Performance Gain when using CL7 and CL9 RAMs

[bim27142]

Folks, theoretically how much or how significant will be the performance gain between having CL7 RAM over a CL9 one?

I'm particularly eyeing on these 2 G.SKILL RAMs

(G.SKILL) Ripjaws-X 8GB DDR3 1600Mhz (F3-12800CL7D-8GBXH) vs. (G.SKILL) Ripjaws-X 8GB DDR3 1600Mhz (F3-12800CL9D-8GBXL)

The price difference is quite significant so just checking if it is really worth it...

In addition, which is probably better, getting a faster CL9 RAM or just the lower speed CL7 RAM?

(G.SKILL) Ripjaws-X 8GB DDR3 1866Mhz (F3-14900CL9D-8GBXL) vs. (G.SKILL) Ripjaws-X 8GB DDR3 1600Mhz (F3-12800CL7D-8GBXH)


Appreciate your inputs!

 

[entropy13]

Although there are certainly cases where pairing Sandy Bridge processors with low-latency or high-frequency memory can yield impressive gains, it's hard to find a common desktop application or game whose performance improves enough to justify the additional expense. If you're looking to set benchmarking records or to compensate for personal shortcomings, K-series Sandy Bridge CPUs at least make it easy to run exotic DIMMs at blistering speeds. Everyone else can rest assured that using relatively inexpensive DDR3-1333 memory won't cost them much performance in the real world.

http://techreport.com/articles.x/20377/4

This uncertainty together with pretty wide range of DDR3 SDRAM prices on modules with different specifications do not allow us to give specific recommendations regarding the best memory choices for Sandy Bridge platform. However, in general terms, you should keep in mind two things. Firstly, the memory frequency is of greater importance for the overall system performance than the memory timings. Secondly, the additional financial investments into faster memory may not pay back in the long run. In particular, high-speed DDR3-2133 and DDR3-1866 modules may cost 1.5-2 times more than the ordinary DDR3-1333 SDRAM.

Therefore, we believe that inexpensive DDR3-1600 SDRAM with not very aggressive timings would be the most reasonable choice for contemporary LGA1155 systems: in our opinion, memory like that offers the best price-to-performance ratio today.

http://www.xbitlabs.com/articles/memory/display/sandy-bridge-ddr3_8.html#sect0



Ideally DDR3 1600MHz CL9 is the "best" price/perf wise, but even 1333MHz RAM are good already for any real-time use.

 

[bim27142]

thanks kabayan!!! very informative... i just realized that indeed, faster RAMs are not so significant anymore since SB K series are multiplier unlocked, so old-school overclocking is like a thing of the past for these modern procies'

 

[twilyth]

That was very informative, but I don't see how they can make such a blanket statement when the importance of latency varies with the block size.

 

[erocker]

thanks kabayan!!! very informative... i just realized that indeed, faster RAMs are not so significant anymore since SB K series are multiplier unlocked, so old-school overclocking is like a thing of the past for these modern procies'

If this is for your AMD rig, get the lower latency stuff without a doubt.

 

[entropy13]

That was very informative, but I don't see how they can make such a blanket statement when the importance of latency varies with the block size.

The importance of latency...is nonexistent when you aren't benchmarking 24/7, that's what they said, so what has block sizes got to do with anything?

Except for this.

If this is for your AMD rig, get the lower latency stuff without a doubt.

 

[twilyth]

The importance of latency...is nonexistent when you aren't benchmarking 24/7, that's what they said, so what has block sizes got to do with anything?

Did you read the post at my link? If you understand that chart, it's pretty obvious.

 

[wolf]

(G.SKILL) Ripjaws-X 8GB DDR3 1866Mhz (F3-14900CL9D-8GBXL) vs. (G.SKILL) Ripjaws-X 8GB DDR3 1600Mhz (F3-12800CL7D-8GBXH)

they are the better choice without a doubt in my mind, even for a system enthusiast/gamer you will be hard pressed to notice much between 1866mhz ram and 1600mhz. cas 7 to 9 should have a bigger impact.

I'd say it's worth the price difference, but I'm not too sure the cost of both kits. that choice is really one you have to make yourself.

 

[entropy13]

Did you read the post at my link? If you understand that chart, it's pretty obvious.

It's pretty obvious that there are differences but rarely translates to actual differences in actual performances. Would be being 479 ns faster really be significant if you're not benchmarking?

 

[twilyth]

It's pretty obvious that there are differences but rarely translates to actual differences in actual performances. Would be being 479 ns faster really be significant if you're not benchmarking?

That's not the point. If most memory access is say 100 words, then there is a huge difference between CAS4 and CAS9. But once you get to the 1000th word (sequential, not random - I think), there is almost NO difference.

Do you think any of these tech sites have the foggiest notion of what the average sequential block size is for the average user? No, they do not. But I would be happy to be proven wrong since my google-fu is not good enough to try to prove myself wrong.

 

[bim27142]

If this is for your AMD rig, get the lower latency stuff without a doubt.

yeah, i agree...

 

[slyfox2151]

subbed to hear more on this debate

 

[manofthem]

I have the 8G 1600 set cl9, and then oc'd it to 7-8-7-22@1.6v. Works fine with my amd build