Dear Reader,

A little bit of fun to start and then wham! another cerebral note from Mark Block to get us thinking. In this issue

• Spectroid: A Truth Serum for Audiophiles
• Beyond the Baffle Part 2: Mics Mess With Our Minds
  ​By Mark Block

Cheers, Merrill Wettasinghe/Chief Music Officer 𝄞𝄞

Spectroid: A Truth Serum for Audiophiles

Merrill Wettasinghe

Is it the room? Is it the speakers? Is it the preamp?

Questions worthy of Shakespeare. Questions that have cost audiophiles countless hours of sleep, a fair amount of money, and, in some cases, a noticeable amount of hair.

The quest is always the same: How do I achieve a flat frequency response?

First things first: measure it.

This isn't a new discovery, but it is a gem. Spectroid is a real-time audio spectrum analyzer that has become one of my favorite tools. It offers a superb full-screen display of live frequency content while simultaneously capturing frequency history, allowing you to visualize exactly what is happening over time.

Play any track and you'll quickly see the overall shape of its frequency response. Yes, it's another toy—but it's an exceptionally useful one. More importantly, it's a toy that doesn't lie.

Developer Carl Reinke has done an outstanding job. Spectroid provides significantly higher frequency resolution than many analyzers I've tested. It also includes a number of useful audio capture settings, allowing for extremely high-resolution measurements—though for most of us, that level of detail may venture into the realm of glorious overkill.

This isn't merely a tool for impressing your non-audiophile friends (or particularly patient dinner dates). It can also serve as a valuable reference when evaluating systems. Use your favorite recordings as a baseline and compare how different systems reproduce them across the frequency spectrum.

Test Setup

The images below show a 20 Hz–20 kHz sweep using the test bench speakers.

Phone: Samsung Galaxy S23 Ultra

Speaker: Cheap $100 2 way speaker used for burn up testing.

Sample Measurements taken:

The Bad News

For Apple users, Spectroid is currently available only on Android.

The More Complicated News

For Android users, the Google Play Store version is often restricted to older devices.

The Good News

Drum roll, please...

There is an alternative download that works perfectly on current Android devices. I'm running the latest Android OS, and Spectroid runs flawlessly. As of this writing, the most recent update was released on February 20, 2026.

Download it here:

​https://spectroid.en.softonic.com/android​

And if you'd like to join discussions with fellow users, visit the Spectroid community group:

​https://groups.google.com/g/spectroid​

Measure first. Debate later.

Enjoy!

Beyond the Baffle Part 3: Beyond Blumlein

By 🍄Mark Block 🍄

In Part 1, we discussed flanging and its ability to fool our pea brains into hearing sounds swirling around us with just two speakers up front. In Part 2, we discussed how “the absolute sound of acoustic instruments in real space” can be affected by microphone technique, and explored whether it’s possible to capture beyond-the-baffle sound in purely acoustic recordings without post-production studio trickery. We got deep into Alan Blumlein’s brilliant coincident figure-8 technique. We determined that when a Blumlein recording is played back with the speakers spaced at roughly the same angle as the microphones—much wider than is typical in many listening rooms—the solidity of the stereo illusion can be remarkable. But that solidity comes at a price: with a few notable exceptions, the stage generally does not extend beyond the speakers. For better or worse, Blumlein gives you solidity more than lateral spillover. You can hear this image specificity in audiophile recordings by Kavi Alexander for Water Lily or Bob Katz for Chesky.

Personally, I love the illusion of the stereo soundstage extending beyond the loudspeakers. YMMV. To do that with microphones alone, some decorrelated or out-of-phase information has to be baked into the recording—but it has to be the right kind of phasiness. Spaced-omni setups, such as the classic Mercury and RCA approaches, are prone to a different kind of phasiness. They can produce grandeur, depth, bass weight, and hall ambience, but they usually do not combine Blumlein-like focus with beyond-the-baffle magic.
​
Think about it. If two or three omni microphones are spaced several feet apart, the same sounds from the performers reach the microphones at significantly different times. Those time differences can create spaciousness, but they also bake comb-filtered and contradictory localization cues into the recording. With widely spaced microphones, the stereo system reproduces timing differences far larger than the ones our heads normally generate, given that our ears are only six or seven inches apart. The listener then hears the normal interaural timing and level cues created by the loudspeakers in the room, plus the much larger timing discrepancies captured by the microphones. The result can be exciting, spacious, and beautiful, but it is not lifelike or three-dimensional in the ways I’m chasing here.

