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: Tenchusatsu who wrote (119447)11/28/2000 2:06:36 PM
From: Dan3  Read Replies (2) | Respond to of 186894
 
Re: As long as current applications don't get replaced by new ones, Athlon is still ahead of Pentium 4 in the majority of situations [from TH]

I'd still call it pretty much a performance tie between 1.5 P4 Dual Channel RDRAM and 1.2 Athlon Single Channel DDR.

Now let's discuss manufacturing costs...

Dan



To: Tenchusatsu who wrote (119447)11/28/2000 5:26:21 PM
From: dybdahl  Read Replies (1) | Respond to of 186894
 
But he isn't right. On Linux, it has been common for a long time, that the kernel and the gzip library are compiled for the processor generation, that is in use, and even though Red Hat Linux is 386 compatible, it is Pentium optimized. Some distributions even go further and deliver code that is not 386 compatible in order to gain extra performance.

Often, performance critical modules like the Apache webserver, are also re-compiled on the target computer, sometimes in order to gain performance, sometimes just because the system administrator wants a module compiled into the webserver instead of having the module linked to the webserver application.

I don't know what Microsoft does, but since you cannot insert an extra processor in a dual-CPU motherboard with Windows NT, the kernel for dual-CPU and single-CPU must be different, and hence compiled for the CPU configuration.

I'm not sure how processor specific the Linux compiler (gcc) optimizes the code, but saying that compiled code never gets optimized for a specific processor in the real world is plain wrong.

But as long as we buy Windows Me and Windows 2000 computers, he got a point.



To: Tenchusatsu who wrote (119447)11/28/2000 8:59:06 PM
From: Paul Engel  Read Replies (1) | Respond to of 186894
 
Tenchusatsu - Re: 'Another recount - Tom's fourth article on Pentium 4:'

Yet another confirmation that Tom Uberclockermeister os nothing more than a hack with a bunch of equipment lying around.

Tom will always come up with some Benchmark - out of millions - that will support whatever agenda he has to put forth.

Paul