Crucial Ballistix 4GB DDR3-1866 BL2KIT25664FN1869
Feb 16th, 2011 | By JaredDetailed below are the specifications of the system used during testing:
- Control
- CPU: Intel i5 760
- MB: MSI P55-GD65
- GPU: HIS HD 5870 iCooler V Turbo 1GB
- HD: WD 250GB SATA
- Case: Cooler Master HAF932
- PSU: Ultra X3 1000W
- OS: Microsoft Windows 7
All benchmarks and tests were executed five times and an average was taken to derive the most accurate results. The tests will be conducted at stock frequency (935 due to multipliers), highest stable frequency at CAS 9 and stock frequency with the tightest possible stable timing.
- Memory & Frequency
- Crucial BL2KIT25664FN1869 – 9-9-9-24 @ 1000 MHz / DDR3 2000
- Crucial BL2KIT25664FN1869 – 9-9-9-24 @ 935 MHz / DDR3 1870
- Crucial BL2KIT25664FN1869 – 8-9-8-23 @ 935 MHz / DDR3 1870
Now that we have that out of the way, let’s take a look at the first set of benchmark results. Everest consistently delivers good information about memory performance.
You can see here that upping the clock naturally returns better numbers here while there is almost no gain by small adjustments with timings at stock speeds. The only real odd result is slightly worse results with lower timings though it is such a small number we can call it a wash.
Next up, FutureMark takes the stage for testing with 3DMark Vantage and PCMark Vantage.
Within 3DMark Vantage there isn’t too large of a gain in overall score but we see a large jump in the CPU score when the Ballistix kit is run at 1000 MHz. The boost we got in tightening the timing was minimal at best. Oddly enough, with PCMark Vantage the performance difference by overclocking and or tightening the timing was mostly unnoticeable. The overall score increased for both overclocked settings but the memories score was not what I expected. The other applications were a mixed bag though oddly we have the largest gains in productivity and communications scores.
Last up is HyperPi. This modified version of SuperPi allows us to run multiple threads at the same time. It still calculates the value of pi to a ridiculous decimal place. Keep in mind a lower time is better.
HyperPi returns results of what I would expect, a slight gain with lower timings and a larger gain when the speed is overclocked.