 |
 |
 |
||
|
|
||||
 |
|
|
| Introduction and Specifications | ||||||
In PC enthusiast circles, Corsair is a name that needs no introduction. For years, the company has consistently produced some of the most sought after, high speed memory modules available at the time. Their XMS line of DDR memory was incredibly popular when it was initially released, as were the company's Pro and Xpert modules. Today it's Corsair Dominator memory kits that are enticing PC enthusiasts, thanks to their innovative cooling solution and high-performance. In this article, we'll be focusing on Corsair's latest Dominator memory kit, the TWIN2X2048-10000C5DF (2GB, 2 x 1GB DIMMs). The PC2-10000 DIMMs used in this kit are high-capacity and are capable of running with a CAS Latency of 5, a RAS to CAS Delay (tRCD) of 5, a Row Precharge (tRP) of 5, and an Active to Precharge Delay (tRAS) of 18 (5-5-5-18-2T) at an impressive 625MHz (DDR2-1250MHz).
"The Dominator Series Twin2X2048-10000C5DF is a 2048 MByte matched pair of DDR2 SDRAM DIMMs built using Corsair's latest high performance heat sink with Dual-Path Heat Xchange (DHX) technology coupled with a Corsair Dominator Airflow Fan. This part is designed to unleash performance of the new NVIDIA nForce 680i SLI-based platform. This memory has been verified to operate at 1250MHz at latencies of 5-5-5-18. The Twin2X2048-10000C5DF comes with Enhanced Performance Profiles (EPP), the open standard for performance module SPD's jointly developed by Corsair and NVIDIA. EPP SPD's on Corsair modules allow users to automatically configure EPP enabled motherboards with aggressive memory performance settings, for maximum memory and system performance."
The "DF" designation in the TWIN2X2048-10000C5DF kit's name mean that this kit includes one of Corsair's active memory coolers, also known as the Dominator fan. As we've mentioned in the past, the Dominator fan assembly is constructed of aluminum, with three 4cm fans. It mounts to the clips that typically hold system RAM in place and connects to a single 3-pin fan plug. We should note that the fans can be throttled if your motherboard has the capability to control fans based on temperature. In practice, we found the Dominator fan to be easy to install and relatively quiet, but in our open-air test bench its effect on temperature was difficult to quantify. Corsair claims the Dominator fan will reduce operating temperatures by a few degrees (from 15.4oC to 11.3oC above ambient in Corsair's testing), which will increase stability, overclockability, and longevity. The only issue we had with the Dominator fan is that it cannot be installed properly on motherboards where the DIMM slots are in-line with the PEG slot and an extra long graphics card is installed. This is a mistake fewer and fewer motherboard manufacturers are making lately, however, so it shouldn't be an issue for many of you. |
| Inspecting the Modules |
At first glance, Corsair's flagship Dominator memory looks much the company's "Pro" branded products minus the activity LEDs, but the Dominator heatsinks are actually quite different than anything previously offered by Corsair.
The memory chips on both sides of the PCB are also outfitted with their own dedicated, oversized heatsinks to help cool the chips as well. Having two paths from which heat can be drawn out of the modules is what Corsair is calling DHX technology, or Dual-Path Heat Xchange. And the Dominator fan featured on the previous page is the third component that aids in cooling, which in turn enhances stability, longevity, and overclockability. In addition to DHX, Corsair is meticulously binning the memory chips used on the Dominator kits and is using only select samples that can operate reliably at their rated latencies and frequencies. The culmination of Corsair's efforts is a 2GB memory kit that's rated for operation at over 1.2GHz. We should note, however, that the TWIN2X2048-10000C5DF memory requires 2.4v to operate at its rated speed. That's a lot of voltage, and many motherboards don't offer DDR2 voltages quite that high. Lastly, we should mention that the Dominator TWIN2X2048-10000C5DF memory is equipped with EPP, or Enhanced Performance Profiles. Enhanced Performance Profiles can increase performance by taking advantage of additional memory parameters added to the unused portion of a standard JEDEC Serial Presence Detect, or SPD ROM. The JEDEC specification only calls for small amount of data to be stored in a standard SPD, which leaves a significant amount of unused space. EPP takes advantage of this space to store specific information about the modules, like their maximum supported frequencies and timings.
Although the Dominator TWIN2X2048-10000C5DF memory has four distinct heatsinks and a taller-than-usual PCB, the DIMMs aren't all that much bigger than more traditional DIMMs, as you can see in the pictures above. Of course, the Dominator DIMMs are taller then most others, but they're actually a little thinner than memory with typical heat spreaders. We don't think clearance will be an issue in any standard system, SFF user's however, may want to ensure you've got the room in your chassis. |
| Compatibility and Our Test System | |||||||||||||||||||||||
Compatibility Testing: Before we sat down in front of our test machine to evaluate Corsair's TWIN2X2048-10000C5DF memory modules, we installed them into some of the systems we had available at the time to see if there were any compatibility problems to speak of. We tried the sticks in a total of 4 different systems, powered by various popular chipsets. We tested these modules on the following platforms...
