Cardioid vs Omnidirectional: One Is a Default, One Is a Switch
MicTested
Two microphones can't settle this comparison, because the shop barely sells the second one. Of the 82 microphone listings collected here, exactly one answers the polar pattern field with omnidirectional and nothing else, and it's a wireless lavalier. Omnidirectional isn't usually a microphone you buy. It's a switch position on a microphone that also does cardioid.




The comparison is between a default and a switch position
Cardioid is what a microphone is unless somebody decided otherwise. The full census of polar patterns across this collection is set out elsewhere, and it's lopsided in a way most comparisons of this kind aren't. The part that matters here is narrower. Of the 82 microphone listings, 12 name omnidirectional anywhere in the pattern field, and eleven of those name it alongside at least one other pattern.
Although twelve sounds like a row, reading them by name rather than by category changes the picture again. Four are inexpensive USB microphones, being the Blue Yeti, the Yeti Nano and two HyperX QuadCasts. Five are studio condensers at four figures, being the Neumann U 87 Ai, U 67 and M 149 and the AKG C12 VR and C414 XLII. One is a camera-mount shotgun. One is the Sony C-800G, and the last is that wireless lavalier. There's no ordinary omnidirectional microphone in the middle of the row, because the row has no middle. Every one of the eleven that switches is a condenser, and not one dynamic microphone in the whole collection offers a second pattern at all, which makes the transducer row the one that decides whether the pattern row can move. The pattern row returns the favour with nothing.
Omnidirectional costs about five times as much
The price gap is the practical answer, but it's larger than the argument usually admits. Half the 39 cardioid-only listings that carry a price cost under $186, whereas half the twelve that can do omnidirectional cost over $874. The cheapest cardioid-only microphone here is $19.90 and the cheapest one that offers omnidirectional is $99.99.
That $99.99 is the Yeti Nano, which puts the pattern switch on a USB body and prints one sensitivity figure for both positions. So the entry price to omnidirectional isn't really $99.99 either, though. It's $99.99 for the switch and nothing whatever for the measurement, since the listing carries no self-noise row at all.
What the switch actually costs, measured on one capsule
Three listings in this collection publish a number labelled with the pattern it belongs to, and those three are the only place the comparison stops being an opinion.
| Microphone | Self-noise, cardioid | Self-noise, omni |
|---|---|---|
| Neumann U 87 Ai | 12 dB-A | 15 dB-A |
| Neumann U 67 | 17 dB-A | 21 dB-A |
| Sony C-800G | 18 dB | 21 dB |
All three agree on direction, though they disagree on size. Switching the same microphone from cardioid to omnidirectional costs between three and four decibels of self-noise, which is the difference between a quiet capsule and an ordinary one. Sensitivity moves the same way on the two Neumanns, from 28 down to 20 mV/Pa on the U 87 Ai and from 24 down to 15 on the U 67, so the microphone gets both noisier and quieter at once.
On a dual-diaphragm condenser the omnidirectional pattern is made by combining two back-to-back membranes rather than by having a native omni capsule, and the combination is what the numbers are paying for. But one figure runs the other way, and it's worth naming. The C-800G's maximum SPL is 131 dB in cardioid and 134 dB in omnidirectional, so what omnidirectional loses in quiet it gains in headroom.
Nine listings out of twelve can't answer the question at all
Only 3 of the 82 listings here label any figure with a pattern name. Everything else prints one number and leaves you to guess which position it describes. That guess isn't a small one either, since on a multi-pattern microphone it falls between two answers that Neumann and Sony have already shown differ by three decibels.
The C414 XLII lists five patterns and gives one sensitivity of −33 dBV and one noise figure of 6 dB-A for all of them. The Neumann M 149 is stranger still. It lists five patterns, prints three sensitivity figures and six noise figures, and labels none of them, so the counts don't even match up well enough to guess.
And the omission reaches the manufacturers. Neumann's own page for the U 87 Ai gives a single sensitivity of 28 mV/Pa and a single signal-to-noise ratio of 82 dB above a line that reads "three polar patterns: omni, cardioid, figure-8". That 28 is the cardioid number from the dealer's table, but nothing on the maker's page says so. Here the shop is the more complete document, which is not how this usually goes.
The words themselves don't hold still
One card in the set names the forward pattern three different ways on one page. The Sony C-800G's specification row reads "Omnidirectional / Unidirectional", its description says "Cardioid and unidirectional polar patterns are user switchable at the front", and its self-noise and SPL rows are labelled Cardioid and Omni. Unidirectional is a category containing cardioid, so the row and the prose are answering at different levels of the same question.
