John Darko
When it comes to swallowing bitter pills in hi-fi, few leave more of a sour taste than learning that the room is the largest contributor to a loudspeaker system’s sound quality and that getting the room to sound good – so that it doesn’t mess too much with the sound of our loudspeakers – can be a struggle. Every room has two behavioural zones: resonances in the bass and reverb in the mids and highs. A room’s acoustic behaviour transitions from resonance to reverberation at the Schröder frequency, which, depending on the room’s three dimensions, can sit anywhere between 100Hz to 300Hz.
Last week, I detailed the midrange and treble reverberation issues found in each of my three listening rooms and how, for each one, I tamed that reverb with acoustic panels fitted to the walls and the ceiling. Such explanations don’t come without a degree of audience blowback. The most common complaint relates not to price but to my public image: that a reviewer working in a room that sounds better than the majority of his audience renders him less of a man of the people. “Look at him in his Ivory Tower!”.
I can certainly feel the frustration buried within this kind of response. Kitting out a listening room with acoustic panels is a luxury. For those sharing their listening space with others, it might be an insurmountable challenge. Getting a full acoustic fit-out past the aesthetics committee is no small undertaking. For anyone renting a home outside of mainland Europe, it might be contractually impossible.
And yet we cannot deny that a reverberant room colours the sound arriving at our ears from a pair of loudspeakers. Some sound will arrive directly but the reverb will show up late and distorted, having bounced around the room thousands of times. This late, distorted sound can lower the resolution of what we hear and in a reviewer’s room, hearing less from a hi-fi system can negatively impact the amount (or type) of insight offered to readers or viewers. This is especially true when auditioning upstream components like amplifiers, DACs and streamers where smaller deltas are present.
A reviewer’s decision to treat his/her listening room strikes at the heart of a dichotomy that sits between a) offering proper insight into how something sounds and b) maintaining a relatable public image. Decisions, decisions. Having treated three rooms in as many years, I hope it’s obvious that I’ve chosen to prioritise the former.
Why?
I know that much of this publication’s audience depends on the findings generated by side-by-side comparisons. I think it’s safe to assume that those same people want those findings to be as reliable as possible. An acoustically-treated room helps enormously with reliability. I find it much easier to audibly split the differences between hi-fi components when working in an acoustically treated space. Any differences now jump off the page where previously I’d have to read it several times over. Could the Ivory Tower be a lighthouse?
The other, more substantial argument in favour of dialling down a room’s midrange and treble reverb is more vanilla but no less important. Listening to music in an acoustically treated room is more enjoyable than listening to music that has been left as is. A better-sounding room ups the fun factor!
Now let’s pull a 180.
Acceding to the idea that a reviewer’s room should present with a similar acoustic to that of his/her audience members so that the reviewer hears what his audience hears, we have to ask: where should that room’s problems reside? Getting specific with an answer ignores that all rooms are different — as different as the people who sit in them.
Perhaps we might consider building the picture of an average room but what does the average listening room look and sound like? Clothing companies have bell curves for arm lengths, inseams and waistlines but do hi-fi manufacturers maintain a similar database of room dimensions? Doubtful. Besides, when measuring our own to contribute to the data pool, do we factor in the adjacent dining area or hallway? Let us also not forget ceiling height where some are sloped or vaulted. Where would they sit on any normal distribution curve?
If all rooms are as different as people, the probability that one of the world’s handful of full-time hi-fi reviewers is working in a room that sounds similar to yours is vanishingly low.
My coverage of room acoustics has thus far focussed on the reverberant behaviour of a room. This takes place above the Schröder frequency but I’ve said very little about the bass resonance issues that we face below Dr. Manfred Schröder’s Plimsoll line. A video on this topic is already on the slate for 2024 but (with brevity in mind) a room’s bass issues are a direct function of its dimensions.
For example, a 4m x 3m room with a 2.5m ceiling height will see its axial modes sit at 57Hz, 43Hz and 69Hz. Why? Because 57Hz has a wavelength of 6m, twice that of our room’s width; 43Hz has an 8m wavelength, twice that of our room’s length; and 69Hz has a 5m wavelength, twice that of room’s height. Technically speaking, a room’s low end should be treated first.
Back with the reverb that sits above the Schröder frequency, we’ve seen how the RT60 of a good-sounding listening room should remain consistent and sit somewhere between 0.3 seconds and 0.6 seconds through 300Hz to 4kHz. What should the RT60 graph of a more relatable room look like? Should it blow out to over one second or should it dance between 0.7 seconds and 0.9 seconds?
