Welcome Stranger to OCC!Login | Register

Mushkin XP3-15000 8-9-8-20 3x2GB Review

ccokeman    -   May 18, 2009
» Discuss this article (4)

Testing:

To find out just how well a product performs, you have to test it in a real world environment so that you don't just blindly believe what the manufacturer says the product will do. Some are right on the money, while others fall somewhat short. On the other hand, there are products that exceed the manufacturer's specifications and will perform at a higher level than what the specifications lead you to believe. To find out what kind of performance Mushkin's XP3-15000 modules deliver, I will be running them through the OverclockersClub suite of benchmarks to see how the performance compares to that of modules at both the rated 1866MHz, as well as some lower rated modules running at 1800MHz as a comparison.

Testing Setup:

Comparison Modules:

CPU-Z: This application shows us the settings that we have chosen in the BIOS. Items shown in this application include CPU speed and bus settings, motherboard manufacturer, BIOS revisions, memory timings, and SPD chip information.

 

Task Manager: We use this utility to show physical memory, kernel memory, page file, and processor usage.

 

Overclocking:

Overclocked settings:

  • Processor: Intel Core i7 920 167x20 = 3348MHz
  • Memory: Mushkin XP3-15000 8-9-8-24 DDR3 2021MHz

Trying to get over the 2000MHz mark was an exercise in futility with the first i7 920 and motherboard I used. Once I moved to a second CPU and motherboard, I was able to eclipse the 2000MHz barrier with the Mushkin XP3-15000 memory, at its rated timings of 8-9-8-24. This, of course, required adjusting the CPU voltage, the memory voltage, subtimings and most importantly, the QPI - or Integrated Memory Controller - voltage. The specifications call for a vDIMM of 1.65v and QPI volts of 1.475v for operation at 1866MHZ with latencies of 8-9-8-24. With these voltage settings, the modules were rock stable, even when tightening the timings to 7-8-7-24; just a slight increase in the memory voltage to 1.75v was required for stability. This is above the specification of 1.65v set by Intel, but the memory speed is above what most people run as well. By dropping the memory speed to just over 1600MHz, the timings were tightened up to 6-7-6-18. At this level I was able to drop the voltage to the modules to 1.70v, with a reduction in the voltage to the memory controller as well. Ultimately, the highest stable point I could reach was 2021MHz at 8-9-8-24 with 1.84 volts to the modules, and 1.665v to the memory controller; both of these voltages are well above what some would call "unsafe" for the life of your hardware. But hey, this isn't computing for the safe, now is it? Nope! Overclocking takes additional volts to most parts of the system, it's just what level of safe are you willing to live with?

 

 

The benchmarks used in this review include the following:

Benchmarks:

  • CPU-Z Version 1.49
  • Windows Task Manager
  • PCMark Vantage
  • SiSoft Sandra 2009
  • Left 4 Dead

 




  1. Intrduction & Closer Look
  2. Specification & Features
  3. Testing: (Setup, CPU-Z, Task Manager, Overclocking)
  4. Testing: PCMark Vantage, Sandra 2009
  5. Testing: Left 4 Dead
  6. Conclusion
Random Pic
© 2001-2014 Overclockers Club ® Privacy Policy

Also part of our network: TalkAndroid, Android Forum, iPhone Informer, Neoseeker, and Used Audio Classifieds

Elapsed: 0.0301959515