecture are eventually going to take over Intel's entire lineup, or so say the roadmaps. Even the high-end Xeon's market segment will be filled by such chips by the end of 2007. Turns out Intel can design a pretty good chip when it's not buying into its own MHz myth (which it isn't with this architecture, unlike with the P4).

Bingo.

Hopefully, we'll see an end to 30+ stage pipelines that blow away huge
numbers of in-process instructions on every "surprising" branch, and fewer four-level deep caching systems that only work well for little
loops with little data sets--but get the GHz rating up.  ;-)

"Whatever it gains in the straightaways, it gives up in the turns."

-michael

8-voice music synthesizer using NadaNet networking!
Home page:  http://members.aol.com/MJMahon/