From owner-freebsd-hackers Fri Mar 15 12:34:53 2002 Delivered-To: freebsd-hackers@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by hub.freebsd.org (Postfix) with ESMTP id 430EB37B402; Fri, 15 Mar 2002 12:34:50 -0800 (PST) Received: (from dillon@localhost) by apollo.backplane.com (8.11.6/8.9.1) id g2FKYma77667; Fri, 15 Mar 2002 12:34:48 -0800 (PST) (envelope-from dillon) Date: Fri, 15 Mar 2002 12:34:48 -0800 (PST) From: Matthew Dillon Message-Id: <200203152034.g2FKYma77667@apollo.backplane.com> To: Terry Lambert Cc: Doug White , Jordan Hubbard , Michael Smith , hackers@FreeBSD.ORG Subject: Re: Interesting sysctl variables in Mac OS X with hw info References: <20020315120943.T71602-100000@resnet.uoregon.edu> <3C9257F2.5C0EE18F@mindspring.com> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG :Doug White wrote: :> I've been asked several times about how to get CPU speed information for :> inventory purposes. :> :> People would really like the speed number printed on the chip, not what :> it's currently running at, if that's retrievable :) : :Can't mask the speed number. : :Chips with a lower printed number are just chips that failed :testing at higher clock rates. Sometimes, they don't even :fail, if they have a big demand swing. 8-). : :I guess they could laser it out... : :If Intel really didn't want overclocking to happen, they would :put the clock on board the CPU, and make it an output, not an :input... 8-) 8-). : :-- Terry Actually, Intel did just that on their Celerons a little while after they were first introduced. People realized that Intel was stamping chips that tested at higher clock rates with lower clock labels in order to keep their lower-rated distribution pipelines full. That created a major overclocking craze on the Celeron line as well as no small amount of grey-market relabeling of chips. This also led to a certain percentage of grey-market relabeled chips failing since not all the lower-rated chips passed the higher rated tests. Enough did, however, and Intel actually began losing market share in their higher-rated chips to their lower-rated chips. Oops! To combat this Intel actually started either lasering or fuse-blowing the PLLs on the chips to make them match their labels and prevent them from being too seriously overclocked. I don't know if they bother to do it anymore, though, as the chip speed testing gap is now back to normal... packaging and heat dissipation has become important again. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message