Page 1:LoVo Means Low Voltage
Page 2:Low-Power RAM: Kingston HyperX DDR3 LoVo
Page 3:Test System Details
Page 4:Benchmark Results: Synthetics
Page 5:Benchmark Results: Applications
Page 6:Benchmark Results: Audio/Video
Page 7:Benchmark Results: System Power Consumption And Single-Thread Efficiency
Page 8:Benchmark Results: Multi-Thread And Overall Efficiency
It clearly doesn't make sense to go for a low-voltage memory product as the first (or only) step in reducing power consumption. We found differences of up to 1 W during idle and up to 4 W at peak. Although we used an AMD machine for testing, we also tried an Intel P55 platform with XMP support that allows for automatic memory configuration. The results were very similar. Exchanging the graphics card or the motherboard will likely have a larger impact on power consumption, so it makes sense to look at more power-hungry hardware first and optimize your memory last.
Our results should be considered a worst-case scenario for Kingston’s HyperX DDR3 LoVo, as the overall system power is so high that the LoVo's potential power savings get lost in the mix. This would apply to any mainstream or high-end system requiring 70 W or more idle power. Spend your money on other components before going after low-voltage RAM, because the impact will certainly be more noticeable.
Once you optimize your system with low-power primary components (CPU, motherboard, drives, etc.), the only way to further decrease power consumption will be through tweaking and undervolting, which can cause reliability issues. As an alternative to this, Kingston’s HyperX DDR3 LoVo provides an additional step for further reducing system power and temperature. It’s like life in general: getting closer to perfection comes at a price.
- LoVo Means Low Voltage
- Low-Power RAM: Kingston HyperX DDR3 LoVo
- Test System Details
- Benchmark Results: Synthetics
- Benchmark Results: Applications
- Benchmark Results: Audio/Video
- Benchmark Results: System Power Consumption And Single-Thread Efficiency
- Benchmark Results: Multi-Thread And Overall Efficiency