Most engineers pick stereo mic techniques based on what sounds good. But the difference between XY, ORTF, and Blumlein isn't aesthetic preference. Each array privileges a different psychoacoustic mechanism, and your ears use two completely separate systems to localize sounds. Knowing which system each technique activates explains results that would otherwise seem arbitrary.
Two Localization Systems, One Pair of Ears
Your auditory system uses two independent mechanisms to place sounds in space.
Interaural Time Difference (ITD): The delay between when a sound arrives at your left ear versus your right. For a source directly to your left, that delay reaches about 660 microseconds, the time for sound to travel around your head (approximate head radius: 87.5 mm). The auditory system can detect differences as small as 10 microseconds. ITD works primarily below 1.5 kHz, where wavelengths are long enough that the phase relationship between ears stays unambiguous.
Interaural Level Difference (ILD): Your head casts an acoustic shadow for high frequencies. The far ear receives a quieter, filtered version of the signal. ILD dominates above 1.5 kHz, where wavelengths are short enough for your head to create meaningful attenuation.
The 1.5 kHz boundary has a physical basis. The maximum ITD the auditory system ever encounters is about 660 microseconds, corresponding to a period of roughly 1,515 Hz. Above that frequency, the maximum phase difference between ears spans more than one full cycle, making the direction ambiguous. The brain cannot tell whether the left ear is a quarter-cycle ahead or three-quarters behind. ITD stops being a usable cue, and the system shifts to ILD.
Coincident Pairs: Pure ILD
XY and Blumlein arrays place both capsules at the same point in space, or as close as physically possible with one centered directly above the other. Zero spacing means zero ITD. All spatial information comes from level differences only.
XY with cardioids points both mics outward, typically 90° apart (each mic 45° from the center axis). A source directly ahead gets identical responses from both mics and appears centered. As the source moves off-axis, one cardioid response rises and the other falls, creating an ILD.
At a 90° mutual angle, the stereo recording angle is roughly 196°. Sources quite far off-center still resolve somewhere within the stereo image rather than pinning hard left or right. For close-up sources that fills the field nicely. For an orchestra it can feel slightly vague, because nothing feels truly at the edge.
Here is the counterintuitive part: widening the mic angle to 120° or 135° actually narrows the stereo recording angle. When the mics are spread further apart, level differences build up faster as a source moves off-axis, so instruments reach the edges of the panorama sooner. A tighter image, despite a wider angle between the mics.
Blumlein pair: Substitute figure-8 mics for the cardioids and the geometry changes entirely. Figure-8 patterns have a null at 90° and maximum response at 0° and 180°. With two crossed at 90°, each mic's null points directly toward the other mic's axis. The math gives a stereo coverage angle of 68°, substantially tighter than the 196° of a 90° XY. Sources more than about 34° off-center move quickly toward the edges of the stereo field, which is why Blumlein recordings have such specific imaging. Instruments feel placed rather than distributed.
The trade-off is rear pickup. Figure-8 mics capture both front and back, so the room behind the ensemble is always present. In a good hall that adds a convincing sense of depth. In a live venue with noise behind the array, it adds problems.
Alan Blumlein described this configuration in UK patent 394,325, filed December 1931, the same patent that introduced stereo recording as a concept.
Near-Coincident Arrays: Both Cues at Once
Human ears are separated by about 17 cm and angled outward. The ORTF technique, developed at the Office de Radiodiffusion Télévision Française around 1960, deliberately mirrors this geometry: two cardioids, 17 cm apart, angled 110° outward from each other.
The 17 cm spacing is small enough that comb filtering when summed to mono stays above a usable frequency range, but large enough to introduce meaningful ITD below 1.5 kHz. High-frequency localization comes from ILD; low-frequency localization comes from ITD. This matches exactly how human hearing works, which is why ORTF recordings often feel dimensionally convincing in a way that pure coincident pairs don't quite achieve. Not just wide, but spatially accurate.
ORTF is the most common choice for classical and orchestral recording for exactly this reason: it recreates the cue balance the auditory system expects.
The Nederlandse Omroep Stichting (NOS), the Dutch national broadcaster, developed a related near-coincident configuration: 30 cm spacing at 90°, with a theoretical stereo recording angle of 81°. The wider spacing delivers more spaciousness and captures more of the acoustic environment, at the cost of greater phase sensitivity when folded to mono. NOS suits applications where spatial immersion matters more than pristine mono compatibility.
Spaced Pairs: Pure ITD, With Consequences
Wide spaced pairs (AB technique), typically two omnis a meter or more apart, capture primarily ITD. Omni mics produce no level differences between themselves, so localization at all frequencies depends on timing alone. This produces very wide, immersive stereo with generous room sound.
The problem is mono compatibility. When two signals with significant time offset are summed, frequencies whose half-wavelength equals the capsule spacing are completely cancelled. A 1-meter AB pair removes 170 Hz and all its odd-multiple frequencies when collapsed to mono. This is why spaced pairs appear regularly in music production but rarely in broadcast, where mono remains a real output.
Choose the Mechanism, Not the Habit
Understanding which cue each technique exploits gives you a principled basis for decisions:
- Close source, controlled room: XY cardioid. Zero phase issues, clean ILD.
- Precise instrument placement needed: Blumlein. The 68° capture angle positions sources crisply within a tight field.
- Natural orchestral or choral sound: ORTF. Its geometry matches human hearing and produces spatially convincing results.
- Maximum spaciousness, stereo-only context: AB spaced omnis.
- Stereo width as a post-recording variable: MS. The mid cardioid guarantees mono compatibility; the side figure-8 captures spatial information with adjustable gain.
A Blumlein pair on a piano in a good hall sounds exceptional because the tight 68° field maps the instrument precisely and the rear pickup captures the room authentically. The same technique on a solo vocalist outdoors disappoints: the 68° capture angle pushes the singer toward the stereo edge before much movement has occurred, and the rear lobe picks up nothing useful. The technique isn't wrong. The physics just don't fit the situation.
That mismatch, and knowing in advance when it will happen, is what separates deliberate recording choices from lucky ones.