The number one goal when choosing a power supply is to select a model with sufficient wattage to power the system utilizing it with stability. Currently, a high-end system with Intel Core i7 CPU, an X58 based motherboard, two GTX 295 graphics cards in SLI will draw around 800W at full load. Normally the choice for such a system will be a 1000W power supply, which is sufficient in providing full stability. So we often hear the question, "do I really need a 1500W power supply?" If you are using a system such as the one we just described, then the answer is “yes!” When high-level computer enthusiasts are looking for a power supply, sufficient wattage is only one of the basic requirements. There are other important requirements that warrant considerations as well.
Having high efficiency means reduced power consumption and electricity cost. If we use the above described high-end system as an example, equipping it with a 1000W power supply should be sufficient, but let’s take a look at the below efficiency graph in detail:
Nearly all power supplies achieve their highest efficiency during 50% loading condition. The graph shows two power supplies with 80 PLUS Silver level efficiency with the only difference between the two being their wattage rating, one is 1000W and the other is 1500W. We can clearly see that at around 800W loading, the 1500W power supply is more efficient, therefore drawing less power. Some may argue that during system idle, the 1000W power supply is more efficient so overall power draw is still lower than 1500W. But if we calculate in detail by using the aforementioned high-end system as an example, which has idle power draw of around 100W to 200W, the 1000W power supply will be more efficient by 1% to 2 %. This translates to around 3W to 1.5W of saving (100/0.81~0.82 – 100/0.8 = 3~1.5) compared to the 1500W power supply. Doing the same calculation with 800W system power draw, the 1500W power supply is more efficient by also 1% to 2%, which translates to around 21W to 10.5W of saving (800/0.88 – 800/0.86~0.87 = 21~10.5) over the 1000W power supply. So the power saved at full load using a 1500W power supply is at least seven times greater than what a 1000W power supply can save during system idle. Even if an user spends more than 2/3 of his/her time idling and only 1/3 of the time at full or high loading condition (such as playing games), the 1500W will use less electricity overall.
Modern high-end power supplies have vastly improved efficiency compared with ones produced just a few years ago, which has resulted in noticeably lower heat output. These power supplies are usually coupled with large diameter fans, so they need very low fan speed to cool properly, thus resulting in lower operating noise as well. But as loading increases on these power supplies, their internal heat output will increase proportionally so they are equipped with fan controller to change speed in response to heat output.
If we compare a 1000W power supply with a 1500W power supply with similar fan control scheme (fan start ramping up at 60% loading) on a system drawing around 800W, we can produce the following graph to compare the expected noise level:
At idle, both power supplies have the ability to stay very quiet, but at around 800W, which corresponds to 80% power draw for a 1000W power supply, the fan has already ramped up noticeably higher and beginning to sound noisy. The 1500W power supply, on the other hand, is still operating at around half of its total capacity so the fan does not ramp up. This shows that you can expect a 1500W power supply to be as quiet during full load as it does in idle condition when powering a typical high-end system.
Component temperature rises proportionally with increased loading so components inside the power supply are exposed to constant high temperature if subjected to high loading conditions. High temperature causes faster aging of components and reduced overall lifetime. Using the 1000W and 1500W power supplies as reference again, we set the expected maximum operating temperature to 90% of the component temperature rating (component temperature rated at 85℃, expected max.
Operating temperature at 76.5℃) to compare their temperature rise in the graph:
When drawing around 800W, the 1000W power supply’s internal component temperature increases to 68℃ (80% of the expected max. operating temperature) and under the same condition, the 1500W power supply’s component will only rise up to 59℃ (69% of expected max. operating temperature). So with system power draw at the same level, we can expect the 1500W power supply to have longer component and overall lifetime.
From the above summaries, we believe the use of higher wattage power supplies have tangible advantages for computer enthusiasts or professionals using high-end system builds. The results will nearly always favor higher wattage power supply running at half of its capacity compared to lower wattage unit running into higher power levels.
For more information on the SilverStone’s 1500W power supply, the Strider ST1500, please visit the following link:ST1500