“Hard to drive” is a sensitivity problem, not an impedance one

The phrase is almost always attached to the wrong number. Ask which headphones are hard to drive and you will be told about 300Ω Sennheisers and 600Ω Beyerdynamics. Ask this database the same question — 280 sources, 567 headphones, every figure traced to whoever published it — and the answer comes back the other way round.

Of the twelve models here that the fewest sources can drive to 100 dB SPL, 9 are below 50Ω and none is at 250Ω or above. The hardest headphone in this database to reach 100 dB on is a 3Ω in-ear.

Read the last three columns across. These are not voltage-hungry headphones; they are current-hungry, and the current column is the one that sorts them. The Symphonium Titan asks for 187 mA at just 0.56 V — a voltage any phone can produce and a current most cannot.

The number that decides it

How much power a headphone needs for a given loudness is a function of its sensitivity and of nothing else. That is arithmetic rather than a finding — impedance does not appear in the formula at all.

What is a finding is what happens when you stop asking about the formula and start asking about real equipment: how many of the 280 sources here actually reach 100 dB on each of the 567 models. Sensitivity still predicts that, with a rank correlation of 0.46. Impedance predicts it with a rank correlation of -0.01 — which for practical purposes is none at all.

Sensitivity only gets to 0.46 rather than to one because the second half of the story is in that first table: below about 20Ω an amplifier can run out of current while it still has volts to spare, so a few low-impedance models fall further down the list than their sensitivity alone would put them. That effect makes low impedance harder, not easier — which is still the opposite of the folklore.

Impedance does not decide how much work an amplifier does. It decides what form the work arrives in. Same watts, more ohms: more volts, less current. Same watts, fewer ohms: more current, fewer volts. A source runs out of one or the other long before it runs out of watts, and which one it runs out of first is a property of the source, not of the headphone.

Two headphones that need the same power

These two want the same thing in watts and nothing like the same thing in everything else. The impedance ratio between them is 25 to one.

The Sennheiser HD 600 needs 1.41 V and 4.7 mA. The HiFiMAN Edition XV needs 0.28 V and 22.9 mA — 5.1× less voltage at 4.9× the current. And the count in the last column goes the way nobody expects: more sources in this database reach 100 dB on the 12Ω model than on the 300Ω one.

The comparison that settles it

Take the two ends of the impedance range and ask how many of the 280 sources here drive every model at that end:

• 261 of 280 reach 100 dB on all 30 models at 250Ω or above.
• 85 of 280 reach 100 dB on all 358 models at 32Ω or below.

And the asymmetry is total: not one source in this database manages the low-impedance set while failing the high-impedance one. Every source that copes with the difficult low-impedance models copes with the high-impedance ones as well, which is the opposite of how the folklore runs.

Where the folklore came from, and why it is not nonsense

It was true once, about particular equipment. Output-transformerless tube amplifiers of the 1970s genuinely preferred high impedance, and a lot of them could not deliver current into anything else. The OTL kit in this database still behaves that way — it has an output impedance of 120Ω, which under the one-eighth rule suits a 600Ω headphone and nothing much below 960Ω. That is a real constraint and it is still real today. It is just not the same constraint as “hard to drive”, and it applies to a handful of products rather than to headphones as a class.

The other half of the folklore is true about one particular kind of source, and got generalised to all of them. A phone-class output is voltage-limited: past a certain load it simply has no more swing to give. The Apple USB-C dongle (EU) in this database tops out at 0.77 V however much impedance you give it, which is why it reaches 2.0 mW into 300Ω and exactly half of that into 600Ω. The Sennheiser HD 600 needs 1.41 V, so that dongle does not get there. High impedance is genuinely a problem there — not because the headphone is inefficient, but because the source has a ceiling in volts. Plug the same dongle into a 16Ω in-ear and the ceiling never comes near.

So both halves of the phrase describe a limit of the source. One end runs out of volts, the other runs out of current, and calling either of them a property of the headphone is what turns a solvable question into folklore.

What this page is not claiming

Reaching 100 dB SPL is not the same as sounding its best, and this site has no way to measure the second thing. Nothing here says a high-impedance headphone will not benefit from a better amplifier; it says the benefit is not that the amplifier is finally loud enough.

And the ordering above rests on published sensitivity, which 175 of these 567 models do not have in a usable form — their makers printed a decibel figure and no reference unit. Every model page says which case it is in. The audit counts them.

If you want the full ordering rather than the twelve worst, the shortlist has it, and the matcher answers it for the headphones you actually own.