RAM clock cycle time

[DAW]

In order to make a more realistic comparison of RAM modules, I want to convert CL (CAS) latency to True Latency measured in nanoseconds. The formula is:

True Latency (ns) = Clock Cycle Time (ns) x Number of Clock Cycles (CL)

The problem is that I don't know if the Clock Cycle Time is calculated from other factors or if it is a constant for each given SPD speed (e.g. 2133 MHz,  2400MHz, etc). How do I determine the Clock Cycle Time in nanoseconds for any given RAM module?

Can someone please help?

[DaveLembke]

Why not just use benchmark results of RAM sticks as a means of knowing, as for the main board that the RAM is in and the CPU that has memory controller and other things also affect performance?

I have a Socket AM3+ board for example and got different RAM benchmark results between same system with an AMD Athlon II x4 620 2.6Ghz and a AMD FX-8350 x8 4.0Ghz. Both CPUs have their own integrated memory controllers, and both CPUs handled the RAM differently at the same FSB of 1333Mhz. The Athlon II x4 maxes out with 1333Mhz FSB support. The FX-8350 can take faster, but it was running at 1333 which is what the RAM was rated at. Got this 8GB 1333Mhz RAM 2 x 4GB for free from a dead system, but have since upgraded to 1600Mhz DDR3 16GB. the benchmark results between the Athlon II x4 620 2.6Ghz and the FX-8350 x8 4.0Ghz was almost 10% difference, where the RAM performed better in the faster FX-8350. Benchmark used was Passmark.

[Geek-9pm]

Good benchmark results emulate the way something would be used in the real world. A memory test that did random reads and writers will give results that differ from  long sequential reds.