Logo   Banner   TopRight
TopUnder
Transparent
OCZ Reaper Heat Pipe Conduit DDR2
Transparent
Date: Jun 18, 2007
Section:Misc
Author: Michael Lin
Transparent
Introduction
For the last several months, DDR2 memory development has been largly stagnent. The fastest DDR2 memory kits from last summer were, for the most part, still the fastest DDR2 memory kits this spring. We haven't seen any truly significant developments in DDR2 since Micron stepped up to the plate with their D9 chips. Although manufacturers continue to release new DDR2 memory, speed increases are fewer and harder to come by. It seems that even the mighty D9 has run out of headroom and it's becoming very difficult to produce faster DDR2 memory while still maintaining acceptible production yields.

With the relative standstill in DDR2 development, manufacturers determined to keep pushing DDR2 speed boundries have resorted to increasing voltage and improving cooling. Voltages on performance DDR2 modules now far exceed JEDEC's specification of 1.8v and values of 2.2v and higher aren't uncommon. As a result, these modules need better cooling and the standard heatspreaders that we've seen on performance RAM since the hayday of DDR are being replaced with more effective cooling methods.

We looked at the current DDR2 speed leaders back in March; Corsair's Dominator series and OCZ's Flex XLC series. Both of these product lines are good examples of recent attempts at pushing DDR2 speed boundries and they both utilize high voltages and relatively drastic cooling techniques. While Corsair went with the traditional big heatsink + fan method with their Dominator series, OCZ decided to go the water cooling route with their Flex XLC line. In recent years, heatpipe cooling has gained immense popularity in PC cooling and it was only a matter of time before they would also be applied to our memory. The wait seems to be over because the distinguishing feature of OCZ's latest line of DDR2 memory is the use of a large heatpipe assembly.

small_Packaging_front.jpg       small_Packaging_rear.jpg

OCZ Reaper HPC Performance DDR2 Memory
Features and Specifications
OCZ_Reaper_HPC_Banner.jpg

OCZ's latest solution for cooling their high performance memory modules is the Reaper Heat Pipe Conduit (HPC) system. The Reaper HPC memory uses a combination of an aluminum heatspreader, copper heatpipe, and aluminum fin array to quickly dissipate heat. Heat generated by the memory is absorbed by the heatspreader, where it is then transported to an aluminum fin array which then dissipates the heat into the surrounding air. In theory this method should cool the memory modules much more efficiently than the flat heatspreaders most other memory modules sport. The Reaper HPC currently comes in four flavors and we've summerized the juicy details in a table for your convenience.

OCZ Reaper HPC Edition DDR2 Memory
Reaper_Chart.png

While not as well specified as OCZ's Flex XLC series, the Reaper HPC series of memory modules certainly won't be mistaken for generic value RAM and they occupy the next highest spot on OCZ's product line. Like the Flex XLCs, all Reaper HPC memory modules get the benefit of an EPP (Enhanced Performance Profiles) programmed SPD which allows them to automatically boot at the rated specs when used on NVIDIA SLI chipsets. All Reaper HPC units are also covered by OCZ's lifetime warranty.

OCZ's Reaper HPC series originally consisted of two high performance memory kits rated at PC2-9200 and PC2-8500 speeds. Two new kits were recently added to the Reaper HPC line-up, an affordable PC2-6400 kit and a low-latency Enhanced Bandwidth Edition PC2-6400 kit.
For the purposes of this article, we have obtained the original PC2-8500 Reaper HPC kit and the new Enhanced Bandwidth Edition PC2-6400.
Transparent
Design and Features
Thanks to it's unique cooling solution, the OCZ Reaper HPC memory modules are very distinctive in appearance which sets them apart from the masses of memory modules making use of flat heatspreaders. The black heatspreaders and copper heatsinks give the Reaper a very slick look, although it also makes it nearly twice the height of most memory modules. While there are a couple of manufacturers offering after-market memory cooling solutions that make use of heatpipes, OCZ is the only memory manufacturer to use such a design. OCZ calls their cooling solution the Heat Pipe Conduit, abbreviated to HPC.