The Shure MV5's card hedges the same way, answering "Unidirectional (cardioid)" and naming both levels at once, which is the only entry in the set that admits the two words sit at different levels. The counting problem goes further up. AKG's own page for the C414 XLII advertises "nine selectable polar patterns" while the listing for that microphone records five, and the page won't show its specification table even with the tab clicked, so the nine can't be checked against anything.
Which one the recording asks for
Cardioid rejects what's behind it while omnidirectional doesn't, and every practical difference follows from that one sentence. A cardioid in a poor room records less of the room; an omnidirectional in a good room records the room on purpose. A cardioid on a boom near a wall rejects the wall, and on the SM58, whose listing does print the angles, the rejection reaches 20 dB directly behind; an omnidirectional doesn't care where you point it, which is why the one dedicated omni here is a lavalier clipped to a person who keeps turning their head.
What the specification sheets add to that is a price, and it's payable twice. Once in money, at roughly five times the median, and once in noise, at three to four decibels on the same capsule. Neither is a reason to avoid omnidirectional. Both are reasons to know which position the number on the card was measured in, and on 79 listings out of 82 you can't.
What the cards answer about cardioid and omnidirectional
Is omnidirectional better than cardioid?
Neither is better. Omnidirectional records the room and cardioid rejects it, and on the same capsule omnidirectional is three to four decibels noisier.
How many omnidirectional microphones are there here?
Twelve listings name the pattern, though only one names it alone, and that one is a wireless lavalier. The other eleven switch between patterns.
What is the cheapest microphone here that does omnidirectional?
The Blue Yeti Nano at $99.99 new, or $59.95 in the same shop's used department. It has two positions and publishes one sensitivity figure for both.
Does a specification sheet say which pattern its numbers describe?
Rarely. Three listings out of 82 label a figure with a pattern name, and one manufacturer page publishes a single number for a three-pattern microphone.
Is unidirectional the same as cardioid?
Not quite. Unidirectional is the category and cardioid is a pattern inside it, and two cards here use both words as if they were interchangeable.
What the comparison comes down to
Omnidirectional is rarely a product and usually a setting, so the honest form of this question isn't which microphone to buy but which position to select. The listings answer the first question easily and the second one almost never.
Yet three of them do answer it, and all three say the same thing. The pattern you choose changes the numbers on the sheet, and the sheet almost never tells you which pattern it was describing.
Sources
- B&H Photo Video's listing for the Neumann U 87 Ai, read row by row: the polar pattern row giving Cardioid / Figure-8 / Omnidirectional, sensitivity given as "20 mV/Pa (Omni) / 28 mV/Pa (Cardioid) / 22 mV/Pa (Figure-8)", and equivalent noise level as "15 dB (Omnidirectional, A-Weighted) / 12 dB (Cardioid, A-Weighted) / 14 dB (Figure-8, A-Weighted)".
- Neumann's own product page for the U 87 Ai: the line "three polar patterns: omni, cardioid, figure-8" alongside a summary block giving one maximum SPL of 127 dB, one signal-to-noise ratio of 82 dB and one sensitivity of 28 mV/Pa for the microphone as a whole.
- B&H Photo Video's listing for the Sony C-800G, with the specification table open: equivalent noise level of "18 dB (Cardioid) / 21 dB (Omnidirectional)", maximum SPL of "131 dB (Cardioid) / 134 dB (Omni)", a polar pattern row reading "Omnidirectional / Unidirectional", and a description calling the same two patterns "Cardioid and unidirectional".
- B&H Photo Video's listing for the AKG C414 XLII: five polar patterns in one row, with a single sensitivity of −33 dBV and a single equivalent noise level of 6 dB A-weighted covering all of them.
- AKG's own product page for the C414 XLII: the claim of "nine selectable polar patterns", against the five its dealer listing records, on a page whose specification table did not render on either of two visits.
- B&H Photo Video's listing for the Blue Yeti Nano: $99.99 new with a used unit at $59.95, a polar pattern row of Cardioid / Omnidirectional, one sensitivity of "4.5 mV/Pa (at 1 kHz)" for both, and no self-noise row.
- Pattern counts, price medians and row sizes come from listings of 200 audio products collected the same way. How those are read and compared.
All six pages were read on 27 August 2026.
What we did not doNo microphone was switched between patterns for this page. Every figure above is a manufacturer's or a dealer's, and what we add is the count: how many of the microphones collected here can do omnidirectional at all, what that costs, and how few say which pattern their numbers describe.