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: dybdahl who wrote (119179)11/25/2000 1:45:47 PM
From: Tenchusatsu  Read Replies (3) | Respond to of 186894
 
Dybdahl, <There are many types of calculations out there, where you can choose between a very precise Floating Point version and an approximate SSE2 version.>

I thought SSE2 supported full IEEE-1394 floating-point. You must be thinking of 3DNow, which uses "approximate" floating-point because of its focus on games and other non-scientific applications.

<Toms Hardware chose a DVD encoding benchmark for testing P4 against Athlon, and chose an SSE2 method to do this first, but after some external input he changed his mind to use Floating Point, because the results were better.>

Wrong. Tom originally used MMX encoding, not SSE2. Here's the link to the e-mail sent by one of Tom's readers:

tomshardware.com

I think Tom will change his tone once his beloved MPEG4 decoders start supporting SSE2, along with other software. Right now, all the AMDroids can do is deny that anyone will ever support SSE2.

Tenchusatsu



To: dybdahl who wrote (119179)11/25/2000 2:19:42 PM
From: Paul Engel  Read Replies (2) | Respond to of 186894
 
DieBall - Re: "Tom thinks that nobody can justify to buy a P4, because no market needs extreme SSE2 performance on an otherwise slow processor. Has anybody examples that shows he is wrong? "

In case you are terminally confused, the Pentium 4 has been "out" for only 5 days.

Tom - don't forget - is some diesenfranchised dork in Germany that has the influence of a mosquito on the PC buyng public.

Re: "Toms Hardware chose a DVD encoding benchmark for testing P4 against Athlon, and chose an SSE2 method to do this first, but after some external input he changed his mind to use Floating Point, because the results were better."

Precisely - the results were better for Tom's agenda.

And Tom's agenda - to bash Intel - is an old phenomenon that most people recognize and laugh at - or just ignore.

People buy AMD CPU/PCs for one of two reasons:

1. They are not Intel
2. They are usually cheap - and they want CHEAP (and they get it).

After 31 years, AMD has a modest 17% market share - and Intel's current product line up looks more than adequate to maintain or decrease AMD at the 17% level.

You might saunter in to a retail PC store during the Christmas buying season and watch consumers buying patterns - and you will most certainly spot the dominant trend to buying mobile and notebook PCs.

And 98% of these have Intel Inside.

Paul