Introduction

Last month Intel officially announced and released their new desktop CPU, the Core 2. With rave reviews from all over the web it's hard to imagine this CPU not being a success with the hardware enthusiasts.

With the spotlight on the CPU and sometimes the motherboards it seems the DDR2 memory modules are mostly being left in the dark. We aim to answer the following question in this article: Do you need expensive higher rated DDR2 modules to get the most from a Core 2 setup?.

With AMD's AM2 we've seen that you need DDR2 800 with rather strict timings to achieve optimal performance and match that of the older S939/DDR1 combo. Prices for these high speed DDR2 memory modules can easily be double of the value minded products, yet the performance increase is unfortunately not doubled. Will Core 2 + high speed DDR2 tell a different story?

In order to test a wide range of memory speeds and timings we needed high quality modules, OCZ Technology helped us out with 2x 1GB DDR2 PC2-7200 Platinum SLI-Ready.


Rated at DDR2 900 at very low timings (for DDR2) of 4-4-3-15 they should scale well from bottom to top of the different memory speeds/timings available.

What speeds did we test?

We tested 4 different memory speeds with 3 different timings and compared performance using six different benchmarks (including synthetic, games and applications). Using pricing info from Newegg.com and Alternate.de for 2x1Gb modules we created the following table as a brief overview:


Depending on where you shop you will run into similar prices, $/€150 seems to be a popular price point for the value line products.
Test Setup

Test Setup
CPU	Intel Core 2 Duo E6700
Mainboard	Intel Desktop mainboard D975XBX ?bad axe?
Memory	2 * 1024 Mb DDR2 PC7200 EPP OCZ
Other	Connect 3D X1900XT Silverstone Zeus 750 Watt PSU Nec 2050 DVD writer Seagate 120 gig 2x Seagate 200Gb SATA

Let the benchmarks begin ->

Benchmarks: Synthetic


We start off with Sisoft Sandra?s memory benchmark, it measures the maximum bandwidth of your system, as expected, the increase in memory speed and the decrease in memory timings has a considerable impact on the results.



Changing timings has a 5-10% impact, increasing memory speed results in a 10-15% bump. The biggest jump can be seen by going from PC3200 to PC4200.

If we throw the CPU into the equation we measure memory bandwidth impact in SuperPi 1Mb.



Not quite the same increase, maximum ~5% going from slowest to fastest. Remarkable is the drop in performance going from PC4200 (533) to PC5300 (667), even with tighter timings PC5300 can not catch up with PC4200, the difference is only ~2%, but it?s strange to see a decrease when you expect an increase in performance.

Now let?s add the graphics card to the whole in our last synthetic test, 3DMark05 from Futuremark.



Up to 3% increase, not quite note worthy is it? Going from PC4200 to PC5300 with CL3-3-3-8 makes no difference.

Game and Application benchmarks next ->

Benchmarks: Applications


The Photoshop benchmark from Driverheaven performs a series of actions on a large photo and displays the time it takes.



Tight timings are the key for this benchmark, which give you a boost of up to 4%... not impressive.

Benchmarks: Games


F.E.A.R. has an in-game performance benchmark which we used here, details were set to low game and resolution to 800x600, this offloads the video card and stresses the CPU/Memory/Motherboard more.



Even at this none GPU limited setting the maximum increase is only 6%, and you can see it becoming a trend with PC5300 performing lower than the expected.

The ID Software engines always been very good benchmark tools to measure system performance, at 800x600 we got these results:



Well, here the results almost reflect those got from the synthetic Sisoft Sandra benchmark, if you play Quake 4 at these settings with your high end video card, be sure to get high end memory! ;-)

However if you do want to turn up the eye candy and get the most from your system, you might find that the increase in memory bandwidth for your Core 2 is not paying off:




Let?s warp things up ->

Conclusive thoughts



So does Core 2 need high speed memory to shine? The answer is a resounding no! This is different from AMD AM2 where more expensive memory is needed to get the most out of the system.

On Intel Core 2 pure synthetic memory bandwidth benchmarks show a 30+% increase, but this does not translate in a noticeable performance bump in games and applications, where the increase, at best, is ~6% and this going from cheap high latency PC3200 to expensive low latency PC6400, and while these expensive modules do take the performance crown, their lead over the mostly cheaper PC4200 rated sticks is smaller than 3% in real world benchmarks.

So it doesn’t matter much what memory speed/timings you buy, the value line will suit the Intel Core 2 system just fine, but do keep one thing in mind, as in our testing we found performance actually decreases a bit going from PC4200 (533) to PC5300 (667)!

The Core 2 has a front side bus (FSB) speed of 266Mhz x 4 (Quadruple) “1066Mhz”, the ram is running at 266Mhz x2 (Dual Channel) x2 (DDR) = “1066Mhz”, so with PC4200 memory and FSB are running synchronized. When you use PC5300 you are no longer running synchronous with the FSB and a memory divider of x1.25 (5/4) has to be used. The older Athlon XP from AMD also displayed this decrease in performance when running memory asynchronous due to its short pipeline, where memory latency is more important then memory bandwidth. With the Pentium 4 the pipeline was longer and the effect of running asynchronous which increased latency was masked. Core 2 technology marked Intel's return to a shorter pipeline and thus is more similar to the Athlon XP than the Pentium 4.

By running the memory synchronous to the FSB you have the least amount of latency and thus performance is at its best. Why is PC6400, which is also running asynchronous, faster then? Because the memory speed is now that much higher that it compensates for the loss of running asynchronous and overall performance does increase. PC5300 worked great with Pentium 4 but it should be avoided when running Core 2, except when you plan to overclock. PC5300/6400 will give you the extra headroom to increase the FSB while keeping memory synchronous.

We hope this article was useful, if you have any suggestions or comments please don’t hesitate to drop a line in our forums.

Questions/Comments: forum thread