And what of our relatable room’s behaviour below the Schröder frequency where the battleground transitions from mid/high reverb to bass resonance? At what frequencies should the bass peaks and nulls reside and at what magnitudes?
Above is the frequency response of a pair of Piega 701 Wireless Gen 2 sat not in my lounge room or office but in the guest room. This measurement was taken with REW and a Umik-1 microphone placed at the listening position. Despite the room measuring roughly 3m x 4m, there are no significant peaks or nulls below the Schröder frequency, which makes light work for Piega’s room correction smarts. It’s in the low end – below the room’s Schröder frequency – that DSP’s power can be felt. More on that in a moment.
If there’s one thing I’ve learnt about room acoustics these past few years it’s that a frequency response graph can hide a room’s true character. Even though the frequency response of the 701 Wireless Gen 2 in my guest room is far from a car crash, a large portion of the Darko.Audio YouTube audience would complain about the sound of my voice were I to record a video in this room — because its reverberant behaviour (above the Schröder frequency) clocks in at between 0.7 seconds and 0.9 seconds. Yikes. On the upside: look ma, I’m a man of the people once more:
We filmed most of the b-roll for our upcoming Piega 701 Wireless Gen 2 review video in this room because I first wanted to test Piega’s room correction software in what many would refer to as a ‘real world’ space. Per our preview post, these are active speakers whose streaming and source routing are handled wirelessly by an outboard box.
Piega’s plastic Connect Plus box plays catch on cloud and LAN streams (and hard-wired signals) before handing them off wirelessly to the left and right speakers. If this rings a Buchardt bell, you’re in the right neighbourhood. Buchardt’s Platin hub uses the WISA protocol to stream to its loudspeakers but Piega’s Connect Plus box performs the same feat with the Kleernet protocol. Functionally speaking they are more-or-less identical.
Piega’s room correction component works similarly to Buchardt’s: the Connect Plus plays white/pink noise through the speakers whilst we move an iPhone running Piega’s app around the room. The iPhone’s microphone readings are processed by the app to acoustically map the room and generate a correction curve. That correction curve adjusts each speaker’s frequency response to better match the room. It’s why I think ‘compensation’ is a truer descriptor of this process than ‘correction’.
In my untreated room, Piega’s room correction didn’t bring forth a night-and-day improvement, only a small uptick in midrange intelligibility: an uplift borne out by the Piega loudspeakers’ in-room frequency response curve:
Piega’s room correction software is curtained at 500Hz. Per the frequency response graph, it changes nothing above that point. This doesn’t bode well for any reverb concerns. Never mind that reverb is a time-domain distortion that’s baked into the room’s dimensions where some frequencies arrive at the ear later than others. Lifting the curtain wouldn’t change how compensating for reverb and time domain errors in the mids and highs from within a stereo hi-fi system’s DSP requires more loudspeakers and more powerful software.
Here’s the rub: more loudspeakers not only sit beyond the scope of this website but they might cost as much as a full acoustic fit-out, which would also benefit the sound of the room when no music is playing. As anyone who has sat in a busy restaurant can attest, it’s easier to make out what someone is saying across the table when the room itself isn’t creating a delayed/distorted facsimile of every diner’s voice.
Back in my guest room, I measured the RT60 with Piega’s room correction turned on and off. There is little point in labelling which one is which when not even a hair’s width separates ’em.
Just as we might expect from a problem generated by wall, ceiling and wall reflections, Piega’s room correction software does nothing to improve the room’s reverb time. With RC ON, the RT60 still dances between 0.7 seconds and 0.9 seconds. This isn’t Piega’s failing. The blame lies squarely with the room itself whose RT60 isn’t quite as bad as my office and lounge were before they were treated (over a second!), but in my experience, anything above 0.6 seconds is instantly noticeable.
Room reverb in the mids and highs cannot be fixed in software. This leaves wall and ceiling panels as the only option for someone wanting predictable and reliable results when trying to reduce a room’s RT60.
Before I shoot the Piega review video’s talking head component, I will take a few days to compare the 701 Wireless Gen 2 to the Buchardt A700. I have time to conduct this side-by-side in one room only. Should I compare the pair in an untreated guest room or the acoustically treated lounge room? I’ll answer that question with a question: what good is a reviewer’s listening space whose acoustic make-up sits closer to the majority of a reviewer’s audience when that acoustic make-up could mask (some of) the differences between two sets of loudspeakers?
Further information: Piega