Reaper_front.jpg

The Reaper Heat Pipe Conduit system consists of a large aluminum heatspreader attached to an aluminum fin array by a copper heatpipe. The aluminum fin array is basically a thin and long heatsink. The entire assembly resembles an elaborate handle, not unlike what you might see on a closet or dresser, although the space between the heatspreader and the fin array is too small to fit your fingers through. In order to increase the surface area, which improves heat dissipation, the aluminum fin array that constitutes the "grip" of the handle has a large number of notches cut into it. When viewed as a cross-section, the "grip" is actually in the shape of a X. This shape was chosen to further increase the fin array's surface area. 

small_Reaper_PC2-8500.jpg small_Reaper_PC2-6400.jpg
Reaper HPC PC2-8500
Reaper HPC PC2-6400 EB

The heatspreader, which is clamped to both sides of the memory module, has a number of channels cut into its otherwise flat surface, increasing its surface area. On one side is a shiny badge proudly displaying the OCZ logo. The other side of the module is completely plain except for the afromentioned channels cut into the heatspreader and two screws, which secure the heatspreader assembly in place. This side of the module also has a sticker in the upper left corner which specifies the model, speed and timings for the module.

OCZ_Reaper_HPC_Diagram.png

The HPC system works in two stages. First, the heat generated by the memory modules is absorbed by the heatspreader. The heatspreader does exactly what its name implies and spreads the heat accross its surface. The copper heatpipe then moves some of this heat away from the heatspreader to an aluminum fin array where the heat is dissipated into the surrounding air. This design uses the same basic principles used by other heatpipe systems and in theory it should greatly increase the cooling system's thermal capacity.

Remember that heatpipes are used in cooling applications to transport heat from one point to another, quickly and relatively efficiently . Heatpipes, on their own, don't have any noteworthy cooling capabilities. In this, as well as many other heatpipe systems, the heatpipe is used to connect a primary heatsink to a secondary heatsink. Due to its high efficiency, the heatpipe effectively joins the two heatsinks together, increasing the total surface area that can be used to dissipate heat. A larger surface area means the heatsink can dissipate heat quicker and handle larger thermal loads.

It's for this reason that you can't simply slap some heatpipes onto a heatsink and expect it to work effectively. The heatpipe needs to be well integrated so it can transport as much of the heat from the heatspreader to the fin array as possible. This, unfortunetly, is where the HPC's design falters. The heatpipe in the HPC assembly is simply tacked on to the top of the heatspreader. From this position, it will only have access to a small fraction of the heat collected by the heatspreader. It would have been much more effective if the heatpipe was attached to either side of the heatspreader near the center, right above the memory chips, where the heat is most concentrated.

The current position of the heatpipe reduces its effectiveness since it can only reach the relatively small amount of heat that rises to the top of the heatspreader. While this doesn't make the heatpipe assembly useless, it does mean that it won't reach its full potential.

Overall, the Reaper HPC is very well constructed. When we first saw images of the Reaper HPC, we hypothesized that i n order to seat the memory modules, you would need to apply a significant amount of pressure on the heatpipe assembly which may weaken it, causing it to break after several installs. After several dozen cycles of installation and removal during benchmarking, the assembly remains securly affixed. We never experienced any give or flexing while installing or removeing the modules. We're happy to report that our initial fears were unfounded and the heatpipe assembly is well constructed and very sturdy.
Transparent
Overclocking
Except for the cooling solutions they use, the Reaper HPC line-up is very similar to the Flex XLC series and memory modules belonging to either series utilize the extremely popular Micron D9GMH memory chips. Considering their similarity, we expect the Reapers to overclock well, just like the Flex XLCs did.

All of our overclocking was done on the NVIDIA 680i SLI platform since it allows pseudo-synchronous memory clocking. This means we can overclock the memory without changing the FSB frequency, which eliminates a potential bottleneck. It also means that the CPU clock is not affected by the memory overclock, allowing direct performance comparisons to be made in benchmarks since memory benchmarks are somewhat effected by the CPU. All of our overclocking was done with the FSB locked to stock frequencies.

