Reviewer Auditioning and Reviewing at 50 - 55db max

sergeit -- Thu, 09/10/2009 - 22:53

I'm curious what others think. A colleague of mine recently observed a professional reviewer from a popular rag audition a product entirely at 50 - 55db and no higher, including live recordings.

I have my own thoughts but do others have any comments about such a practice?

Robert Harley -- Fri, 09/11/2009 - 16:36

I don't think that I could adequately evaluate a product without listening at realistic (live) levels.

Tom Martin -- Fri, 09/25/2009 - 15:29

If the noise floor in a quiet room is 40db, then this would give you 15db of dynamic range. In addition, very few users will use equipment this way. Both of these are huge drawbacks for reviewing.

CEO and Editorial Director, Nextscreen LLC

Jonathan Valin -- Sat, 09/26/2009 - 08:55

Although they do play a part, higher SPLs all by their lonesome won't give you "more dynamic range." The dynamic range on a recording is determined by the program, the instrumentation, the engineering, the mastering, and the medium it was recorded on and to, as well as by the playback level. IMO, playback levels should be set to give you the most lifelike dynamic scale for a given piece of music--which is to say that instruments that typically play more softly or instruments playing pianissimo or pianississimo passages should be as clear, articulate, and natural in timbre, volume, and dynamic as instruments that play more loudly or instruments playing fortissimo or fortississimo passages. This average playback level will vary widely with program and recording, though it will never be as low as 50-55dB average SPLs (unless, of course, someone is playing music purely for "background listening" while he or she is doing something else).

BTW, I think many listeners tend to "turn up the juice" automatically to make the softer instruments and softer passages more audible and the louder ones more exciting and impactful (or simply to give themselves the impression that they're getting more clarity and dynamic range, particularly in the bass, out of murky, compressed, no-dynamic-range rock recordings). The truth--at least on much classical and acoustic music--is that far from increasing dynamic range they're actually reducing it by turning notes or passages that were meant to be sounded as pianissimos into mezzofortes. They're skewing dynamic scale to the f-to-ffff side of the spectrum--and thereby killing the piece and the performance. Playing back Joan Baez singing "Gospel Ship" at 100dB+ average SPLs would be insane, IMO; OTOH, playing back The Stones' "Gimme Me Shelter" at 100dB+ average SPLs (or louder) would be appropriate.

Vade Forrester (not verified) -- Wed, 09/30/2009 - 11:29

Not many listeners listen to their systems at such low levels, so the reviewer that does so isn't evaluating the equipment under the conditions it's likely to be used. Louder listening levels may stress the equipment to reveal flaws that you won't hear at 50-55 dB.

Robert Harley -- Wed, 09/30/2009 - 13:32

Vade Forrester is absolutely correct; many aspects of equipment performance are revealed only at higher listening levels.

Norman Varney -- Sat, 10/09/2010 - 23:45

Though we don't know what music the reviewer was listening to, we also don't know how the dB level was measured. The microphone should be at ear height, at the listening position, facing the ceiling, and "C" weighted, slow. Measuring levels at 50-55dB might have been measured "A" weighted, who knows, and the difference is about -30 dB at 50 Hz.

Measuring your average dB level is not as easy as it sounds for the layman. I have fancy test equipment that can average over a long period of time, etc. But equiped with the Radio Shack SPL meter, the answer is more difficult. First, you have to select music that remains at a somewhat constant level. I found even typical rock & roll tunes easily swing 15dB "C" slow. Once I set my volume for "Rock & Roll" I found the meter averaging around 90-95dB SPL "C", slow. Is this repeatable? Hardly. A different recording on a different day could easily be off by a few dB.

Yes, the playback level will vary with the music type in order to make it seem real. That said, I believe I tend to play solo acoustic guitar a tad louder than real-world, and I definitely play live rock & roll music below real-world levels.

I also agree that listening to music at too low a level will not only bury the expressiveness of the music itself, but will limit the performance potential of the equipment under review. Low frequencies, low-level resolution, dynamic responses, even emotional impact would be limited even if the reviewer had an ultra quiet room of say an NC 5 rating, and especially so if he has a typical NC 25-30.

Why is SPL so important to the reviewer? I have been to many reviewers homes to help them get the most out of their systems so that they can offer more honest reviews. Regarding performance, first is set-up, then comes calibration. For the reviewer (or consumer comparing equipment), the outputs of the devices under test should match each other within 1dB SPL across as wide a bandwidth as possible in order to be fair. If not, the louder device will likely be favored.

Norman Varney
A/V RoomService, Ltd.