To add lateral spaciousness without completely sacrificing the image specificity of coincident mics, engineers had to get a little tricky. Enter ORTF.

Developed in the 1960s by the French broadcasting network Office de Radiodiffusion Télévision Française, ORTF abandoned figure-8 microphones in favor of two directional microphones: cardioids with a heart-shaped pickup pattern, spaced 17 centimeters apart (about 6.7 inches). You can probably grok that this is not a random number. It’s roughly comparable to the spacing between human ears, though ORTF is not binaural recording and should not be confused with a dummy-head technique. The mics are angled outward at 110 degrees, roughly mimicking the way our ears receive sound from slightly different directions. Because the capsules are no longer occupying the same point in space, ORTF puts back into the recording what Blumlein eliminated: interchannel time-of-arrival differences, which mimic one way our ears localize sound.

Example: When a bassist on the right of the stage saws away at their instrument, the sound hits the right microphone first, then travels roughly 17 centimeters farther before reaching the left microphone. The secret sauce of ORTF is the small acoustic time delay in addition to the level difference. Because the 17 cm spacing introduces time-of-arrival differences in the range our brains interpret as spatial information, the illusion of space expands outward. The strict left/right boundaries of the loudspeakers soften, and the hall’s natural reverberation will sound wider, even in a 60º equilateral setup.

I must note that because ORTF was a French invention, the Dutch thought they could do it better. Of course they did. It’s called NOS (Nederlandse Omroep Stichting). In the Dutch system, the mic spacing is wider: 30 cm apart (11.8 inches), so the interchannel timing difference is greater than with ORTF. To prevent a hole in the middle, the angle is narrowed to 90 degrees. Anatomically, the Dutch math is a little suspect—no one’s ears are a foot apart unless they’ve been taking steroids for years. (I’m looking at you, Barry Bonds.) But NOS was not designed as a dummy head. The engineers arrived at it subjectively through careful trial-and-error testing. The resulting stereo image is wider and more spacious, although the focus is somewhat less defined.

So is ORTF the Holy Grail of acoustic recording techniques? Maybe not. Taking things in yet another omnidirectional direction, the Jecklin Disk, or OSS (Optimum Stereo Signal), uses roughly ORTF-like ear spacing—about 16.5 to 17 cm—but places an acoustic baffle between the omni mics to recreate a "head shadow" like a human skull. (P.S. The basic baffled-microphone idea goes back to Blumlein’s 1931 patent; Jürg Jecklin didn’t introduce his refined OSS version until decades later.) Jecklin Disk recordings offer a taste of the spatial freakiness we hear with binaural audio while still translating acceptably to a standard 60-degree loudspeaker triangle.

But if the original Jecklin Disk started with ear-like spacing, didn’t NOS suggest that a little extra spacing might produce a bigger, more spacious image? My friend Barry Diament of Soundkeeper Recordings did some trial-and-error experimentation with a Jecklin setup and reached a similar conclusion: a little wider is a little better:

“My own experiments told me I was getting a better sense of the space with a 15" spacing. I looked into the commercially available Jecklin Disks sold to recordists and determined that, in addition to having the microphone mounts locked in at ear spacing, these had surfaces I deemed too reflective.... This led me to build my own variation on the theme, a 12" circle made of soft absorbent material.... I heard the sonic images solidify in a way I'd not heard before,... [appearing] to emanate from just where [they were] when I made the recording…. I have since heard it said Mr. Jecklin now uses a mic spacing of 15–16 inches.”

