Testing (continued):
Performance Test v6.1:
This final suite of tests is new to me, and I decided to run them after seeing them used on a few other sites. Performance Test v6.1's memory benchmarks provide six different values, and higher is better for all of them. The tests in this package are obviously geared towards the use of more memory. Whether overclocked or not, the 4GB 667MHz kit performs about as well as, or better than, the 2GB 800MHz kit in all phases of testing. And in some tests, like "Large RAM" and "Mark" it isn't even close. This suggests what many may already understand, and that is that in some situations the capacity of the RAM is more important than the speed of the RAM.
Real World Evaluation:
After running through the benchmarks I wanted to take a look at a real world example of how 4GB of memory might come in handy, even if it was running at a lower speed. The benchmarks showed that this kit didn't perform poorly, and was able to hang with the faster memory for the most part, but benchmarks can't paint the complete picture.
The test was simple, although somewhat subjective. I have no hard figures to express the performance, but the results were rather binary; the 4GB kit completed the task and the 2GB kit did not. I had tried the following a few weeks ago, and based on my system (with 2GB of memory) hanging before completion, I realized the task had to be split into smaller groups that the system could handle.
I wanted to open roughly 1150 PHP files in order to do a "Find / Replace" for a link that needed to change on certain sections of Bigbruin.com. The text editing program I use has a tabbed interface, allowing multiple files to be opened at once. With 2GB of memory in place, I tried to open all 1150 at once, and after several minutes of what I assumed may have just been processing time, the mouse became unresponsive and I had to hit reset. I then tried the same thing with the 4GB kit installed, and although there was also a minute or two where I thought the same thing might happened, I was able to take control of the system again and complete my edits.
Like I said, not a subjective test, but it does confirm that the capacity of a system's memory may be more important than the speed and timings. For a desktop or workstation that can't take advantage of anything higher than PC2-5400, adding more memory may be just the ticket to keeping it working at its best.
|
|
