Date: Mon, 28 Aug 2000 19:09:23 -0700 From: Kent Stewart <kstewart@urx.com> To: Mike Meyer <mwm@mired.org> Cc: Marc van Woerkom <marc.vanwoerkom@science-factory.com>, questions@FreeBSD.ORG Subject: Re: FreeBSD is being extremely slow.. Message-ID: <39AB1B53.91113A36@urx.com> References: <14762.54705.346152.495600@guru.mired.org> <39AAE9EC.DFD5E4E@urx.com> <14763.2428.985901.162062@guru.mired.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Mike Meyer wrote: > > Kent Stewart writes: > > Mike Meyer wrote: > > > Marc van Woerkom writes: > > > > > I refuse to support overclocking, please fix your system and > > > > > then repost if you continue to have problems. > > > > A wise decision. > > > > I was once tempted to overclock a P166 to 180 or somethig MHz. > > > > There were several weird errors due to overclocking that did never > > > > show up under W95 but only under FreeBSD at that time. > > > What's really wierd is that overclockers seldom go to even as much as > > > 10% more CPU. For anything but very long-running cpu-bound tasks > > > that's not enough to be noticeable! > > That isn't true. You go from a FSB of 66 to 100 and clock for clock > > that is a 1.5x gain. > > I've never heard of anyone doing that one before(*). The ones I see > are more like the one here (166 -> 180), which is less than 9%. One of the favorites was taking a Celeron 300a running it a 100MHz and getting 450 out of it. They were supposed to be good for 500Mhz. Another was running a 450 @ 133 yielding a 600. I've been told that is why the price of the 450 is still so high. I think it is safer to use one of the .18 micron cpu's instead of over clocking. When it came to cranking out setiathome wu's, a Celeron 300a@450 was still slower than a P-III 450. The Celeron system was running FreeBSD 3.4 and the other was W2K Server. FreeBSD runs wu's about 3% slower than W2K on the same system. I have two multi-boots and I see 29-30K second wu's on W2K and 33-34K second wu's on FreeBSD. The 3% was averaged across more than 100 wu's on each system for each OS. > > However, what I normally see are CPU speeds, which might be a > different ball of wax. If you go from 66 to 100 FSB with 4x cpu > multiplier and a 366MHz CPU, then the *CPU* clocked at 400MHz, which > is right at 10%. About 30 seconds after I hit the send button, I wanted to add a comment agreeing with what you said. When they raise the speed of the FSB, it is a real improvement but there are cards that don't work at those odd speeds. Going from 500 to 550 with a change of cpu's is 10% but because of the memory funnel, you won't see a %10 increase in through put. Increase the FSB and you can. Is a 10% increase in through put worth dropping the warranty, I doubt it unless you think of the cpu as a throw away. > > I'm not into this stuff. The damn things are flaky enough without > going out of my way to make them worse. Murphy sits at my shoulder just waiting for a chance :). > > <mike > > *) The exceptions are the guys doing liquid-cooled systems, and > getting 2 or 3x. On the other hand, they admit they're doing it for > hack value, and are spending more on the system than it would have > cost to buy a system running at the resulting speed. One of the prettiest computer's I ever used was a Cray 2 and it was totally immersed in an inert fluorocarbon liquid. You would see bubbles of gas from the section of the computer that was being used. The other technique is the Peltier (?) (refrigeration style) but if you let the cpu go to sleep, you will drop it off the bottom of the temperature range and basically freeze it. One of the Cray SE's that I met, was called back to Saudi Arabia because the chiller died on an early Cray 1/S and the alarm failed, which resulted in the soldier melting, dropping surface mounts. They actually made it run again because flying in replacement cards were too expensive and it was cheaper to fix the broken cards. Kent -- Kent Stewart Richland, WA mailto:kbstew99@hotmail.com http://kstewart.urx.com/kstewart/index.html FreeBSD News http://daily.daemonnews.org/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?39AB1B53.91113A36>