Date: Wed, 25 Feb 1998 14:21:14 -0600 (CST) From: Kevin Day <toasty@home.dragondata.com> To: mika@cs.caltech.edu (Mika Nystrom) Cc: freebsd@atipa.com, adoane@eagle.ais.net, freebsd-smp@FreeBSD.ORG Subject: Re: Dual proc PII MB of choice? Message-ID: <199802252021.OAA29766@home.dragondata.com> In-Reply-To: <199802251955.LAA16742@stun4p.cs.caltech.edu> from Mika Nystrom at "Feb 25, 98 11:55:05 am"
next in thread | previous in thread | raw e-mail | index | archive | help
> This is interesting. I would tend to think that if the CPU in question > exists in a version that supports the higher clock speed, and if the > same cooling is used as on a standard installation at the higher > clock speed to dissipate the heat that is generated at that speed, > the only problems you might see from overclocking are soft errors. > This is because the only Bad Things that really increase on the > CPU when you up the clock speed are current draw and power consumption, > and if the chip in question exists for the higher clock speed, it is > obviously designed to handle that. The reason it is sold as a slower > part is because of variations in manufacturing that makes the > transistors slightly slower, variations that may be offset by > superior cooling or fewer heat sources in the vicinity of the > CPU, allowing a lower operating temperature than called for in the spec, > and consequently higher clock speed. > > Mika AFAIK, you are partially corrent, but there *are* major differences in chips, even in the same family. The pentium has gone through 3 die shrinks, and several changes along the way. A P/200 chip is completely different, than say a P/100. But, within the same revision, you are correct. They mass produce chips, test them, the ones that run at a higher spec without a problem are sold as such. If not, they lower the speed at test again. (At least, this is they way they were doing it before, if this has changed, let me know) Some older Pentiums don't have some of the divisor pins hooked to anything, even. :) Kevin To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-smp" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199802252021.OAA29766>