With the exception of the EVGA nForce 680i SLI, which we used for the benchmarks in this article, to quickly test these modules all we did was install them, power up the systems, made sure Windows booted and then we ran SiSoft SANDRA's burn-in wizard for a few minutes. Thankfully, we didn't experience any issues whatsoever. The memory worked perfectly in all of the platforms we tested. This limited compatibility testing wasn't extensive by any means, but it does bode well for the TWIN2X2048-10000C5DF memory modules. The fact that they worked properly on the popular platforms listed here, means most consumers shouldn't have any major issues with them either.
How we configured our test systems: When configuring our test system for this review, the first thing we did was enter the system BIOS and set the motherboard to its default configuration. The hard drive was then formatted, and Windows XP Professional (SP2) was installed. When the installation was complete, we installed all of the necessary drivers for our components, and removed Windows Messenger from the system. Auto-Updating and System Restore were also disabled, and we setup a 1024MB permanent page file on the same partition as the Windows installation. Lastly, we set Windows XP's Visual Effects to "best performance", installed all of our benchmarking software, defragged the hard drive and ran all of the tests. Throughout all our benchmarking, we had the memory voltage set to 2.4v, unless otherwise noted. |
|||||||||||||||||||||||
| Stock Performance (By SPD) | ||||||||||||
We began our testing with SiSoftware's SANDRA, the System ANalyzer, Diagnostic and Reporting Assistant. SANDRA consists of a set of information and diagnostic utilities that can provide a host of useful information about your hardware and operating system. We ran SANDRA's Memory Bandwidth and Latency tests on a Core 2 Extreme X6800 powered test bed with three different brands / types of memory installed. The BIOS was set to AUTO for these tests, so the different RAM modules were being run at standard JEDEC DDR2-800 specifications here.
With all of the memory running in similar configurations, they all put up virtually identical scores in the two SANDRA tests we ran. There was no significant different between the kits when running at standard DDR2-800 speeds.
For our next round of benchmarks, we ran the Memory performance module built-into Futuremark's PCMark05. For those interested in more than just the graphs, we've got a quote from Futuremark that explains exactly what this test does and how it works... "The Memory test suite is a collection of tests that isolate the performance of the memory subsystem. The memory subsystem consists of various devices on the PC. This includes the main memory, the CPU internal cache (known as the L1 cache) and the external cache (known as the L2 cache). As it is difficult to find applications that only stress the memory, we explicitly developed a set of tests geared for this purpose. The tests are written in C++ and assembly. They include: Reading data blocks from memory, Writing data blocks to memory performing copy operations on data blocks, random access to data items and latency testing."
PCMark05's memory performance module didn't report any major differences either. This is expected behavior, however, considering all of the kits were configured by at standard DDR2-800 specifications.
We continued our testing with some low-resolution F.E.A.R. tests. Despite the fact that this is a game benchmark that can be used to test the relative performance of video cards, frame rates are strongly influenced by processor speed and available memory bandwidth, especially at low resolutions, which is how we ran the tests to get the frame rates listed below.
We had another dead heat with the low-res F.E.A.R. benchmark. On the pages, ahead, however, we'll test the three memory kits at their rated speeds and timings, which should yield much different results. |
| Performance @ Rated Speeds | ||||
Next, we continued testing with SiSoftware's SANDRA, but we manually set the memory timings on the Corsair TWIN2X2048-10000C5DF modules to their rated settings of 5-5-5-18 (CAS Latency = 5T, RAS to CAS Delay (tRCD) = 5T, Row Precharge (tRP) = 5T, Active to Precharge Delay (tRAS) = 18T) at 1250MHz. We achieved these speeds by dropping our CPU's multiplier to 7x and raising the FSB to 1666MHz (416MHz quad-pumped) with the memory and FSB linked via the EVGA nForce 680i SLI motherboard's BIOS. The end result was a CPU clock speed of 2916MHz. We performed a similar procedure for the Corsair TWIN2X2048-8500C5 and OCZ PC2-8000 Platinum memory kits as well, but there were slight differences in the CPU speed in the end. To hit 1066MHz with the Corsair 8500C5 kit, we dropped the multiplier to 8x and raised the FSB to 1422MHz, which resulted in a 2844MHz CPU clock. And for the OCZ kit, which is rated for 1000MHz at 5-5-5-15, we dropped the multiplier to 9 and raised the FSB to 1333MHz, for a final CPU clock of 2997MHz.
