|OCZ 6GB 1600MHz CL7 DDR3 OCZ3P1600LV6GK|
|Reviews - Featured Reviews: Memory|
|Written by Olin Coles|
|Tuesday, 24 February 2009|
Page 6 of 8
Performance Test Results
Testing system memory (RAM) is a subject which requires a bit of technical knowledge. I have recently found a few other sites using nothing except video games to benchmark the RAM product they are reviewing, and this whole practice made no sense to me at all. Video games are GPU dependant when they have a decent processor behind them, and RAM has a very small impact on frame rate performance - as I will prove to you in my own tests. So keeping all of this in mind, I use the tools that belong in a system memory review; I use system memory benchmarking applications. Additionally, I won't spend three pages discussing how I overclocked or how I made it to tighter timings. Every memory module comes with its very own limit, so one size doesn't fit all and this kit may not be the same as the next.
As it turned out, the OCZ 6GB DDR3 Memory Kit OCZ3P1600LV6GK memory kit couldn't overclock much beyond the factory 1600 MHz from OCZ without adding any additional voltage and keeping the default clock latency values of CL7-7-7-24. Keep in mind that the default memory voltage is set to Intel's recommended maximum of 1.65V, which means that there isn't much room for stretching it. What this 1600 MHz kit could do though, was drop latency down to CL7-7-7-20. I know, it's not much, but lower latency means more than a higher clock speed with trip-chan kits anyway.
First up was the PassMark Performance Test benchmark which runs several different system memory tests in a row. Although some of the tests are specific to the performance of the RAM, others take the CPU clock speed and front side bus into account when developing a score. Most important are the memory read and write tests, and the score based tests are bias towards CPU speed and other hardware factors.
Passmark's Performance Test offers the most consistent memory test results of the entire group, with each of the test runs resulting in a score nearly identical to the previous test run. Ideally, all of these programs should be this consistent, but until they are I would consider Performance Test to be the best tool available for testing system memory bandwidth. Oddly enough though, there's very little performance difference between the kits we've benchmarked. As an example, the 1066MHz CL6 kit performs actually performs better than the 1764MHz CL8 kit but the 1333MHz CL7 kit outperforms them all. Memory, similar to processors, is going to benefit most by an efficient transport architecture. When it comes to triple-channel configuration, and specifically using DDR3, there's a penalty for higher latency cycles.
Our next test results were recorded from Lavalys EVEREST using the Cache and Memory Benchmark tool. The results shown below represent the average measurement obtained from the 6GB Triple-Channel 1600 MHz OCZ DDR3 kit. The results for the average read, write, and copy bandwidth from EVEREST are displayed below using several different kits for comparison.
Everest is among my most trusted benchmark programs, and the Cache and Benchmark tool is one of the more reliable in terms of consistent results. The chart above shows that the difference betwen 1333 and 1600 MHz amounted to a sizable increase in added bandwidth while keeping the same latency timings. Everst makes it appear as though there could be a significant difference in the performance between the various kits we used for testing, but the real question is whether this difference will translate into a notable real-world experience.
SiSoftware Sandra Lite offered results nearly the same as EVEREST in regards to increased performance; according to the chart above. Although the bandwidth tests are of a different nature, the OCZ 6GB DDR3 kit performed in the same fashion here as it did in the EVEREST tests.
Sandra, similar to EVEREST, prefers memory clock speed and then factors in latency. Because of this, the 1066MHz CL6 kit scores lowest, with the 1764MHz (overclocked) CL8 scoring highest in the Internal Buffered test. The Float Buffered tests actually put the 1764MHz CL8 nearly equal to the 1600MHz CL7 kit. Once again, the synthetic benchmarks indicate that the faster (clock speed) kits are considerably better than the slower kits. However, it has been my experience that lower latency is better suited for real-world performance, not overall clock speed.
Finally, I tested the OCZ 6GB Triple-Channel 1600 MHz DDR3 kit in the game Far Cry 2. Realizing that games can be either CPU or GPU bound, this made it difficult for me to compare all of the memory sets since the clock speed of the processor would change as I adjust the BIOS settings for the desired RAM speed. So by testing with a more-than-capable GeForce GTX 285 video card with moderate game settings, the system becomes more CPU/RAM dependant.
Even despite the CPU/RAM dependence, less than 2 FPS in average frame rate separate 1066MHz from 1764MHz. Making my argument more pointed is the microscopic difference of 0.4 FPS between 1333MHz and 1600MHz memory, both running at the same CL7 latency. This mediocre improvement is evidence of how insignificant the system memory speed is in relation to video game performance. It also proves that the marketing campaigns meant to focus on ‘gaming memory' are a complete joke... unless you're the punch line buying into it. But don't misunderstand me, because system memory could have a much larger impact on game performance if you use it to overclock the processor.
Obviously, if you're using an X58-based computer system, you probably aren't going to keep the CPU at the stock speed anyway; and you probably also use a video card that more than makes up for the very small gains of overclocked system memory. This means that faster RAM allows for a faster CPU, and in turn produces a faster frame rate. But in the world of system memory benchmarks, comparing the different sets of RAM in a game is pointless.