Articles: CPU
 

Bookmark and Share

(94) 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 ]

Overclocking

Phenom overclocking tests is another interesting part of our today’s discussion. Although it was an engineering sample, the obtained results may serve as a preliminary reference indicating what we could expect from overclocking of quad-core AMD processors with B2 core stepping.

Overclocking experiments were performed in the same testbed as the performance tests. The processor was cooled with Scythe Infinity cooler. System stability was checked with Prime95 utility.

The default processor Vcore for our sample was 1.3V. Without pushing it any higher we got it to run stably at 2.8GHz.

Having increased the voltage to 1.45V, we managed to get our processor to work at 3.0GHz, however, the system stability was pretty questionable at first. Unfortunately, further voltage increase didn’t improve the stability, but we just couldn’t give in so easily, as we were very curious to see what the Phenom could do at 3.0GHz. Luckily, our processor proved capable of running flawlessly at this speed. As we found out later, it was failing because of insufficient second voltage that can be adjusted independently on Socket AM2+ mainboards. Once the voltage of the memory controller and HyperTransport bus was raised from the default value to 1.25V, the system passed all stability tests impeccably.

Of course, we couldn’t pass on checking out the 3.0GHz Phenom performance. Therefore, the diagrams in the next part of our review will contain not only the results for Phenom 9500, 9600, 9700 and 9900, but also the performance of the overclocked processor.

 
Pages: [ 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 ]

Discussion

Comments currently: 94
Discussion started: 12/20/07 11:29:54 AM
Latest comment: 05/20/08 12:39:59 AM

View comments

Add your Comment