High end high quality int. amp for low level listening


Hello to all Audigon members.  I'm quite in a dilemma weather I should upgrade my amplifier. Currently I own Pathos classic one MK3 driving Sonus Faber Sonetto's speakers, and I must say I'm very happy with sound filling my 35 square meters room. However, it's known that the speakers are power hungry as they rated at 86db sensitivity and 4 ohm impedance and I think they will surely benefit from a bigger power supply. With all that being said I'm not sure if I will hear any improvement mostly because 90% of the time I listen at ~60dBs SPL. My budget is around 5k $ and these are the amplifiers I've been considering:  Hegel H390, Anthem STR, Cambridge Audio Edge A, McIntosh MA5300/MA252, Accuphase e280, Rotel Michi x3 or used Pass Labs INT 25, Mark Levinson 5805.

What do you think guys, will any of the amplifiers make ay difference at 60dB SPL ? 
celestial__sound
@holmzI think ditusa above answered this better than I can. I’m not an electrical engineer, but it has to do I think with the quality of the first watt. Yes 60 decibels is the same volume in two different setups, but the amount of distortion and  resolution heard at that level can be vastly different.  I think you want dynamic headroom listening even at 60 dB. 
IMHO, keep the Pathos and get high efficiency speakers or one with a benign impedance curve (e.g., Fritz Carrera Be, ProAc Response D2, used Joseph Audio Pulsar). I think high efficiency speakers sound better at low volumes than low to medium efficiency speakers, but just my opinion — based on tube Coincident Dynamo 8 wpc amp driving Klipsch Heresy IIIs (99 dB/1 m) versus 125 wpc solid state Belles Aria Signature integrated driving Vandersteen VRL CT (86 dB/1 m). 
Maybe you’re right. But there would have to be a reason why the sounded better.

It could also be the Fletcher Munson is getting somewhat adjusted with speaker that has more low and high end than the other set?
It would be interesting to see REW curves for those two sets @dcevans That assumes that both amps don’t have a loudness correction feature.

The speakers are linear, so it is not like the more sensitive set comes off the blocks faster, other wise they would be non linear at low volume levels… and also they would then be non-linear with small changes in signal level. One would have to make them sticky or something like that, as the motor force is a linear product of current and magnet flux strength. The main non-linear part with a speaker, I think, is compression.
’’Maybe you’re right. But there would have to be a reason why the sounded better.’’

Because efficiency and sensitivity are not the same. See: tech note
below page 1 section C:

/www.lansingheritage.org/html/jbl/reference/notes/tech1-3a.htm