Tom's Guide | Tom's Hardware | Tom's Games
![]() |
![]() |
![]() |
Abit and Asus came out with new BIOS yesterday and today ...... lowering CPU temps by more than a few Degrees C.
Let me tell you when Intel demands they deliver. Sorry for the Heatsink makers with the temp hyping by boards makers so that they could sell heatsinks.
For months after the IC7 series board released, users have been complaining about CPU temp being too high as reported by BIOS (check Abit forum. Nothing was done until the Prescott came out with reviewers blooding the review sites with 70C temp.
After loading the lastest P4P800 deluxe bios the CPU temp dropped another 3 C. I'll see how far the IC7 will drop.... I'd expect 10C.

you sure the temps actually went down or the sensor just giving you false reading - or they're just fixing temp readings from previous bios

My my, what an awful lot of hassle just to get a CPU stable ;-)
I don't suffer from insanity, I embrace it.

I don't understand: if the CPU dissipates the same amount of heat and the heat removal is the same, how do the boards report lower temps by the new BIOS?

"Let me tell you when Intel demands they deliver"
Deliver a BIOS that's been reprogrammed to lower the temp gauge by 5 degrees so it looks like the Prescott is approaching a normal temperature.
What a cynic.............
I don't suffer from insanity, I embrace it.

P4 is using diode and computation in the BIOS codes. You are correct on you assumption. Intel specs requires the measurement of CPU temp at the center of the heat spreader where it is impossible to put a sensor. So, they do some computation to calculate the approximate temperature. This is now, I believe is used in the A64 chip also.
99.9% of boards are made in Taiwan and 99.9% of heatsinks are also made in Taiwan (or China). So, if the board reports sligtly higher temp, you rush to buy a "better" heatink. They get all your money plus tips which is the heatsink.

<<what an awful lot of hassle just to get a CPU stable ;-)>>
Hooner;
There is no stablility issue here. While with all the temps talk, the systems are stable.
This is CPU overclocking forum.<<What a cynic.............>>
You are slow but you finally got the message.

In the old BIOS, they must have a way to calibrate the calculated temp to actual temp using some measured data and I bet the temps are pretty close (with some exceptions in some Asus boards). Otherwise what's the point to report the temp in the first place? Now after a BIOS upgrade, suddenly the CPU temp drops a few degrees. Do they have measured data to back it up? If yes, does that mean the temp on non-Prescott chips reported by the old BIOS are not correct? If not, what's the point for this suddenly low temp for the Prescott?

TXH
1)That's why you can't find any temp reporting from Dell, HP or Compaq, Emachine, Soney etc.
2)Like you, I was led to believe the same years ago. Until you have put together 10 to 20 systems and done a few overclockings.
3)As I said before, go to Abit-USA.com forum, you would read users having temp reporting issues from the same model board. Yet, you also read enough the "good" temps from P4P800 and P4C800.The best is to get a good external temperature probe (not very expensive), you'll be into a big surprise.
You decide or what I've detail here:
http://www.home.earthlink.net/~lazyman/data/2.E.htm

Well, despite my moanings I think that's an excellent report, kudos to you........
I don't suffer from insanity, I embrace it.

My prescott 2.8 E runs at around 50c, is stable, fast and can overclock to about 3.2 without a sniff of a problem. I use the boards core cell chip to oc the system. All components are stock. It benchmarks better than Athlon 64s compared on the futuremark website.
MSI 865 neo 2PE
Crucial 512mb DDR400
Geforce FX5200
Maxtor 80 gig hdd SATA
400w PSU
Case with a couple of fansWhats all the fuss about, buy one and see for yourself!

![]() |
![]() |
![]() |

This post is quite old and has been locked from receiving new replies. Please create a new posting instead.
| Ads by Google |