Regarding frequency response, sure but bear in mind on its own it is pretty meaningless
I don't know about meaningless, because frequency response is the best measure of tonal color. But I agree it's not the only thing. There are four parameters that affect audio reproduction:
Frequency response
Distortion
Noise
Time-based errors
Of course, there are subsets, such as hum and buzz and LP crackles under noise. I explain these four parameters more fully here:
The four parameters
Audiophoolery
I personally felt though off-axis response falls under the same basic FR measurement just at different angles.
Sure, but on- versus off-axis response affects imaging, which some people incorrectly believe is not possible to relate to measurements.
Does the simple FR tell you about cabinet resonance or driver breakup-issues in detail, or potential decay time variance over the FR?
Cabinet resonance falls under both frequency response and time-based errors. A resonance is accompanied by a peak in the response, and also an extended decay time. But driver breakup is definitely distortion.
It would help on that debate if you could provide two different speaker measurements including how distortion affected what the reviewers reported
The main problem here is "what the reviewers reported," which to my way of thinking is more or less a random result.
Seriously, I have no use for subjective impressions because they are not reliable and vary all over the map. What one reviewer loves another may hate. Even the same reviewer might love something one day and hate the same thing the next day if the listening was not done blind.
Distortion in loudspeakers is very complex. Heck, loudspeakers generally are very complex to assess for quality. But sticking with distortion, each driver has its own unique type of distortion that varies with frequency and volume level. Passive crossovers add their own distortion that's unrelated to what the drivers do. To properly measure
everything that affects loudspeaker quality is a huge task. Listening alone is inadequate because the speaker's distortion varies depending on frequencies present in the source. So it's not that measurements can't tell us everything needed. Just that seeing a complete set of measurements is very rare.
One reason why I am more suspect of distortion being the only reason is when you look at preamps (ignoring power amps as that really complicates the issue) and say compare a cheap implemented global negative feedback SS to a highly spec-implemented tube zero feedback design and both have very low distortion and flat FR.
The four parameters I listed above indeed tell everything needed about an amplifer circuit. Amplifiers are much simpler than loudspeakers! So for an amplifier all that's needed is frequency response, distortion, and noise. In this case I'll put ringing under frequency response. Regardless, if two amplifiers have a response flat to within 0.1 dB from 20 Hz to 20 KHz, and the sum of all distortion is at least 80 dB down, then both amps will be audibly transparent and thus sound the same when auditioned properly (level-matched and blind). Looking at the data for those two preamps I'd assume they sound the same. Of course, the specs may not apply to everyone's system. For example:
Arc Ref 5 said:
THD achieved into 60kohm (2nd-4th harmonics) re. 1000Hz
Here the distortion is spec'd into a 60k load, but what if your power amp's input is only 5k? And I don't understand the "2nd-4th" restriction. What is the total level of distortion? More important, what is the distortion at other frequencies such as 20 Hz and 100 Hz etc? Differences in specs that are
not reported could account for a different sound. I don't see IM distortion listed, and IMD is generally more audibly damaging than THD.
Often when I list my "four parameters," someone will say there's more to audio fidelity than that. But every time I ask what else there might be, I never get an answer. I am ready to accept that there's more to audio fidelity than these four parameters, as soon as I see credible evidence.
--Ethan