How do you evaluate how fast a stick of ram is? Many people look at the raw frequency, others CAS latency, and others transfer rate.
As this article outlines, you need to consider both speed AND latency.
For speed (MT/s), this is fairly straightforward, higher is better. But that is only half the equation. The latency of access also matters. Latency is reported as the CAS time – but that is reporting only the total number of clock cycles before access. This can be misleading by itself, because it’s just a number of cycles – not how LONG that time is.
To get an apples-to-apples comparison on latency, we need to look at latency in terms of nanoseconds – not clock cycles. To calculate a module’s latency in nanoseconds, simple multiply clock cycle duration by the total number of clock cycles.
latency (ns) = clock cycle time (ns) x number of clock cycles
Otherwise, you may not be getting much of an improvement at all. Youtube testers often find little difference between different sticks of ram because they might be focusing on faster CAS latency but not doing the whole equation. The table below shows some examples and why you need to pay attention to both speed and CAS timing.
|Technology||Speed (MT/s)||Clock Cycle Time(ns)||CAS Latency||Latency (ns)|
So what is the recommendation?
- Step 1: Identify the highest memory speed supported by both your processor and motherboard (including overclocking profiles).
- Step 2: Select the lowest latency memory that fits within your budget at that speed, remembering that a superior (i.e. lower) latency means superior system performance.