CAS Latency is all the rage nowadays because of all the attention on high performance RAM from OCZ, Corsair, and Mushkin.
What is CAS Latency?
CAS stands for
Column
Access
Strobe (conversely, RAS stands for Row Access Strobe) and CAS Latency is the amount of delay that you have to wait between the CAS signal and the data being pushed from the RAM pins. CAS2 latency essentially means we have a 2 cycle delay, and CAS3 has a 3 cycle delay: hence the lower the CAS, the lower the latency, and the better your performance.
The sequence of events is something like this:
- The "row" of the memory address is accessed and RAS is activated
- RAS-to-CAS delay
- The "column" of the memory address is access and CAS is activated
- CAS delay
- the data is pushed to the RAM pins
Is CAS Latency a HUGE factor in performance?
Memory timings in general account for a performance difference. The proper memory timings can result in as much as a 2-4% difference in application performance, but CAS settings alone will probably only account for an even smaller performance increase.
CAS is not always an important factor either: different applications will have different memory requirements and CAS only comes into play at one stage of the memory access: you still have to worry about the RAS-to-CAS delay. Also, your CAS will only be effective for memory intensive situations or situations where memory is accessed a lot.
In consideration for whether you should choose CAS2 or CAS3 DRAM, I'd suggest going for CAS2 whenever possible (ie: when the cost is about the same). The performance difference is not huge but it's still there. Also, even if you have RAM that's not rated for CAS2 operation, you *might* be able to try it out. We've known many cases where CAS can often be tweaked a little with no ill effects.
Always check your memory stability using
memtest86 when tweaking your RAM!
contributed by:
| Is this answer useful? Yes No
So far 420 of 678 person(s) think this answer is useful.
|
