dcsimg
RealTime IT News

Intel Patent Aims to Stop Overclocking

Overclockers beware! Intel is defending its property.

The Santa Clara, Calif.-based chip making giant now has a patented anti-overclocking technology designed to automatically shut down a system before an overworked processor burns it up.

The U.S. Patent and Trademark Office granted the No. 1 chipmaker patent #6,535,988 on March 18 for use in future Intel products. Which particular chips, however, remains to be seen

"We apply for many patents due to the amount of research we do. We have not made an announcement on any kind of implementation for this particular patent," Intel spokesperson George Alfs told internetnews.com.

In its patent summary with the USPTO, the company said its i386, i486, Celeron and Pentium processors are its most widely overclocked chips.

Overclocking (also known as speed margining) is a popular technique for eking out a little more performance from a system. In many cases, you can force your CPU to run faster than it was intended simply by setting a jumper on the motherboard. Overclocking does come with some risks, including over-heating, so experts say you should become familiar with all the pros and cons before you attempt it.

Entitled "System for detecting over-clocking uses a reference signal thereafter preventing over-clocking by reducing clock rate," the 30-point text outlines ways that Intel technology detects if a system is in danger and then takes appropriate action.

If overclocking is identified, Intel said its technology disables the operations of a computer system or significantly undermines key operations of a computer system.

The practice of overclocking is becoming a mainstream business. Scores of Web sites like the Overclocking Store, and Extreme Overclocking offer tips and sell products that help computer users go beyond a chip's designated speed. Even internetnews.com sister sites Sharkey Extreme and SysOpt.com have pages devoted to teaching people how to boost a system's performance.

The most common problem of overclocking relates to bit errors and data corruptions. Usually, chipsets and/or hardware components which need the system clock frequency for computing operations may incorrectly interpret electrical signals between "1" and "0" due to timing violations. More serious problems of over-clocking relate to advanced chipsets, which use a random number generator (RNG) for security applications such as cryptography, digital signatures, and protected communication protocols.

The other major problem, according to Intel, is that "unscrupulous" resellers and/or distributors may purchase less expensive processors that are rated at lower clock frequencies and then remark those processors at higher clock frequencies for resale at higher prices.

With so many people trying to overwork its chips, Intel said it had to take some type of internal action.

"A need exists for a more secure and mechanism for detecting and effectively deterring overclocking of a system clock signal so as to prevent resellers, distributors and/or end users from operating the processor at a clock frequency that is greater than a rated clock frequency," Intel said in their patent summary.

David Poisner, a component architect in Intel's Desktop Products Group (DPG) in Folsom, California, invented the technology. Poisner also has a patent for "Method For Recovering From Computer System Lockup Condition."