Installation and Testing:
 

Using the very trusty Abit AB9 Quad GT (I just can't recommend this motherboard enough) SLI Motherboard, I powered down and cleared the CMOS. Starting with a fresh BIOS setting, the system booted right up in AUTO mode with the memory defaulting directly to its 1066MHz specified frequency. This in turn clocked the E6600 up to 3.2GHz right out of the gate. Typically, I like to run stock 800MHz frequency benchmarks, but this isn't that kind of memory. It's rated to PC2-8500 or 1066MHz, so that's where I will test it first off.

CPUZ reported the correct stock operating timings that the Reaper kit uses. The last OCZ kit I had the pleasure of testing unfortunately didn't pass CPUZ muster unable to overclock to anything respectable, nor reporting correct timings and proved to need manual RAM timings. The OCZ Reaper modules didn't need any manual adjustments except where extreme tweaking required. One other nice convenient bonus was how the motherboard and memory worked so well together from 800MHz to 1000MHz. The system automatically detected the tighter frequencies and adjusted the voltages accordingly. It's so nice when the two major components that affect the stability of a system work so well.

The system appeared stable for about 20 minutes during initial testing until it suddenly rebooted without warning. Rebooting in to Windows, I immediately ran CPUZ only to discover that the memory was clocking in at 1066MHz with 4-4-4-15 memory timings!! What's this? Did I accidentally get some VX2 memory? The memory is rated for 5-5-5-15 at 1066MHz, but not with lower, more aggressive timings. The BIOS was set to "By SPD" in the memory timing setting as well as the processor which is the reason the system insisted on automatically running at 383MHz Front Side Bus. To see if the system was fully stable, I raised the CPU and Memory voltages to see if the system could be stabilized. The system rebooted fine and then used to benchmark the memory bandwidth, processor, and Battlefield 2142. The system remained stable and allowed for several hours of torture and a few rounds in game. Next, the 3DMark05 benchmark was used to get the overall system score.

Just to be thorough, the memory was down clocked to 1000Mhz, 900MHz, and 800MHz where it performed perfectly using 4-4-4 and 5-5-5 memory timings. Orthos, again, didn't report any errors under 1000 to 1060MHz frequencies using stock voltages. At a manually set 1066Mhz with stock memory voltages, the memory seemed a bit unstable. The system wouldn't crash, but I did notice a few oddities and signs of some form of instability here and there. It also could be my perception after some time staring at the screen entering and exiting program after program. None the less, the performance of the PC2-8500 Reaper  kit straight out of the package is still quite impressive. It's obviously aggressive memory with very good modules.

Memory latency performance was very good and was able to best the PC2-9200 kit due to its default 5-5-5 timings. It is miniscule as scores go varying only just a little, but it indicates how well lower latency modules benefit the PC2-8500 Reaper's performance. SANRA's memory bandwidth benchmark indicate a performance gain of only about a 100MB/s which equates in to a little more system performance gains sometimes noticeable in games such Fear where average frames per second are calculated. You may not notice them normally, but to those of us who like the quick, responsive, and extra few frames per second in a video game, we definitely appreciate the gains.

So, how does the kit perform compared to another pair of PC2-8500 memory modules? The results are different in that the Reaper kit managed to out perform because of the tighter timings. Resetting the Reaper kit to it's 5-5-5-15 timings, scored almost identical results compared to the PC2-9200 memory at equal frequencies. Under Auto and stock settings, the Reaper kit does manage to pull ahead just a little which again may not be completely noticeable to everyone. It does count for something especially among the hardcore enthusiasts expecting extreme stability and performance. Besides, if it says "Reaper" on it, it had better be pretty bad @$$!

Conclusion:

The OCZ DDR2 PC2-8500 Reaper HPC Memory kit definitely flaunts its ability to run at 1066MHz at default settings. Performance seemed very crisp and every setting was stable. It's as though the memory works or doesn't as there didn't seem to be any middle ground with the kit. Of course, performance could be a joint result between OCZ and Abit, but that only means this is a deadly duo. The memory performed very similarly on the 650-SLI platform that I have been using for the past week. The only thing that seems could make this particular kit any better would be to grab a PC2-9200 kit instead. The modules are a grade above the PC2-8500 kit and should perform exactly the same except where the 1150MHz performance gains are concerned. 

There are several positive points one can make when choosing the OCZ DDR2 PC2-8500 Reaper HPC Memory kit. Some of the best features are the cheaper and more sensible RAM sinks utilizing the heat pipe design which help to keep the price down. The two most obvious benefits of the Reaper memory is the new, aggressive look and performance beyond specs reaching 1066MHz with 4-4-4 memory timings rather than the default 5-5-5 memory timings. Sure, it isn't to spec, but who's complaining?  No one expects anything to operate above or beyond the specs of their components, but it's really great when something exceeds expectations. Will your kit perform the same? I think the answer comes in the form of two names, OCZ and Reaper. You can expect OCZ to know when to use the right modules for the right name. All things considered, I have to say these are an impressive set of memory modules and worthy of a fellow enthusiast's recommendation.

 

<< A Closer Look | VH Main Page >>