PSU Efficiency Ratings and What They Change

The badge certifies one property, tested at a few load points: how much of what leaves the wall reaches your components. It says nothing about ripple, protection circuitry or build quality — and those are what decide whether a supply is any good.

A note on this guide: this rests on what the certification programmes measure and on how efficiency curves behave, rather than on units measured here. See how we test.

The problem, as it appears

Two supplies at the same wattage, different badges, meaningfully different prices. The higher tier is obviously better, and the question is whether it is better enough to pay for.

Framed that way the answer is usually no — and framed correctly the question turns out to be about something else entirely.

Why the tier disappoints as a quality signal

Because it measures efficiency, and people read it as a proxy for everything.

The certification tests how much power is lost between the wall and the components at specified load points. That is a real, tested, useful figure. It is also the entire scope.

What it does not cover: how clean the delivered power is, whether the protection circuitry actually works, what components are inside, and how the unit behaves at the moment something fails. Those are the properties that determine whether a supply protects the machine or takes it with it.

So a highly certified unit from a brand with no track record is exactly what the badge says — efficient — and nothing more.

What actually matters

The saving is smaller than the price gap suggests

Efficiency waste is the difference between what the wall supplies and what the components get, and it shows up as heat.

On a domestic machine that idles most of the day, moving a few percentage points is a small annual figure. Whether it repays a price difference is arithmetic you can do — but only once you know what the machine actually draws, which is a measurement rather than a guess.

This page contains affiliate links. If you buy through them, Ugly Fixes may earn a commission at no extra cost to you. Products are selected independently.

The method is the same one used for UPS sizing in how much power a gaming PC actually uses.

The heat argument is the better one

Waste does not vanish — it becomes heat inside the case, which the case fans then have to remove.

A more efficient unit adds less to that load and generally runs its own fan less as a result. That is a small, real, everyday benefit, and it is more persuasive than the electricity saving for most people — the fan side of it is in PSU fan noise and zero-RPM modes.

Efficiency peaks near the middle

The curve is not flat. Efficiency typically peaks around half load and falls away at both ends.

That is the argument for sizing so ordinary use lands near the middle of the unit's rating — a heavily oversized supply spends its life down at the inefficient end, which quietly undoes some of what the higher tier bought. The full sizing argument is in sizing a PSU without guessing.

The correlation is real, and it is a correlation

Manufacturers do tend to pair higher certification with better capacitors, better protection and better build, because a unit engineered to hit a demanding efficiency target is generally engineered carefully throughout.

That makes the tier a useful hint. It does not make it a guarantee, and the distinction matters when comparing an established brand's lower tier against an unknown brand's higher one.

What the badge cannot tell you

Protection circuitry — over-current, over-voltage, short-circuit — is what stands between a fault and the rest of the machine, and none of it is measured by efficiency certification.

This is the site's standing rule applied to the clearest example on it: a component whose failure takes other things with it is where you spend more, and the number that justifies spending more is not the efficiency tier.

Upstream matters too

Efficiency describes what happens inside the unit. What arrives at it is a separate question, and a supply cannot protect against everything on the incoming side.

The decision, plainly

• Buying new, similar prices → take the higher tier, for the heat and the correlation
• Buying new, large price gap → spend it on brand and protection, not on the badge
• Existing supply works fine → do not replace it for efficiency
• Machine idles most of the day → the saving is small; measure before assuming
• Unknown brand with a high tier → efficient, not necessarily safe

The mistake to avoid

Reading the tier as a quality score. It is a tested claim about one measurable property, and it has become shorthand for "good supply" because it is the only number on the box that looks technical. The properties that decide whether a supply protects your machine are not on the box at all.

The second mistake is upgrading a working supply to a higher tier. The saving is small, the old unit is not recovered, and the risk of the swap — a machine that will not POST afterwards, and a new suspect list to work through — is not nothing. Efficiency belongs in a purchase decision, not in a replacement one.