source : tomshardware

PC Power Consumption Has To Be Observed Over Time

Tom's Hardware was one of the first publications to not only criticize ridiculously high heat dissipation, but also excessive power consumption where it's totally unnecessary. AMD and Intel have been in the news for years because their products either weren't efficient enough, or because of the advances they both made in reducing their power requirements. Measuring minimum (idle) and maximum power requirements of components, or of the overall system at the plug, is a good way to determine whether or not a device is thrifty. However, these measurements leave an important factor out of the equation: performance.

The delay of new processor products Phenom X2 and X4 forced AMD to shift its focus from "look at how fast our products are" to "look at the value we provide". The reason is Intel's Core 2 processor family beating the daylights out of the aged Athlon 64 family: on average, Core 2 Duo is not only faster, it's also more efficient than the Athlon 64 X2. It's not as bad as it sounds for everyone, though, as a decent Athlon 64 X2 system is still fast enough for the majority of users. A typical system certainly still provides good value, but the enthusiast crowd typically hasn't had much of a reason to saddle the foundering Athlon horse.

Intel has the advantage of having the faster and more efficient processors, as it has made several advances while adjusting and fine-tuning processor manufacturing. Compared to the first Core 2 Duo generation, at an idle power of up to 24 W, today's processor steppings are clearly more efficient and stay at around 10 W. In addition, most Core 2 processors, on average, are more overclockable than Athlon 64 X2 models.

Let's not forget that the processor is only one component inside the PC. Other hardware items also require power and add to the total power consumption, which eventually is also greatly influenced by the power supply. These components include the motherboard and chipset, the main memory, the graphics card - which can consume more than a high-end processor - the hard drive, optical drives and expansion cards. High-efficiency power supply units reach an energy efficiency of almost 90%, while average products stay at below 80%. This means that the remaining power is converted into heat, which is useless for you, even though you have to pay for that waste on your electricity bill. Knowing this, the impact of the CPU on the overall system is put into new perspective.

Our take is as follows: power consumption measurements of system idle power, or while a system is busy working on a heavy workload, create a snapshot of minimum/idle power and of the maximum power requirement. As already mentioned, though, this will only help give a feeling for these two extremes. The idle power is the baseline, which people use to start measuring power consumption. Meanwhile, the system or component performance is not considered at all - but it should be, as real-life application scenarios will always create a certain workload. This will always put components into a more energy-demanding state, but also result in quicker completion of a workload when faster hardware is used.

AMD Vs. Intel: Who Is The Real Winner?

The answer to this question certainly depends on the system configuration. On average, an Intel motherboard will require more power than an AMD board, as the chipset includes the memory controller, which AMD integrated into its processors. This happened for performance reasons, but the side effect clearly is better efficiency. However, the differences in voltage regulator layout and high-end vs. budget chipsets on a motherboard can also negate my statement - especially as you go into the high-end, where motherboards are filled with additional components.

If we assume that all other components can be used for either platform, differences can only be found on the processor and motherboard side. The power supply, RAM, the hard drive, and the graphics card are standard components for both AMD and Intel systems.

The question now is: how will the power consumption be if we assume certain usage scenarios? It is not difficult to predict this for systems that run idle or when they're at full load, but this does not reflect typical PC usage. Instead, we decided to use SYSmark 2007 Preview to simulate typical user behavior, and looped runs of 3DMark06 and PCMark05 to stress the system a bit more.

We didn't select high-end processors, because Intel's Core 2 Extreme would leave the AMD Athlon 64 X2 family in the dust. We decided to use reasonable mainstream processors that are known to be somewhat energy efficient: an Athlon 64 X2 5000+ using the firm's 65 nm manufacturing process, and an Intel Core 2 Duo E6400.

Performance Benchmark Results

We only ran a few typical benchmarks to show the differences between our AMD and Intel test systems.

3DMark06, PCMark05

The 3D standard benchmark 3DMark06 shows similar performance for the two systems.

In the overall score..., the Intel system looks better.