bit-tech.net

First Look: Corsair CM3X1024-1333C9DHX

Published on 28th May 2007 by Richard Swinburne

Xvid Encoding:

We tested video encoding performance using VirtualDub-MPEG version 1.6.15 and a multi-threaded version of the Xvid codec, along with the LAME MT MP3 encoder for encoding audio. We did a two-pass encode of a 15-minute 276MB digital TV recording with a target file size of 100MB.

DDR3-800MHz 5-4-4-8
DDR3-1066MHz 7-7-7-15
DDR3-1333MHz 9-9-9-24
DDR3-1520MHz 9-9-9-24
DDR2-800MHz 3-4-4-9
DDR2-1066MHz 5-5-4-12

100

200

300

400

500

600

700

Time in Seconds (lower is better)

DDR3 is finally a winner, with 1333MHz+ out performing DDR2 at both speeds we've tested here. Again, 1066MHz DDR3 isn't really worth considering, you might as well invest in DDR2 at the same clock speed.

MP3 Encoding:

We used LAME MT for our audio encoding test - it's the multi-threaded version of the popular LAME MP3 encoder. We ran tests with both Intel's and Microsoft's compilers - naturally, the Intel compiler resulted in some performance increases on Intel's processors. We converted all 18 tracks from Moby's popular Play album to a 192kbps variable bit-rate MP3.

DDR3-800MHz 5-4-4-8
DDR3-1066MHz 7-7-7-15
DDR3-1333MHz 9-9-9-24
DDR3-1520MHz 9-9-9-24
DDR2-800MHz 3-4-4-9
DDR2-1066MHz 5-5-4-12

100

125

Time in Seconds (lower is better)

MP3 encoding only benefits slightly from memory bandwidth, and despite the (continuous!) anomalous result, DDR2 and DDR3 perform within just a couple of seconds of each other.