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 : Apple Inc.
AAPL 273.40-0.1%Dec 26 9:30 AM EST

 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext  
To: Robert Salasidis who wrote (789)1/6/1997 1:22:00 PM
From: Meghan Richards   of 213177
 
You are correct ... the MMX situation is not a bug.

The facts are as you stated ... the Pentium-MMX requires several
clock cycles when switching between floating point and MMX mode.
It takes longer to switch in one direction than the other (but I
forget which one is longer).

The number of clock cycles between the two modes will be considerably
less on the upcoming PPro releases. So this 'problem' goes away
shortly.

The 20% speed improvement in the Pentium-MMX comes about because
they increased the size of the L1 cache (not because of the additional
MMX instructions). I've heard performance numbers that are all over
the map when the software utilizes the new MMX instructions.
Report TOU ViolationShare This Post
 Public ReplyPrvt ReplyMark as Last ReadFilePrevious 10Next 10PreviousNext