Date: Thu, 29 Mar 2001 08:11:06 -0500 From: Bill Cohane <wcohane@banet.net> To: "Hartmann, O." <ohartman@klima.physik.uni-mainz.de> Cc: <freebsd-stable@FreeBSD.ORG>, <chad@DCFinc.com> Subject: Re: Overheated PIII in SMP system Message-ID: <5.0.2.1.2.20010329074136.026d62f8@mail.verizon.net> In-Reply-To: <Pine.BSF.4.33.0103291238430.3932-100000@klima.physik.uni-m ainz.de> References: <200103290112.SAA00301@freeway.dcfinc.com>
next in thread | previous in thread | raw e-mail | index | archive | help
At 05:46 03/29/01, Hartmann, O. wrote: >I do not think FreeBSD's behaviour is the cause for "overheating" the CPU, >and I do not care about what FreeBSD is doing while in an idle loop. >If the >system is stable, it is all right. > >In my case, something seems to be very wrong. I monitored now for >about two >days all parameters I can obtain by the "HEALTHD" daemon and I I did >several >changes, e.g. swapping both CPUs. Now the system runs "stable" (but >how long?). >I swapped the CPUs again and I realized, that my first observation, that >heat follows one specific CPU, was an err. Heat is located in SLOT 1 for >CPU #0. Obviously is the othe CPU not that sensitive to heat than the >other one, >but a measuremnt of 52 degrees Celsius on the heat sink is really to much! > >"Healthd" gives a lot of warnings to me due the fact that SLOT 1 delivers >a core voltage of 2.06 to 2.08 and not 2.05 as the other, second slot. I >think this is like an evidence to accuse the mainboard. > >This will getting changed today (ASUS P2B-D against ASUS CUV4X-D with two >new boxed 733B-Coppermines). Coppermines seems to be much "cooler" than >old KATMAI - and the risk of a dying system due old hardware is to >high ... Hello I'm rather curious how you are measuring the temperatures of the two processors. I was under the impression that the P2B-D (indeed the whole P2B series) was designed before the Pentium III came out and has no on board means of detecting the temperatures of the CPUs. Are you using special thermal probes attached to the processor cartridges (or heat sinks) and plugged into the motherboard? My point is that if you aren't using the add on temperature probes, how are you measuring the temperatures? (I am not aware of a way to monitor the thermal diode that's inside the PIII without modification to the motherboard socket or adaptor card.) Also, the Asus P2B-D (prior to motherboard revision 1.06, possibly even prior to late shipments of the 1.06...as marked by the characters "D03" showing on a white bar code label on the end ISA slot) have a hardware bug that (at least in MS Win2K) keeps one processor bus at 50% to 80% at all times. (Something about thousands of spurious interrupts being detected each second.) I have three of these boards running dual PIII-800...but they are all running Windows. (I have to use the MS software for my math research. Sorry about that.) All my boards show slight voltage fluctuations (one or two hundredths of a volt) and I think it's normal. (Note that I'm using PCP&C 400 or 450 Watt power supplies...which should provide good power to start out with.) Regards, Bill Cohane To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?5.0.2.1.2.20010329074136.026d62f8>