PSU Wattage Calculator

How this is calculated

START FROM THE TWO PARTS THAT MATTER. A CPU and a graphics card account for most of a build's power, and both publish a figure - TDP, or board power, on the manufacturer's specification page. Everything else together - motherboard, memory, drives, fans, pumps and peripherals - is usually somewhere around 100 W and is not worth agonising over, because it is small next to a graphics card and small next to the headroom you are about to add anyway. ⚠️ TDP IS NOT PEAK DRAW, AND MODERN GRAPHICS CARDS SPIKE HARD. A card rated at 250 W can pull far more than that for a few milliseconds during a load change - transients well above the rated figure are normal, not a fault. A power supply that is technically adequate on paper can trip its own protection on those spikes and shut the machine down, which presents as a random reboot under load rather than as anything obviously power related. This is exactly what the ATX 3.0 specification was written to address, and it is the single strongest argument against sizing tightly. ⚠️ EFFICIENCY PEAKS AROUND HALF LOAD, WHICH CHANGES THE QUESTION. 80 PLUS certification tests at 20%, 50% and 100% of rated load, and the 50% point exists in the standard because that is where power supplies perform best - an 80 PLUS Gold unit is specified at 87% efficient at 20% load, 90% at 50% and 87% at full load. So the aim is not merely to have enough watts; it is to land your NORMAL load somewhere near the middle of the unit's range. AND YOUR NORMAL LOAD IS NOT YOUR PEAK. A system peaking at 475 W does not sit there - gaming might be around 70% of peak and desktop work far less. Sizing so that PEAK equals half the supply would mean buying roughly twice what you need and then running at 25% typical, which is back down the efficiency curve. The typical figure is the one to centre on, which is why it is an input. THE TWO APPROACHES CONVERGE, WHICH IS REASSURING. On the defaults, 30% headroom over peak gives 618 W while centring typical load at 50% gives 665 W - both landing on a 650 W unit. When a rule of thumb and a physical argument agree, the answer is probably right. RAILS AND CONNECTORS ARE NOT MODELLED. A supply with enough total wattage can still be wrong for a build if it lacks the right PCIe connectors, or if a multi-rail design cannot deliver enough on one rail. Check the connector list against your card before ordering, particularly for cards using the 12VHPWR connector. WATTAGE IS NOT QUALITY. A cheap 850 W unit is a worse buy than a good 650 W one - the rating describes capacity, not how cleanly it delivers it or how it behaves when something goes wrong. The 80 PLUS badge speaks to efficiency, not to build quality or protection circuitry.

Common questions

Why not just buy a supply that matches the total draw?

Because the sum of rated figures is not the highest your machine will momentarily pull. Modern graphics cards produce transient spikes well above their rated board power during load changes - lasting only milliseconds, but easily enough to trip a power supply's over-current protection and shut the machine down. That failure looks like a random reboot under load rather than anything obviously power related, which makes it genuinely hard to diagnose. Headroom is not about the average; it is about surviving the spikes, and it is what the ATX 3.0 specification was written to address.

Does a bigger supply waste electricity?

Barely, and much less than people assume - a power supply draws what the components ask for, not its rated capacity. The only real cost of oversizing is efficiency: 80 PLUS certification tests at 20%, 50% and 100% load, and units perform best around the middle of that range. A machine running at 10% of a very large supply sits in the least efficient part of the curve, and an 80 PLUS Gold unit specified at 90% efficient at half load is only 87% at 20%. So an enormous supply costs you a little in electricity and a lot in purchase price, but it does not consume its rating.

What should I aim for as a load percentage?

Around 50% for your TYPICAL load, not your peak. That is where the efficiency curve peaks and it is why the 50% test point exists in the 80 PLUS standard at all. The distinction matters: a system peaking at 475 W does not sit there, and gaming is roughly 70% of peak for most builds while desktop work is far below it. Sizing so that PEAK equals half the supply means buying about twice what you need and then running at 25% typical - which is back down the curve on the wrong side. Centre the typical figure, not the maximum.

The two recommendations disagree. Which do I follow?

On most builds they barely disagree, which is the useful part. On the defaults, 30% headroom over peak gives 618 W while centring typical load at 50% gives 665 W - and both round to the same 650 W unit. When a rule of thumb and a physical argument about efficiency land in the same place, the answer is probably right. If they diverge sharply it usually means your typical-load estimate is unusual, and it is worth revisiting that rather than picking whichever number you prefer.

How accurate does the 'everything else' figure need to be?

Not very. Motherboard, memory, drives, fans and peripherals together usually come to somewhere around 100 W, which is small beside a graphics card and smaller still beside the headroom you are adding on top. Spending time itemising drives at a few watts each is effort that the 30% headroom absorbs many times over. It is worth raising if you are running something genuinely unusual - a lot of mechanical drives, a custom water loop with several pumps, or heavy USB peripherals - and otherwise worth leaving alone.

Is a higher wattage supply a better supply?

No, and the two are easy to conflate. Wattage describes capacity; it says nothing about how cleanly that power is delivered, how stable it stays under load, or what happens when something goes wrong. A cheap 850 W unit is a worse purchase than a good 650 W one for a build that needs 600. The 80 PLUS badge is also narrower than it looks - it certifies efficiency, not build quality or protection circuitry. Reputable reviews that actually load-test a unit are worth more than either number.

What does this not check?

Connectors and rails, both of which can make an adequately-rated supply wrong for your build. A card may need PCIe connectors your supply does not have, or a specific 12VHPWR cable, and a multi-rail design can be unable to deliver enough current on one rail even while the total wattage is fine. Physical size matters too in small cases. Check the connector list against your graphics card before ordering - it is the most common way a correctly-sized supply turns out to be the wrong one.

Take it further with AI

Copy this into ChatGPT or Claude with your own numbers filled in. It hands over the figures this calculator worked out, so the answer is built on real arithmetic instead of a guess.

I used the PSU Wattage Calculator at https://www.bfcbrilliance.com/tools/psu-wattage-calculator.

What I entered:
- CPU power (TDP) (W): ___
- Graphics card power (W): ___
- Everything else (W): ___
- Headroom over peak (%): ___
- Typical load, as a share of peak (%): ___

What it calculated:
- Power supply to buy: ___
- Peak draw of the components: ___
- — before rounding to a real size: ___
- Typical draw in normal use: ___
- Size that centres typical load at 50%: ___

Use those figures as given — they are already worked out, so please don't recalculate or estimate your own. Help me turn them into a plan: what to buy or do, in what order, roughly what it should cost, and the mistakes people most often make with this job.

More power & batteries tools