Best budget budget and enviro-conscious PSU?

tcaudilllg2 · Post by **tcaudilllg2** » Fri Mar 21, 2008 1:33 am

Which means, lowest wattage and highest efficiency at the least price?

casualsax3 · Post by **casualsax3** » Fri Mar 21, 2008 1:38 am

Go with an Antec Earthwatts. The 430w is a nice sweetspot - saved me $40 a month on my power bill, and they're not too expensive.

tcaudilllg2 · Post by **tcaudilllg2** » Fri Mar 21, 2008 6:42 am

I've got a 300W supply. (and an Athlon XP Thunderbird rig). Would I actually be saving electricity by buying a more efficient power supply (>80%) even if it used more watts than than my existing supply? How so?

Edit: Checked it out.

http://archive.slickdeals.net/showpost. ... stcount=20

Power from the wall = power delivered / efficiency

650 = 455 / 0.7

To give you some idea of the impact of efficiency for home users...

Say you have 2 supplies, each asked to deliver 300 W (which is on the high side, but never mind that) average, continuously every day. One is 70% efficient, the other 80% efficient.

P1 = 300 / 0.7 = 429 W
P2 = 300 / 0.8 = 375 W

Pdiff = P1 - P2 = 54 W

So the more efficient PSU will use an average of 54 W less power. Let's multiply that out over the course of a year to get kilowatt-hours in a year:

(54 * 24 * 365) / 1000 = 473.04

Typical cost per kilowatt-hour of electricity is in the neighborhood of 7 cents or so, depending on where you live.

473 * .07 = 33.11

So in a year, assuming you ran the system full-tilt continuously, you'd save $30. With a more realistic scenario for a home user, about half that. With the system in your OP, more like a third.

By that logic, an 85% supply would save:
300 / 0.85 = 353W, a savings of 22W over the 80%

(22 * 24 * 365) / 1000 = 192
192 * .07 = $13.44/year savings over the 80%

The real comparison of course, is between 85% and 70%, in which the savings come out to be $46, a hefty chunk of the price of the supply.

What it means over the course of the year, in taking the entire supply wattage into account:
((353 * 24 * 365) / 1000) * .07) = $216.46
Compare with wattage at 70%:
((429 * 24 * 365) / 1000) * .07) = $263

The cost of running a computer is collapsed by 18%.

I.S.T. · Post by **I.S.T.** » Fri Mar 21, 2008 11:15 am

I'm using an Earthwatts 380. Given the PC you're running, 380 watts is far more than enough. It's also fairly cheap. They only cost 60 bucks plus shipping: http://www.newegg.com/Product/Product.a ... 6817371005

Getting this at a retail outfit would cost a fair bit more.

Gil_Hamilton · Post by **Gil_Hamilton** » Sat Mar 22, 2008 4:00 am

tcaudilllg2 wrote:I've got a 300W supply. (and an Athlon XP Thunderbird rig). Would I actually be saving electricity by buying a more efficient power supply (>80%) even if it used more watts than than my existing supply? How so?

The simple answer is that the wattage rating on a power supply is a MAX wattage. They only put out as much power as the system needs.

But efficiency varies with load, and different supplies have different "sweet spots."

tcaudilllg2 · Post by **tcaudilllg2** » Sat Mar 22, 2008 7:08 am

Gil_Hamilton wrote:
tcaudilllg2 wrote:I've got a 300W supply. (and an Athlon XP Thunderbird rig). Would I actually be saving electricity by buying a more efficient power supply (>80%) even if it used more watts than than my existing supply? How so?
The simple answer is that the wattage rating on a power supply is a MAX wattage. They only put out as much power as the system needs.

But efficiency varies with load, and different supplies have different "sweet spots."

Where can I find a list of these "sweet spots"?

Gil_Hamilton · Post by **Gil_Hamilton** » Sat Mar 22, 2008 7:28 am

tcaudilllg2 wrote:
Gil_Hamilton wrote:
tcaudilllg2 wrote:I've got a 300W supply. (and an Athlon XP Thunderbird rig). Would I actually be saving electricity by buying a more efficient power supply (>80%) even if it used more watts than than my existing supply? How so?
The simple answer is that the wattage rating on a power supply is a MAX wattage. They only put out as much power as the system needs.

But efficiency varies with load, and different supplies have different "sweet spots."
Where can I find a list of these "sweet spots"?

A decent power supply review will attempt to determine the efficiency at various levels.

http://www.silentpcreview.com/section10.html tends to do very good work, IMO.

Obviously, their focus isn't on full-on performance, just based on the site name. But they DO try to give accurate and detailed reviews, and at least acknowledge the higher-performance part of the market. Hell, they even reviewed a kilowatt supply, which is completely and utterly absurd from every single consumer-level viewpoint.

And their efficiency measurements are comprehensive across the supply's entire range, which is the important part for this discussion.