Test Results – Synthetic
PCMark05, like all of Futuremark’s benchmark suites, is made up of a series of tests designed to stress specific subsystems, such as drive, processor, or memory performance. SimHQ decided to use just CPU and memory among the individual test suites, the latter because some of the individual memory tests stress processor cache.
The collective CPU score shows very little difference between the two quad cores, though both parts scored higher than the dual core as expected due to the threaded design of these tests. The memory tests allowed the QX9650 to nudge out a small win against the older QX6850, with the two last generation parts producing very similar scores.
3DMark06’s two CPU tests were used since both are heavily threaded and allow multi-core CPUs to shine.
Intel’s new CPU again barely outperformed the older QX6850 in these synthetic tests, with the dual core part naturally losing by a substantial margin.
Windows Media Encoder 9 was used for SimHQ’s content creation test. The application is a free download from Microsoft and is heavily threaded. Falcon 4: Allied Force’s intro movie, a 91MB .avi file, was converted into a .wmv file with high definition video and audio settings and the total time required by each processor to convert the file recorded. Shorter conversion times represent a better score for this particular test.
The QX9650 performed the file conversion 25 seconds faster than the older quad core, with both quads showing a significant finish time compared to the dual-core E6850.
More of an environmental aspect of testing rather than synthetic performance, total system power consumption was measured using an Extech power meter, supplied by AMD during Intel’s fiery Prescott years. The recorded wattage was generated for each processor by running the full series of tests for PCMark05. Each score represents average wattage of the entire test system (sans display). The idle power draw was not recorded due to the ASUS Maximus Formula not recognizing the QX9650 and therefore not allowing power management to properly reduce the processor’s clock speed while idle (CPU-Z was used to ensure the CPU was running correctly).
With the QX9650’s unexpected number of 246 watts, showing less power being pulled from the wall than the tested dual core, we re-ran the test and double-checked to confirm that the new part was running correctly. Yet the same result occurred again, showing the new CPU pulling almost 100 watts less than the older quad and less even than a similarly clocked dual core. We have to attribute this eye-opening difference to Intel’s high-k/metal gate process and its reduction of leakage.