Chicago (IL) - Rumors, that Intel's new chipset 915/925 platforms do not offer much of a FSB overclocking margin, have been circulating for some time. As Tom's Hardware Guide was able to confirm, Intel went one step further: The chip giant adjusted its hardware to prevent any overclocking.

It is no secret that Intel was not quite happy that its past chipsets offered huge overclocking ranges which allowed industry and users to use technologies Intel had foreseen to provide at a later point of time. Most recently, the E7205 (Granite Bay) could be overclocked close to FSB 800 months ahead of Intel's 865/875 chipset launch.

Prenventing excessive overclocking so far was limited to cropping all unnecessary options such as additional multipliers. This time, Intel wants to be sure that this does not happen again. This time, the chip giant is not only removing functions that are considered dispensable, but the company is actively adjusting hardware in order to keep anyone from using features like the upcoming DDR2-667 and FSB 1066.

Intel's startegy is to integrate an overclocking limiter into the MCH chips: If the CPU clock exceeds the threshold that, according to our test results, exceeds ten percent of the specification, the required Phase Lock Loop (PLL) will - which is a simple way of throwing a spanner in the works, as it renders the system crashed.

Read details of the overclocking lock in our 915/925 launch article .

Related stories:
Intel launches Grantsdale and Alderwood chipsets, Pentium 4 5xx-series