Memory speed/latency and Sandy Bridge performance

[996GT2]

Do Memory timings and speed affect the performance of Sandy Bridge much? For example, if I was using DDR3-1600 @ 7-7-7 instead of my current DDR3-1333 @ 9-9-9, would overall performance show a noticeable increase?

 

[RobDickinson]

I guess it depends but yeah SB is loks latancy dependant, seems to have plenty of bandwidth to burn whatever.

 

[RussianSensation]

Do Memory timings and speed affect the performance of Sandy Bridge much? For example, if I was using DDR3-1600 @ 7-7-7 instead of my current DDR3-1333 @ 9-9-9, would overall performance show a noticeable increase?

If you were assembling a system from scratch, I would recommend DDR3-1600. However, I wouldn't spend any $ upgrading from DDR3-1333 if you already have that.

Outside of syntethic benches (i.e., Sisoftware sandra), the performance difference is real world apps. is minimal.
http://www.bit-tech.net/hardware/memory/2011/01/11/the-best-memory-for-sandn-bridge/1