Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
Prices: Mainboards ABIT ASUS Chaintech Shuttle Soyo Tyan CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ Search Prices:

In the past, most cases of burned-out CPUs here (even though they were rare) were caused by improper seating of the CPU cooler. Likewise, most cases of spontaneous reboots we have come across could be resolved by reseating of the cooler with proper application of thermal interface material. In so far, a slanted CPU mount a priori raises some suspicions here but as they say, the proof is in the pudding, that is, as long as there is proper contact along the entire die surface, the entire thing may not matter that much since most high-quality thermal interface materials are insensitive to pressure differentials

We used a Taisol CGK760172 cooler because of its relatively mild pressure on the CPU (no point in going to brute force like in the case of the Thermalright SLK800 and cracking the processor). After mounting the CPU and HSF without using any thermal interface material, we used a flashlight on the far side of the CPU and looked "through" the space between the CPU and the HSF bottom to see whether there was any light coming through on top of the CPU die. If such a light slit existed, this would be clear evidence for insufficient contact between the die and the cooler.

   

Left: Low magnification overview. The flash had to be turned off (naturally) in order to capture anything. Right: close-up of the center area. Towards the right (the area where the CPU is raised) there is no detectable light shining through between the CPU die and the HSF. To the left, that is the far left corner of the die, a light-slit tapering off towards the right was clearly visible. Admittedly, the photographic documentation is not very clear here since it would require a much better equipment than the one at hand to show the full detail. Click for larger images.

The light slit test confirmed what we had been afraid of, using the speed strip changes the geometry of the CPU setup such that the cooler was pressed onto one corner of the CPU die rather than making full contact with the entire surface. In real life, some of this will be compensated for by the use of thermal interface materials. The big question in this case will be:

• a: How long will the TIM last before it forms pockets, and what temperature gradient will there be across the CPU diagonal?
• b: How well will the entire setup hold up under load.

We can't answer these questions since they will depend largely on the cooler used in the respective setup, some may follow the CPU tilt better than others, most likely, single-tab designs will be less problematic than others. Does the unlocking work? Sure, we have shown it already awhile ago. However, there is no chance I personally will fire up a CPU under these circumstances.

Conclusion

It is an interesting design that is used to unlock the multiplier and the inventor certainly deserves his brownie points for ingenuity but there are inherent risks associated with its use. I am not saying "don't use it!", this decision is up to every user and there is nothing I can or want to contribute other than showing some of the facts. The design itself can, however, be modified very easily by using a matrix covering all 462 pins, similar to the "Silent Serpent" from a few years ago. In that case, the entire CPU would be raised but remain level and other than a marginal increase in total pressure of the heatsink, no adverse side effects would occur.

Speedstrip as I would design it: It just takes a little bit more plastic but would solve all mechanical issues if it were done as shown above with the metal mask indicated by the red arrow.

Basically, at this point, what it comes down to is the "enter at your own risk", and Speed Strip will not be held responsible for anything that explicitly or implicity occurs from the use of the SSA-1 unlocking strip.

next page:    => more =>