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: Tenchusatsu who wrote (69887)8/25/1999 5:57:00 PM
From: Elmer  Respond to of 1575981
 
Ten, here's an article showing just how dumb the REGISTER can be!

"Cadence Celeron design has double the cache:

Sources close to Intel have suggested that the Mendocino design, which saved the Celeron from utter oblivion, has 256K cache designed in, not just the 128K cache, as widely advertised. According to the source, Cadence was called in after the original Celeron performed disastrously and designed the Mendocino core. The Intel source, based in Israel, said that the external design was good, but Intel did not want to implement everything all at once. That suggests that Intel may well enable the other 128K of trannies at a time best known to it. No one from either the Intel or Cadence press offices was available at press time. © "

EP



To: Tenchusatsu who wrote (69887)8/25/1999 7:19:00 PM
From: Ali Chen  Read Replies (2) | Respond to of 1575981
 
<4-way Profusion .. with eight clients. The 8-way ...with 12 clients.>
So the results are obviously "marketized":
the 4-way has 2 clients per processor,
the 8-way has only 1.5 clients.

With 32+ clients there is no performance difference:
zdnet.com

A system with 12 clients only at a time
is a joke, "mission critical" joke.

Anyway, thanks for courteous response
(not in DRBES sense :) :).