Does Flod Toole think ears trump measurements? In his video he states a concert hall that measures horribly(comb filtering etc.( can be the site of an extraordinary performance .I'm paraphrasing.
Let me make an important point here and then answer your question: there is no such thing as "measurements" as a single concept. Each measurement has a different value to the point of the question being asked. The notion that we should discuss measurements versus the ear is completely wrong. The ear is the ear whereas measurement is a generic term that can refer to anything from weight of a DAC to a psychoacoustic model inside of a codec attempting to minimize distortion. The latter highly correlates with what we hear, the former not much of any.
I mention this because you cannot generalize common measurements in a room to measurements elsewhere. Each has their own potential benefits and flaws.
With that disclaimer out of the way, in acoustics there is an important concept of transition frequencies of a few hundred hertz. Below those frequencies, room measurements absolutely speak the truth. If you see a 15 dB peak at 70 Hz and a note hits it, everyone here and their dog will hear that note way amplified. Coloration is added and is specific to that room. In another room, the same measurement may show a 5 db trough at 60 Hz, and hence guaranteeing that the two rooms have changed the response of the same loudspeaker. No ifs and buts about it.
Identifying the above by ear alone can be hard. The room modes at low frequencies can have resolutions below a single Hertz! This is why you need ultra precise EQ system to dial these out or lots and lots of room treatment. I don't know of anyone, no matter how experienced, that can walk into a room and say, "oh, there is a peak at 72 Hz." You can try to guess and screw around with an EQ by ear but I assure you that it will sound good on some tracks and not others. So the value of measurements here is clear and accepted.
Above transition frequencies we run into a problem when we use a single microphone. We have two ears, not one. When a sound comes from the side, our head creates a shadow in these higher frequencies and hence what the right and left ear hear are not the same. See my article here and this graph:
http://www.madronadigital.com/Library/RoomReflections.html
Notice the big difference in hearing the same sound 30 degrees to one side. So no way a single microphone can pick up what is clearly two different signals.
The other problem is that the resolution of our ear keeps reducing as frequencies go up. Again a graph from my article:
This means that if you make a frequency sweep as is commonly done and show that frequency response, what you see in there is wrong. There will be tons of variations that fall below the resolution of the ear. In this sense, the measurement is way, way too good relative to how deaf the ear is in picking up small changes. Again see the article for details on that. This can be easily fixed by filtering the response gradually as frequencies go up or doing that manually as I do (REW has added the former now but I am not quite sure of its appropriateness).
It is for this reason that we say "measurements" can be totally useless in acoustics. We are not saying all measurements are bad. We are referring to common measurements people make who don't understand the psychoacoustics.
And the last point is key: if you understand the science of how we hear, then you cannot be easily misled by measurements. Now this is not an easy science to learn but it is learnable. Certainly if you spend an hour or two a week studying it, eventually you come out the other side, fully capable of understanding at some point. Once there, there is no danger in using measurements.