CPU-Z_Reaper_8500_OC.png
Reaper HPC PC2-8500 @ 1143MHz & CAS 5-5-5-15

First up is the Reaper HPC PC2-8500. These modules are rated for a stock frequency of 1066MHz at CAS 5-5-5-15 with a stock voltage of 2.3V. OCZ provides Extended Voltage Protection (EVP) on all Reaper HPC memory, which allows their voltages to be increased to a specified limit without invalidating their lifetime warranty. The EVP limit varies from model to model and the PC2-8500 have an EVP of 2.35V. For all of our overclocking, we did not exceed the specified EVP voltage for the memory modules since most people probably aren't too keen on voiding their lifetime warranty. Unfortunetly, in this case, this restriction only gives us an extra 0.05V over the stock voltage to work with due to the module's already high stock voltage.

The highest overclock we were able to coax out of our particular Reaper HPC PC2-8500 modules was 1143MHz at the EVP voltage of 2.35V. At this speed, the modules were able to maintain stock timings. Any attempt to lower the timings or push the frequency higher resulted in significant instability. While this isn't a poor result, it certainly isn't special. It is relatively common for high-end PC2-8500 memory kits to reach overclocks of 1150MHz with nothing more than a minor voltage adjustment. However, we understand that overclocking results are very much a luck-of-the-draw type of deal so we took a quick look around the 'net to get an idea of what kind of results other people were getting with their Reaper HPC PC2-8500 modules. It seems that we were somewhat unlucky with our particular Reaper HPC PC2-8500 modules since our overclock was lower than many of those being reported around the net.

CPU-Z_Reaper_6400_OC.png
Reaper HPC PC2-6400 EB @ 1067MHz & CAS 4-5-5-15

We had a bit more luck with our Reaper HPC PC2-6400 Enhanced Bandwidth edition modules. These modules are rated for 800MHz with timings of 4-3-3-15 at a stock voltage of 2.0V. We were able to push our particular modules to a stable 1067MHz with a timing of 4-5-5-15 at the EVP voltage of 2.2V.
With this overclock, the modules were able to run through our gauntlet of benchmarks several times without incident. We were able to get our test system to boot into windows at higher frequencies, but the system couldn't make it all the way through a run of PCMark05 without crashing; not exactly what we would call stable. Additionally, increasing the memory timings to 5-5-5-15 did not give us any more leway with frequency.

Our result in this case is quite impressive. Our overclocked Reaper HPC PC2-6400 EB modules are superior to stock Reaper HPC PC2-8500 modules in many ways, with lower voltage, tighter timings and a lower price. A quick look around the 'net shows that our results are pretty typical and many other people are achieving similar overclocks although, as always, your milage may vary.
Transparent
Sandra Performance
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.

Each memory kit was tested at three speeds; stock speed, rated speed, and highest overclock. For "stock speed", we used set the motherboard to AUTO, which causes the memory modules to operate at standard JEDEC DDR2-800 specifications. This was done to provide a good "ground-floor" reference point for the modules. For "rated speed", the modules were manually configured to operate at the settings specified by OCZ. Finally, for "highest overclock" we set the modules to the highest stable overclock settings we were able to achieve.

Test System Details
Specifications and Revisions
  • Intel Core 2 Duo E6400 (2.13 GHz) Dual-Core
  • 1 x Nvidia nForce 680i Series Motherboard
  • 1 x Western Digital Raptor 10,000 RPM Serial ATA Hard Disk
  • Microsoft Windows XP Professional (32-bit)

    Test Candidates

  • Reaper HPC PC2-8500 @ Stock (JEDEC DDR2-800 specs)
  • Reaper HPC PC2-8500 @ Rated (1067MHz, 5-5-5-15)
  • Reaper HPC PC2-8500 @ Highest OC (1143MHz, 5-5-5-15)
  • Reaper HPC PC2-6400 EB @ Stock (JEDEC DDR2-800 specs)
  • Reaper HPC PC2-6400 EB @ Rated (800MHz, 4-3-3-12)
  • Reaper HPC PC2-6400 EB @ Highest OC (1067MHz, 4-5-5-15)
SiSoft Sandra XI Synthetic Tests
Higher levels are better for bandwidth, lower for latency.

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.

