Testing Thermal Throttling in Pentium 4 CPUs with Northwood and Prescott cores

[apoppin]

Digit-Life Article

i sure hope this isn't a repost . . . I did a 'search' in GH and also don't remember this discussed (the article is only 10 days old) but am SURprised it wasn't as it is so INTeresting:

The issue of diminishing Pentium 4 CPU performance at overheating has been widely discussed (see, for example Intel Pentium 4 1.6A GHz Overclocking, Thermal Monitor). Still, there are all sorts of myths about the Thermal Monitor and Thermal Throttling technology. That is why we decided to refresh our practical knowledge in this field (dropping the theoretical aspect which is abundant in the above-mentioned articles). For this purpose, we conducted a very simple experiment on two top Pentium 4 3.2 GHz CPUs, one with a Northwood core and the other with a Prescott. With the CPU loaded at 100 percent, we powered off the cooler fan and watched the temperature rise and the performance decrease as time went by. This article gives an account of the experiment and its results.

I learned something.

(don't miss their conclusions - evidently there are a few MISconceptions held here) :Q

 

[apoppin]

Here are the Conclusions:

Now let us sum up the results of our experiments.

1. The Thermal Monitor/Thermal Throttling mechanism is present in Pentium 4 CPUs with Northwood and Prescott cores and functions precisely in line with the specs.
2. Evidently, particular values of time intervals and temperatures measured can hardly be applied to any system. This is due to hardware differences (CPU frequency, cooler and thermal grease pair, motherboard model, ambient temperature, opened/closed case). However, the mechanism of throttling functioning and throttling patterns have been experimentally established and confirmed.
3. The hypothesis that throttling can prevent any system in any circumstances from shutdown didn't bear the test. At standard rate of performance loss (about 50 percent), throttling is not enough to prevent a CPU overheating which leads to an automatic shutdown. On the other hand, now that we know the temperature threshold we can focus on measures that wouldn't let temperatures reach it. It can be efficient ventilation, case coolers, more aggressive throttling options in BIOS Setup, etc.
4. It proved very difficult - if possible at all - to burn a Pentium 4 CPU by stopping the cooler. Of course, providing the heatsink with thermal grease is correctly installed. At least we didn't succeed in it even though sometimes the temperature nearly reached 100°C and kept on for quite a while.
5. A Northwood core has milder thermal conditions than Prescott. However, the manufacturer (Intel) doesn't try to hide this fact, underlining instead a high importance of cooling system and its correct installation. In other words, working at a risky, close-to-throttling rate is declared normal for Prescott, and cooling system is charged with an overheating-prevention task.
6. On the whole, the Thermal Monitor/Thermal Throttling technology deserves to be praised, even taken separately as a bare technology. We liked the very idea of trying to stop or slow down a dangerous temperature rise in the first place, and only then considering a shutdown option. We believe it is much better than just stopping a CPU when it has reached a certain temperature

inteesting, no?

 

[Abhi]

Certainly interesting...