- Back to homepage
- Related articles
  Related articles

High End DDR2 Memory Roundup - Corsair, OCZ, Mushkin, Team

Memory by thorgal @ 2006-11-27

For some people, fast just isn´t fast enough. When you want to build a really high end system, you must search for the best memory out there. Today we take a look at 4 memory kits that are positioned at the upper end of the performance spectrum. Let´s find out which kit comes out on top.

Test Methodology and Latency/Bandwidth Results

Test methodology

Every module we mentioned on the previous page, except for the generic module was put to the test at five different speeds:

PC6400 - 800Mhz performance, at the best possible timings

PC8500 - 1066Mhz performance, at the best possible timings*

Maximum cas 3 performance

Maximum cas 4 performance

Maximum cas 5 performance

*1066 Mhz is tested only for the modules for which this is possible, in this case all 4 of our candidates

For the 800Mhz and 1066Mhz tests, settings are very simple: in bios only the memory speed was turned up to the appropriate setting, while front side bus (FSB) and cpu speed (and multiplier) remained at the stock settings of 266FSB and 2.66Ghz respectively.

As we mentioned a couple of pages ago, the maximum cas "x" performance is measured at other than stock cpu and FSB settings, to be able to give the memory sufficient bandwidth, and to play down the influence of the cpu in comparison to the memory. In our tables on page 8 you'll see that the cpu speeds and FSB setting is always mentioned at the bottom, so please bare that in mind when comparing the results. So to be clear: results cannot be directly compared, as the cpu/Fsb factor accounts for a certain % of the performance shown.

A word on Asus P5B-deluxe performance

The motherboard that we are using for our memory roundup is the Asus P5B deluxe/wifi-AP, a motherboard equipped with the Intel 965 chipset. We chose specifically this motherboard because it is capable of attaining very high front side bus speeds compared to motherboards equipped with the Intel 975 or (older) nVidia chipsets. Among the 965 motherboards, the Asus is also one of the better performing boards, often reaching FSB speeds of more than 500FSB. Of course, when we want to test memory overclocking, we need a motherboard that is capable of this high FSB, and one that is compatible with high frequency ram (the 975 chipset is known to give problems with ram over 1000Mhz, sometimes).

Over the last couple of months, many of the highly respected memory guru's like Tony Bigtoe, Freecableguy (FCG), bachus_anonym and EVA2000 (sorry for all those I forget) have been looking for answers where this high FSB performance, and consequently the high memory performance, was coming from on the P965 chipset. At a certain moment, maximum memory speeds were all of a sudden booming all over the internet, and the general feeling was that some "impossible" memory scores were floating around. That's why some answers had to be found.

You must realize that, when you start pumping up the FSB of your motherboard, you are actually overclocking your chipset. Just like memory, chipsets have standard timings and standard voltages, and when you're overclocking you can give a little more voltage to the chipset, in order to make it scale higher. Normally, once you hit 380-430FSB on the Intel 965 and the 975 chipset, you hit an overclocking wall (unless you do some more exotic, warranty voiding trickery of course...) . How does Asus reach 500FSB than ? One of the things Tony found out was that the P5B automatically relaxes (slows down) the timings of the chipset itself in order to be able to scale above 400FSB. There seem to be certain "sets" of timings which are implemented at different FSB settings, and these sets are called "straps". The Asus P5B now sets a different strap when you boot at 400+FSB, than when you boot at 400-FSB, the 400+ strap being the slower of the two. This obviously has a direct influence on your memory bandwidth and on the entire performance of your PC. This will show up in this review as well, when we're looking into the performance of the Team Group's ram, which scaled beyond 400FSB on our settings, and in the process gives worse results than you would actually expect... To conclude : some answers were found already, but others remain unanswered for the time being. Tony and the others are still working very hard to unravel more mysteries, and try to get the motherboard manufacturers on board in the process.

Take a look at this and this thread by Tony Bigtoe over at the OCZ support forums of BleedinEdge, and also at this thread at the XtremeSystems forums. The threads by Tony are in my personal opinion obligated reading for anyone considering to buy a top performing memory kit, just to understand the performance of your PC setup.

