Benchmark suite

As soon as we ensure full stability on all of the cas settings as shown on the previous page, 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 800Mhz DDR

Before we get started, a quick roundup of the timings at 800Mhz. As the OCZ kit did not manage full stability at cas 3 (it did come close though at 795Mhz DDR cas 3), we had to lower our timings a little. Nor a CAS latency of 3, nor a TRCD latency of 3 ensured us full Orthos stability, so in the end we had to settle for 4-4-3-12 (CAS-TRCD-TRP-TRAS) timings to perform our tests at 800 Mhz.

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.

In all our graphs, please look at the blue bars for the OCZ PC8500 kit, and the yellow bars for the top performer(s) in each test.



The differences you see here are limited, as all tests are performed on the same front side bus and CPU speed.. The small difference you do see is dependent on the cas latency of the memory modules at 800Mhz DDR. The entire top of the memory pack are modules that are capable of running a cas latency of 3 cycles at 800Mhz, the bottom of the pack has to run at 4 cycles. Our OCZ memory kit of the day has to run at 4 cycles, giving a little lower result than the other kits. But as we mentioned, the OCZ kit did come very close to cas 3 performance at 800Mhz, and could probably finish all but the most demanding benchmarks at 800Mhz cas 3 as well.


Latency at 800Mhz DDR

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



When we look at latencies, the cas 3 vs. cas 4 differences are again pronounced in the same way as the bandwidth test. Our OCZ kit is again part of the bottom group with a setting of cas 4 at 800Mhz, but at the top of the bottom group, probably due to its TRP setting of 3 cycles.


Benchmark tests at 800Mhz settings

In this section we take a look at the performance of the memory modules in the above mentioned benchmark programs and real life application tests. We start off with the Futuremark 3D Mark 2001SE test :



As you can see, differences are extremely small here. We're talking about a 1,5% difference only between the fastest (Team Group) and the slowest (the PC5300 generic) memory module. Our test kit is in the middle of the pack.



Next up is 3D Mark 2006, where we take a look at the CPU test only, as well as the total scores. In this case the differences are even smaller between the modules: for the cpu tests only the differences are about 1% between the slowest and the fastest kit, for the total 3D06 benchmark differences are less than 0.5%. As with the 3D01 benchmark, the OCZ Modules are in the middle of the chart.



When we look at the PC Mark test of 2005, results are average once more. The differences between the slowest and fastest kit is about 4% in this case.

Now let's have a look at the Cinema 4D bench, better know as Cinebench :



As you can see, differences are almost none existent. The Reaper kit is at 50s encoding time, the same as most other memory kits.

Time for some Superpi :



Here we see the influence of the slightly higher cas latency of the PC8500, as it sits at the bottom of the older Micron pack, but faster than the Kingston PC8500 micron kit. The slightly tighter TRP latency could be the reason for that. Differences between the slowest and fastest kit is about 5%.


Here are the game tests:



First up is Prey, and Prey tends to give little fluctuations: maximum differences lie around 1.5%. However, our OCZ kit shines here, giving the highest frame rate in the second bench game test of all the kits so far tested.



Finally some F.E.A.R. testing, and this test is clearly giving some more fluctuations than the other tests, as differences up to 7% are possible with this benchmark. The OCZ shines once more with very respectable minimum and average framerates.

Let's have a look how these kits perform on the 1066Mhz memory strap ->