Latency has ceased making a difference years ago. And frequency over 1600 Mhz yields very little benefit.
Latency only matters if your workload won't fit in processor cache and also is not easily buffered in processor cache.
Also, make sure you calculate latency in nanoseconds instead of in CAS.
How does that translate to gaming, and I'm just going by retail specifications, specifically this new set, it's about 60 dollars cheaper than an equivalent corsair set with CAS 9 instead of 11, all else being equal.
http://www.newegg.com/Product/Produc...82E16820233528
Did not test different latency settings, but it might be prudent to watch out for memory speed settings :
Used the same chip, 4.4GHz @ 1.2V (fixed, not adaptive), Antec 920, IC Diamond, Crucial Ballistix Elite 1866 9-9-9-27-2T 1.5V
All runs were using the XMP timings (fixed at 9-9-9-27-2T) - only the speed and voltage were changed. Verified to be error free in memtest86+ 5 RC1.
Highest core temperature during a 25-pass IBT run is reported, averaged across 2 runs.
1333 - 1.4V : 75C
1333 - 1.5V : 75C
1866 - 1.4V : 79C
1866 - 1.5V : 79C
Memory voltage doesn't seem to make a difference, but speed definitely does.
Did not test different latency settings, but it might be prudent to watch out for memory speed settings :
Used the same chip, 4.4GHz @ 1.2V (fixed, not adaptive), Antec 920, IC Diamond, Crucial Ballistix Elite 1866 9-9-9-27-2T 1.5V
All runs were using the XMP timings (fixed at 9-9-9-27-2T) - only the speed and voltage were changed. Verified to be error free in memtest86+ 5 RC1.
Highest core temperature during a 25-pass IBT run is reported, averaged across 2 runs.
1333 - 1.4V : 75C
1333 - 1.5V : 75C
1866 - 1.4V : 79C
1866 - 1.5V : 79C
Memory voltage doesn't seem to make a difference, but speed definitely does.
Yup, forgot to add a better description to my original post.I assume you mean in terms of temperature?
Frequency is more important for performance than memory sub settings with DDR3..
I was reading an interesting thread over at Overclock.net about the impact of memory speeds on gaming:
Thread
The OP claims that while FPS may not necessarily increase, gameplay is noticeable smoother going from DDR3-1600 to DDR3-2133, with higher minimum frame rates and less texture pop in, particularly in Far Cry 3.
I'll be getting my 3930K tomorrow so I'll be testing out my new 16GB DDR3-2133 kit in Far Cry 3, because I know that game had quite a bit of texture pop in due to the craptastic engine and the sheer size of the sandbox.
I dont really believe everything this guy is claiming especially about it feeling smoother and getting no pop in with faster memory. I think if that was really happening then we would have more evidence of that from other users.Frequency is more important for performance than memory sub settings with DDR3..
I was reading an interesting thread over at Overclock.net about the impact of memory speeds on gaming:
Thread
The OP claims that while FPS may not necessarily increase, gameplay is noticeable smoother going from DDR3-1600 to DDR3-2133, with higher minimum frame rates and less texture pop in, particularly in Far Cry 3.
I'll be getting my 3930K tomorrow so I'll be testing out my new 16GB DDR3-2133 kit in Far Cry 3, because I know that game had quite a bit of texture pop in due to the craptastic engine and the sheer size of the sandbox.
CAS Latency means nothing with the Memory speed to determine the real Latency, which is Latency in Nanoseconds of the memory.
I dont really believe everything this guy is claiming especially about it feeling smoother and getting no pop in with faster memory. I think if that was really happening then we would have more evidence of that from other users.
The biggest thing with FC3 is the increase in smoothness. The slower memory feels choppy at times, such as when you get into a firefight. The faster memory never has this problem. Also, with 2133 and above, I get zero texture "pop in", yet it's present with lower memory speeds.
hmm I have not really noticed any pop in with Far Cry 3 but I will pay more attention next time I fire it up. I have only played about 2 hours of the game so try and test it towards the beginning so I can see if I have similar pop in.I'll be testing it tomorrow for sure, as I have a hard copy of Far Cry 3 so I won't have to download it.. While Far Cry 3 is a good game, the amount of texture pop in is unacceptable to me; and this is on ultra settings. Doing a search on the internet reveals this is a common issue that occurs regardless of how much VRAM your GPU has.
Theoretically though, faster system memory could conceivably help with texture pop in, as the GPU uses some of the system memory to store textures and what not.
It's noticeable when you're driving around the island at high speed, or flying in the hang glider.
I've always aimed to get frequencies and timings that give me the lowest real latency (in nanoseconds).
I will not significantly sacrifice the real latency for frequency, but that's just me.
I will also not spend significantly more money on top-of-the-line RAM. The performance benefit is not there. I try to find the most reasonably priced "high-end" RAM.
It's not like the old days where RAM speed could limit your CPU overclock (limits FSB).
Doesn't CAS correspond to latency? How do you measure the "real" latency?
(CAS / Frequency (MHz)) × 1000 = X ns
He's talking about units of latency.
CAS "units" are cycles. These units have a variable amount of time depending on the frequency.
He's pointing out that time is the important latency measurement when comparing memory of different speeds.
example:
1600 MHz at CAS 9 = 5.625
2400 MHz at CAS 12 = 5.000
so in this case CAS 12 is lower latency than CAS 9.
He's talking about units of latency.
CAS "units" are cycles. These units have a variable amount of time depending on the frequency.
He's pointing out that time is the important latency measurement when comparing memory of different speeds.
example:
1600 MHz at CAS 9 = 5.625
2400 MHz at CAS 12 = 5.000
so in this case CAS 12 is lower latency than CAS 9.