Benchmark suite

As soon as we ensure full stability on all of the above settings using Orthos, we are ready to run our benchmark suite. This suite consists out of the following applications:

Sisoft S.A.N.D.R.A. 2007

Lavalys Everest 2.20

Maxon Cinebench 9.5

MadOnion/Futuremark 3D Mark 2001SE

Futuremark 3D Mark 2006

Futuremark PC Mark 2005

Superpi 1.5, Xtremesystems edition

Game 1 : 3D Realms Prey demo ; benchmark by Hardware OC

Game 2 : Monolith F.E.A.R. Multiplayer demo, built-in benchmark

Bandwidth at 800/1066Mhz settings

To measure the bandwidth and latency performance of the memory kit we use two standard applications.

First is the bandwidth for which we use SiSoftware's Sandra application. The System ANalyser, Diagnostic and Reporting Assistant is an information and diagnostic utility which has a very handy benchmark suite, in which you can compare your system to a range of other reference systems. The Bandwidth is one of the available benchmarks :

Madshrimps (c)

As all of the memory kits we review today have similar timings on the 800Mhz memory strap as well as the 1066Mhz memory strap it comes as no surprise that the results are all the same : there are only some statistic differences. 800Mhz memory strap gives us a 14% benefit in bandwidth, whereas 1066Mhz memory strap gives us almost a 20% gain. The fact that Corsair is twice on top might not be a coincidence, as the modules had slightly better timings than the competition (3-3-3-4 vs 3-3-3-8 at 800Mhz, and 4-4-3-4 vs 4-4-4-X at 1066Mhz). The difference is very, very small though...

Latency at 800/1066Mhz settings

Second up is the latency test for which we use the Everest application from Lavalys.

Madshrimps (c)

Latencies are almost the same across the board as well, again with a slight advantage for the Dominator. This is surely the result of the slightly tighter TRP and TRAS timings at 1066Mhz. The gains are 18% and 38% on the 800Mhz and 1066Mhz strap respectively.

Let's have a look if these differences are confirmed in our application benchmarks ->

Comment from Sidney @ 2006/11/27

Very nice article; can hardly wait for the mid-range to come out for comparison.

Comment from geoffrey @ 2006/12/22

Nice round up man, usefull info in there

My 2*1GB TEAMPGROUP don't seem to overclock as well as yours Thorgal. At 3-3-3-8-3 I could not post @ 825MHz @ 2,45V.

Comment from jmke @ 2006/12/22

825 is still very nice

Comment from thorgal @ 2006/12/28

Quote:

Originally Posted by geoffrey

Nice round up man, usefull info in there

My 2*1GB TEAMPGROUP don't seem to overclock as well as yours Thorgal. At 3-3-3-8-3 I could not post @ 825MHz @ 2,45V.

Only saw your post today, Geof. What Team Group kit did you get ?

I've got another team group kit here, the one that I bought myself, the 667Mhz "333" Micron kit, and it also does about 800Mhz cas 3. Haven't thoroughly tested it though, as the kit does much better on cas 4 (over a 1000Mhz, and still pushing).

Comment from geoffrey @ 2006/12/28

Teamgroup Xtreem Kit 2x1024MB DDR2-800 4-4-4-10 (D9)

Comment from thorgal @ 2006/12/28

Quote:

Originally Posted by geoffrey

Teamgroup Xtreem Kit 2x1024MB DDR2-800 4-4-4-10 (D9)

My guess is these are only a different binning of the PC5300 "333" kits (the one I have). Did you test the max on cas 4 already ?

Comment from geoffrey @ 2006/12/28

No, not really, I have P5B Deluxe incomming, I shall test it as soon as I can. Got to instal a few Win Xp setup and still have to do some shopping so it might be for the end of next week.

Comment from blind_ripper @ 2007/03/19

hhhmmm is it just me or dos my kit suck !
max c4 32m is atm 2.6vdimm @ 550mhz .
not really tight timmings but not loos to , not im on chipset 975X or is this normal for this chipset ?
now im testing C3

Comment from geoffrey @ 2007/03/20

Mine seem to pass 525MHz at CAS4 on 965 chipset, keeping in mind my previous results on C3 I should suspect there is more left in yours sticks Blind-ripper.

All information and graphics contained in Madshrimps are sole property of the Madshrimps crew and may not be reproduced or copied in any manner without written permission from us.