Listening to Barry’s recordings, you’d be forgiven for thinking that we’ve found our Holy Grail of mic setups, but I’m looking for the ultimate beyond-the-baffle experience, so let’s consider a few other possibilities. There’s one other mic technique that might be better than Blumlein for achieving extra spaciousness. It’s called Mid-Side, or M-S, and it was devised by… Blumlein. Jeez, what didn’t that guy think of?

In a single 1931 patent, Blumlein laid out an astonishing amount of modern stereophonic theory: how our hearing uses timing and phase cues at low frequencies and intensity cues at higher frequencies; how two loudspeakers can create convincing stereo images; how two channels could be cut into a single record groove using the 45/45 system; and how directional microphones could be used to capture a stereo soundfield, including both the crossed figure-8 Blumlein pair and the sum-and-difference logic behind M-S recording. It is not much of an exaggeration to call the patent one of the foundational documents of stereo sound.

Like the Blumlein pair and X-Y, Mid-Side (M-S) is a coincident technique: the capsules are placed as close together as possible, so there are essentially no time-delay differences between the channels. But instead of pointing two figure-8 microphones diagonally left and right, M-S points one microphone straight ahead at the orchestra—the Mid mic—and places a figure-8 microphone sideways, with its null facing the orchestra—the Side mic. The Side mic mostly captures lateral information: hall reflections, width, and off-axis energy.

The stereo signal is created with a “sum and difference” matrix. The left channel is Mid plus Side; the right channel is Mid minus Side. In other words, the Side signal is added to one speaker and polarity-reversed in the other. If the Mid and Side signals are set to equal level, and the Mid mic is also a figure-8, the result is mathematically equivalent to a Blumlein pair. So why mention it?

Because with M-S, the width is adjustable after the recording is finished. If the engineer raises the Side fader, the stereo image gets wider. It is somewhat like changing the effective pickup angle after the fact. More importantly, it sends more opposite-polarity acoustic information to the left and right speakers. As we saw with flanging in Part 1, some negatively correlated “anti-phase” sounds—especially diffuse or noisy sounds—are hard for our brains to localize precisely. If the goal is a bigger, wider wash of hall ambience, the Side fader can provide it.

When I was doing film and video editing, we had an inside joke about mistakes no one wanted to deal with: “We’ll fix it in the mix!” The stereo soundstage of an M-S recording cannot be fixed exactly, but it can be tweaked in the mix. That’s one of its great practical advantages.

M-S had another benefit that mattered more in the mono era. If you sum the left and right channels to mono, the Side mic cancels itself out completely, leaving only the Mid mic. Technically, the Mid signal is reinforced, but that is just a predictable level change. With many other stereo techniques, mono summing is less tidy: center information can become disproportionately loud, while out-of-phase or time-delayed stereo information may cancel, smear, or comb-filter. For radio, TV, and mono playback systems, M-S offered a very elegant trick: the same recording could produce a spacious stereo version and a stable mono version.

So is M-S the be-all-and-end-all of wider-soundstage recording methods? Not quite. As the Side fader comes up, the center image can start to thin out. The bass may feel weaker, the orchestra may lose solidity, and the violins might begin to sound as if they are coming from somewhere behind your head. That might be cool on a Jimi Hendrix record, but it is probably not what you want when you are trying to enjoy a Beethoven symphony. M-S is powerful, but it is a balancing act—and sometimes the safest move is not to play with the width knob too much.

So where does that leave us? Blumlein gives us solidity. ORTF and NOS add useful timing cues. Jecklin-style baffled omnis add head-shadow trickery. M-S lets the engineer adjust width after the fact. All of these techniques can be wonderful, but none of them fully answers Keith Johnson’s “celestial sound” challenge: How do you capture the three-dimensional space around the musicians and still make it work through two speakers?

That brings us to the Soundfield microphone, Ambisonics, and one of the most “celestial” recordings ever made—in a church, of course. That’s coming up next time.

Links: https://www.barrydiamentaudio.com/recording1.htm​

​
​

Send us your comments, pictures of your systems. ​
​NewsletterEditor@MerrillAudio.net​
With your permission I will print your comments in a future edition.