Reaper_SANDRA_Bandwidth_Float.png

Reaper_SANDRA_Bandwidth_Int.png

Reaper_SANDRA_Latency.png

As expected, at stock speed, both the Reaper HPC PC2-8500 and the PC2-6400 EB modules scores nearly identically in all three tests. At stock speed, the PC2-8500 performed slightly better than the PC2-6400 EB in the bandwidth tests. A similar trend is observed when both modules are overclocked. The difference in speed between the two memory kits is much more noticable in the latency benchmark where the PC2-8500's small lead increases dramatically.

While both memory kits performed significantly better at their rated speed when compared to their performance at stock speed, overclocking didn't produce a significant performance gain. In fact, the PC2-6400 EB's performance actually took a hit in the SANDRA integer bandwidth test when it was overclocked. However, the performance difference between rated and overclock is within benchmark error margins. These counter intuitive results may be due to a number of factors but overall they illistrate that the system doesn't necessarily benefit from very high memory speeds when the front side bus speed isn't increased.

Transparent
PCMark & Gaming Performance
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." 

Performance Comparison with PCMark05
Overall Memory Score
Reaper_PCMark05_Mem.png


PCMark05 paints a more intuitive picture than SANDRA, with higher speeds equating into better performance. There is a notable increase in performance from JEDEC DDR2-800 specified stock speed and the OCZ rated speeds. Overclocking gave a consistant positive result, although it was very minor.

In-Game Performance Comparisons
System Memory Affects Framerates?  You Betcha!

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.

Reaper_FEAR_Low_Quality.png

We can clearly see the effect of high memory speeds with F.E.A.R. The faster speed offered by the rated specifications over JEDEC's standards results in a 5-7 fps performance gain. As with the other benchmarks, overclocking saw almost no tangible improvement in performance. There also wasn't much difference between the Reaper HPC PC2-8500 and the PC2-6400 EB. They posted essentually identical scores regardless of whether they were overclocked or not, once again due to the fact that we didn't overclock the CPU and FSB simultaneously with the memory.

Transparent
Conclusion
Performance Summary: The OCZ Reaper HPC put forth a good showing with consistantly high benchmark results. The benefit of performance memory, like the Reaper HPC, could clearly be seen when compared to "standard" memory running at JEDEC's DDR2-800 standards. Although the performance benefit was sometimes very small, the Reapers aren't at fault. Overall, both of our Reaper kits performed as expected and on par with other high-performance memory kits we've seen.

The use of heatpipes for cooling memory is a recent development. Memory generally doesn't require heavy duty cooling and the simple, flat heat spreaders that dominate the memory landscape are sufficient for keeping temperatures in check. While this has been true for many years, the recent trend towards high performance memory with stock voltages exceeding 2.2V is pushing the thermal output of memory higher than before. As a result we have seen several manufacturers move to new heatspreader designs for their high-end parts.

After their success with the innovative Flex XLC product line, OCZ has succeeded again in producing yet another innovative cooling solution. Although we wished the heatpipe had been implimented differently, the Reaper's HPC remains an effective cooling system and the heatpipe does get warm during use, indicating that it's doing its job. While not quite as potent as the Flex XLC, the Reaper HPC series packs similarly high performance but at a slightly lower price point. Reaper HPC kits can be found for a bit less than comparable Flex XLC kits and the main difference is the cooling solutions they use. If
watercooling doesn't interest you much, the Reaper HPC series is the better buy.

Overall, we thought Reaper HPC series memory performed well, looked great and possibly best of all, they are priced right. The top of the Reaper HPC line-up clocks in at PC2-9200 and a 2GB kit can be yours for $219.50. The excellent PC2-6400 Enhanced Bandwidth edition modules we tested are an even better deal, coming in at just $177.00. Those of you who want the slick Reaper look without the price are in luck because there is a standard non-EB version of the PC2-6400 memory and 2GB of it can be had for the affordable price of $129.99. If your looking for high performance RAM that stands out in the crowd of boring heatspreader clad memory, give OCZ's Reaper HPC line-up a good, hard look.

  • Innovative Cooling
  • Looks Great
  • EPP Support
  • High Performance
  • Tall Modules May Be Troublesome in Some Configurations

Discuss This Article and Share Your Thoughts With Others!



Content Property of HotHardware.com