Although badge is mostly right, timings can make a difference for two main reasons:
1) Throughput (MHz) is limited by the FSB speed, since all data to/from memory has to go through the FSB. Current Intel C2D CPUs mostly have a standard 1066MHz FSB data transfer rate, which can be saturated by a pair of DDR2-533 DIMMs running in dual-channel mode.
2) Not all memory accesses are sequential. Many of the latency values have a bigger effect on non-sequential accesses.
Here's some data so you can decide for yourself what's best:
http://www.madshrimps.be/?action=getarticle&number=1&artpage=1962&articID=472
2) Not all memory accesses are sequential. Many of the latency values have a bigger effect on non-sequential accesses.
OP's question referred to the idea that RAM (PC3200) performing at 3.2 GB/sec data transfer rate could outperform another RAM (PC10600)performing at 10.6 GB/sec data transfer rate 'because' the first one is CAS 2 while the latter is CAS 9. I pointed out the difference in the speed at which data is transfered.
This information is a bit dated (and certainly simple and straightforward), but it may shed light on the question of performance difference between the varying CAS latencies of RAM.
What is CAS latency?
CAS latency (also referred to as latency) is the amount of time it takes for your memory to respond to a command. Specifically, it is the length of time between memory receiving a command to read data, and the first piece of data being sent from the memory.
What is the performance difference between CL2 and CL3?
CL2 DIMMS process data a little quicker than CL3 DIMMS in that you have to wait one less clock cycle for the initial data. However, after the first piece of data is processed, the rest of the data is processed at equal speeds. Latency only affects the initial burst of data. Once data starts flowing, there is no effect.
Although this is very basic defination, I just hope this helps in answering OP's question.