SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Intel Corporation (INTC) -- Ignore unavailable to you. Want to Upgrade?


To: Elmer who wrote (163394)4/3/2002 12:00:25 PM
From: Dave  Respond to of 186894
 
Elmer,

Another analogy to consider is this:

MHz is to Horse Power as a uP is to an engine. Just because one has a high performance engine does not mean any vehicle will perform well with the high performance engine inside it. For example, suppose a high output engine was placed inside a Bus? Do you believe the bus would have comparable performance to say an Indy car?

Most likely, AMD with their "quanti-speed" PR rating describes a "best case" scenario.

I believe it is a bad marketing move for AMD to go back to PR ratings. However, I believe it was AMD's only choice since it is clearly apparent that AMD's uP can not scale up as rapidly when compared to an "equivalent" Intel uP.

The average user could not care less who manufactures the engine of the automobile. Instead, consumers care about the reliablity, or the perceived reliablity, of the automobile they purchase.

IMO, the same goes for computers. MHz is the standard benchmark for computers. Consumers want the best computer they can afford. Of course, everyone wants to talk about "upgradablity", but I believe that is a fallacy.

As long as Intel is the premier supplier to Tier 1 vendors, such as Dell, Intel will remain king.



To: Elmer who wrote (163394)4/3/2002 2:50:50 PM
From: Ali Chen  Read Replies (2) | Respond to of 186894
 
"there is an industry standard benchmark suite out there. It's called SPEC. The only reason that AMD doesn't use it is because of how badly they compare."

Do not twist things. The SPEC benchmark heavily relies
on multi-pass compilation process with profile feedback
taken from the particular benchmark data example. Intel
managed to buy every compiler house that now works
for them doing narrow-focused optimizations. AMD has
lost this vital component of SPEC benchmarking.

If you apply the SPEC-Intel executables to another size
of data, the optimization will likely to fall apart.

The SPEC approach is fine for only handful
of researchers who live off solving few unique
computational problems, where they have control
over the source code.

In the consumer segment, you cannot find a user who
recompiles his Word Processor for every size of
a document he is working on. AMD does not use SPEC
because it is irrelevant to the segment they decided
to compete.

Of course, it was nice to have the performance crown
in all segments, but such is life, lose some, gain
some. They lost SPEC for now, but there are solid
excuses not to focus on this benchmark.
Maybe we should try iCOMP?

- Ali

P.S. By not responding and ignoring my posts you only
show to everyone that you are incapable to refute
any of my statements. Chicken?



To: Elmer who wrote (163394)4/3/2002 3:23:40 PM
From: Charles Gryba  Read Replies (1) | Respond to of 186894
 
Elmer, did you ever get to run my benchmark on your 1.8A?
I am really curious as to the NWoods real-world performance.

C