Crucial DDR2 PC2-5300 (DDR2-667): Performance

Examining the specifications for Crucial Technology's Ballistix DDR2 PC2-5300 2GB memory pair, we see that the memory timings specify lower latency than the higher speed DDR2-800 parts. They are specified to operate at CL-TRCD-TRP-TRAS timings of 3-3-3-12 at 2.2 volts. Here is a screen capture of the timings from CPU-Z:


Click to enlarge.

Default SiSoft buffered bandwidth is depicted in this image below.


Click to enlarge.

Another look under Everest Ultimate Edition 2006 shows the SPD timings with the Crucial PC2-5300 modules.


Click to enlarge.

Stability testing was again accessed through the use of SuperPi as one tool. You'll note that at a front side bus speed of 320 MHz, these modules were able to maintain timings of 3-3-3-8 and complete a 32M run of SuperPi with no issues whatsoever.

By slightly relaxing the timings to 4-4-4-8, performance is shown below of buffered bandwidth from SiSoft Sandra 2005 Professional, of nearly 8Mb/s.


Click to enlarge.

Below, you'll find a screenshot of CPU-Z and buffered bandwidth from SiSoft Sandra 2005 Professional.


Click to enlarge.

Let's take a look at the SiSoft Sandra 2005 Professional unbuffered bandwidth snapped at 340 x 13, which depicts memory bandwidth of approximately 4.5 Mb/s.


Click to enlarge.

This memory and the PC2-6400 both exhibited very stable conditions under stress and at higher clock speeds.


Click to enlarge.

Crucial PC2-6400: Gaming Performance Windows XP Professional x64 Performance
Comments Locked

27 Comments

View All Comments

  • leexgx - Saturday, February 25, 2006 - link

    you Must understand the CPU that tomshardware got was an Test CPU, thay realy should of not done an benchmark as the CPU it self was not fully working and the DDR2 ram that was used was could only work at DDR2 666 and the timeings was stuck at 4-4-4-12 and may have been limted by the broken DDR2 contorler in it as well (you realy should look at it and not trust the resluts at all and thay have told you that its not to be trusted as well as its not an retail product as toms' stats)

    basicly at the time when P4 came to use DDR2 the timeing speed of DDR2 was silly DDR1 would out perform DDR2 this is why amd have not gone to DDR2 as ram speeds are now to the point where it will not be an bottle neck (and Ram price between DDR1 and 2 are not that must £10-£40 $40-$80 depending on size)

    testing the Ram performance in am P4/P5 system is not usefull any way as it can not use it (get some intresting resluts when AM2 fully comes ot)

    (the date is 06/06/06 when it comes out)
    my self i am going to stick my AMD X2 3800+ 939 and mobo and 2Gb of ram in me server (board is dieing caps are leaking on this P4 mobo) and get AM2 when 65mm fabs come out for AM2 and fast ram by then (and some good mobos)
    all that will probly be end of year i can wait my 3800+ X2 is running at 2.4ghz (basicly 4600+X2 as the ram is running at DDr480 as well)and its happy and cool keeps me happy
  • Jedi2155 - Sunday, February 26, 2006 - link

    DDR1 outperforming DDR2 wasn't the main reason AMD didn't go to DDR2 right away. The main reason (if you read AMD's lawsuit against Intel's marketing practices) was because Intel created separate consortium to design DDR2 with the memory manufacturer's with the full intent of keeping AMD out of the loop. Thus AMD was a year or 2 behind Intel in developing the memory controller for DDR2 (as they couldn't see the specs till it was finished) nor did they have any input in the design of the specification.

    And thus DDR2 sucks more than it should had AMD been involved :)
  • DigitalFreak - Wednesday, February 22, 2006 - link

    The AM2 platform launches on June 6, 2006 (6/6/6). Those aren't memory timings.
  • Spacecomber - Tuesday, February 21, 2006 - link

    I'll admit that I mostly skimmed through this article, but since much of the data seemed to be in the form of screenshots that are too small to be easily legible without cliking on them to get the enlarged version, I found it hard to get a quick glimpse of what this article might have to offer.

    I would much prefer results to be presented in traditional graphs and tables, instead.
  • Olaf van der Spek - Tuesday, February 21, 2006 - link

    What is the information about the AMD CPU doing in this test?
    Too much copy/paste?
  • JustAnAverageGuy - Wednesday, February 22, 2006 - link

    Yeah, that does seem out of place.

    Authors also need to go back and recheck their MB\GB , MHz\GHz labels. I sure don't want any 4.5MB\sec RAM.
  • 96redformula - Tuesday, February 21, 2006 - link

    My faith in crucial is gone.

Log in

Don't have an account? Sign up now