I was talking to a buddy about what makes for an accurate speaker, and he suggested that perhaps hyper-accurate EQing could compensate for the imperfections in a loudspeaker's frequency response. That got my wheels turning, and I responded that EQ (even theoretically instantaneous-tracking, perfect editing EQ) would be insufficient, and I tried to come up with a list of what else is necessary. Here is what I have so far:
1) transient response - crucial for realism. I have heard we humans can detect accelerations equivalent to 60-100kHz - sorry CDs...
2) inertness - for realistic decay and consistent frequency response (no slow rise due to resonances)
3) size and spacing - I will include time alignment here, for realistic imaging and scale
4) consistent frequency-spectral performance for the above - tight mids and sluggish bass sounds unrealistic. (I am not a big fan of ports for this reason)
5) radiation pattern optimization - taste- and application-specific. More important for people who share music with visitors, and perhaps a touchy subject for ESL owners.
6) sexiness - totally consumer-taste, but can be any combination of performance, aesthetics, or techno-geekery (I want a plasma tweeter because it is an awesome concept - whatever the performance. It is freaking massless - sorta, almost.)
7) almost forgot dynamics! - you gotta have your snare rimshot sound different from a light brush stroke!
what else makes good audio playback? I would think most of the above apply to everything since the performance left the performers' mouths and hands and feet, from mics to storage medium to amps and cables, etc. I am pretty new, so please educate a brotha!