That is a personal preference. I DO want flat in-room response. It makes the recordings sound more natural and a G and an E are reproduced as the artist played them, not one louder due to the speaker or EQ flaws. I do not hear brightness from flat response.
Oh, I think it's something like 99th percentile (AES papers in the 80's and mid-2000's) find the sloped in-room loudspeaker
power response (all the speaker energy from all directions, summed)
highly preferable to any other curve, including "flat power response", which the vast majority find highly irritating and EXTREMELY thin and shrill.
While REW and other windowed (synchronized FFT) measurement systems can isolate the axial response--what most people inadvertently are referring to-- our hearing is rather an integrating machine that perceives the
power response. A basic RTA with a diffuse-field mic also (more or less) perceives the
power response.
However, a simple RTA is highly subject to comb-filtering from off-axis enegy and difficult (but not imposisble) to provide an accurate power respoonse measurement.
That said, a speaker that has a flat axial response (as measured above, or in an anechoic environment) and a
smooth (but not flat) off axis response--as measured on a turntable or in a typical enclosed space-- will have an excellent down-slope fitted
power response and be perceived as excellent by the vast majority of the population with normal hearing.
