Sign in with
Sign up | Sign in

Benchmark Results: 1080p WMV With VirtualDub And DivX

Tuning Cool'n'Quiet: Maximize Power And Performance, Part 2
By

For DivX encoding, we’ve selected the unconstrained profile and used the highest quality possible. The encoding was done with rate control set to one pass and a fixed quantizer of one. Enhanced multi-threading and Experimental SSE2/4 was enabled; everything else was set at default values.

Like any good video encoding app, the DivX codec can fully utilize more than one core.

It looks like DivX doesn't really like the Phenom II X3 710. It's not a problem with power management; it's the number of cores. DivX and VirtualDub don’t seem to know how to fully utilize all three cores at the same time. So, having that additional core is a liability here. There is a solution--simply disable the third core or force core affinity for VirtualDub to just two cores.

Performance-wise, there is hardly any difference between running with and without power management. Taking twice as long to finish, the Phenom II X3 710 cannot even came close to the Phenom II X4s. If you want to choose the best bang for the buck processor, grab the Athlon II X2 250.

Ask a Category Expert

Create a new thread in the UK Article comments forum about this subject

Example: Notebook, Android, SSD hard drive

Display 2 comments.
This thread is closed for comments
  • 0 Hide
    Anonymous , 8 December 2009 05:19
    crystalcpuid is a better alternative to AMD's CNQ. Multiplier management allows a wide range of multplier/voltages to be selected. A bit of testing and you can use really low voltages at only slightly lowered multiplier settings. Power consumption is related to the square of the voltage so low voltage give massive power savings. AMD CNQ uses very conservative CPU voltages. You dont need to stick to that ridiculously low idle frequency either and its latency penalties.
  • 0 Hide
    pertshire , 13 March 2010 22:34
    Great article. Really open my eyes that quads can actually be more energy efficient that dual cores (On application that ends of course)