Running the Corsair TWIN2X2048-10000C5DF memory kit at its rated speeds and timings resulted in a huge performance increase. While running at its rated speeds, the kit put up bandwidth scores almost 3GB/s higher than stock DDR2-800. Latency was also decreased from 85ns down to 69ns. The results also show the kind of bandwidth increases and latency reductions offered to users who run their memory at 1250MHz, 1066MHz, and 1000MHz. |
| Performance @ Rated Speeds (Cont.) | ||||||||
In this next round of benchmarks, we re-ran the Memory performance module built-into Futuremark's PCMark05 with each set of modules set to their rated frequencies and timings
These PCMark results give some insight into how the benchmark tabulates its final score. The relatively low latency and high-bandwidth offered by the TWIN2X2048-10000C5DF kit put it firmly in the lead by a few hundred points here. But notice that the OCZ kit, which offered less bandwidth than the 8500C5 kit due to its slightly lower clock speed, actually scores higher, because of the lower latency associated with the configuration necessary to get the kit running at its rated speed.
We continued testing these modules with some low-quality F.E.A.R. benchmark runs while the systems were running at their rated speeds as well.
F.E.A.R. didn't show much of an improvement despite the much higher clocks speeds and increased bandwidth offered by all three of the kits we tested. Other games, however, that may be more dependant on memory bandwidth, would show larger improvements. |
| Performance @ Highest Overclock | ||||
For our next set of numbers, we focused on the maximum overclockability of Corsair's Dominator TWIN2X2048-10000C5DF memory kit while set to its rated timings. Using a Core 2 Duo Extreme X6800 CPU, we raised the Front Side Bus speed while concurrently lowering our processor's multiplier. We tried to keep the CPU as close to it's stock 2.93GHz clock speed as possible. For these tests, we locked the PCI Express clock to 100MHz, raised the CPU voltage to 1.4v, lowered the CPU multiplier to 7x, and kept the memory voltage to 2.4v. Ultimately, we left the FSB at 1666MHz - similar to the tests on the previous two pages - but ran the memory in Unlinked mode, where we could raise the memory speed independent of the FSB. The max stable memory speed we were able to achieve was 1295.7MHz. The next highest setting offered by the board, 1309MHz, booted into Windows, but we could not complete any benchmarks.
Our maximum overclock didn't yield any appreciable performance gains, as you can clearly see. Bandwidth did go up by a few megabytes per second, but latency actually got worse, likely due to running the nForce 680i SLI motherboard in Unlinked memory mode. |
| Performance @ Highest Overclock (Cont.) | ||||||||
We also re-ran the PCMark05 memory performance and F.E.A.R. benchmarks again with the Corsair Dominator TWIN2X2048-8888C4DF memory modules overclocked to DDR2-1295.7MHz on our Intel Core 2 Extreme X6800 powered test bed...
Despite the slightly higher latency reported on the previous page, the Corsair TWIN2X2048-10000C5DF memory kit's score went up a bit in the PCMark05 benchmark. The 14 point delta is hardly worth getting excited over, however.
F.E.A.R. didn't show any performance improvement while overclocked. The slightly higher bandwidth offered by the memory in its overclocked configuration just wasn't enough to bump up the framerate here. |
| Our Summary and Conclusion | ||||
Performance Summary: Corsair's new Dominator TWIN2X2048-10000C5DF memory was the fastest of the three memory kits we tested here, thanks to its combination of high-capacity, relatively low latency, and ultra high frequency. In SANDRA's memory bandwidth and latency benchmarks it put up the highest bandwidth numbers and some of the lowest latencies, and it similarly posted the highest scores in the PCMark05 and F.E.A.R. benchmarks as well.
As PC enthusiasts, we can't help but be impressed by Corsair's Dominator TWIN2X2048-10000C5DF memory kit. The TWIN2X2048-10000C5DF kit is arguably the fastest matched pair of DDR2 memory modules on the market today. This memory also features an innovative cooling solutions (DHX), support for EPP, and throughout our testing it was rock-solid stable in every configuration. It wasn't until overclocked and pushed to almost 1.3GHz that this memory faltered, and even then it could very well have been the nForce 680i SLI motherboard holding us back. It's not all good news, however. A quick search using Pricegrabber yielded only a single vendor selling this memory, and it was priced at a wallet-busting $748. We're sure this price is a reflection of this memory's limited available at the moment, but that price is almost double what most 4GB memory kits are currently selling for and it's obviously much higher than the price of other high-speed 2GB memory kits as well. If you want the highest clocked DDR2 memory currently available in your PC, cost be damned, Corsair's Dominator TWIN2X2048-10000C5DF memory kit is it. But you could save a fortune by sacrificing a few MHz and going with one of the many PC2-8500 or faster memory kits on the market today, including Corsair's own TWIN2X2048-8500C5DF.
|
;)
;)
;)
;)
;)
;)
;)
There are four key things at work in Corsair's Dominator memory kits that result in the product you see pictured above. With the Dominator series, the memory's PCB has been heightened and there are thermally conductive vias and traces running through the PCB and connected to the ground plane, to help dissipate heat from within the board itself. Heatsinks are then bonded to both sides of the upper portion of the PCB to further aid in heat dissipation.;)
;)
