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.
Politics : Formerly About Advanced Micro Devices -- Ignore unavailable to you. Want to Upgrade?


To: Jim McMannis who wrote (42378)11/29/1998 10:04:00 PM
From: Yousef  Read Replies (3) | Respond to of 1572777
 
Jim,

Re: "Yea boy...I do this every time... Who is Anand and ZD trying to kid?"

BTW Jim, I tend to run about 3-4 applications at the same time (at work).
These are typically Netscape, MSWord, an E-mail app, and PowerPoint. Most
people in business run in this type of mode ... Obviously most "old guys"
in Florida, don't. <ggg>

Make it So,
Yousef



To: Jim McMannis who wrote (42378)11/29/1998 10:21:00 PM
From: Brian Hutcheson  Read Replies (2) | Respond to of 1572777
 
Jim , re. lost it
I agree , he is making strange comments i.e. k6-2 400 price being kept under Celeron 300a , even the K6-2 350 is $125 and the Celeron 300a is $92 on Pricewatch . Does this guy have a clue ?
Brian
PS I wonder if these self professed tech experts who recommend buying a Celeron 300a and overclocking it to 450mhz will agree to honouring the warranty when Intel refuses to do so . It seem that as long as the chip runs for the period of their test they are satisfied , but the life of those chips subjected to such overclocking will be seriously shortened .



To: Jim McMannis who wrote (42378)11/29/1998 11:03:00 PM
From: Maxwell  Read Replies (1) | Respond to of 1572777
 
Jim:

Frankly I like the Winbench 3D Benchmark from AMD which shows that a K6-2-300 matches the PII-400MHz and a K6-2-333 blows away the PII-450.
That benchmarks show the power of K6-2 in 3D ONLY and that is all it matters these days. I give Intel CPU a few % over K6-2 in 2D applications and Winstone 99 and take the 20%+ in 3D over PII.

Maxwell



To: Jim McMannis who wrote (42378)11/30/1998 3:04:00 AM
From: Scumbria  Read Replies (1) | Respond to of 1572777
 
Jim,

Some people have been turning up the heat on Anand and he defends Winstone 99. I honestly think the guy has lost it...

Some of Anand's explanations indicate that he doesn't have a very thorough understanding of his subject. His comment "the real benefit of write allocate is visible in writes to the write allocated cache line after the initial write allocate." is mostly incorrect.

The real value of write allocate is that it increases the percentage of read hits. Cacheable data that is written (in a single processor system) will almost always be read back. If it is not read back, why did we bother to write it? Turning on write allocate can reduce the cache read miss percentage by as much as 20%.

Scumbria