[koldby]

  Arguably, a very small single driver making all the sound (headphones) and without a room to colour the sound is more accurate by default then a speaker with many drivers, a XO,, the crossover components, the room, the off axis response, ect.
 
However, the sound pressure that speakers do in a room, the soundstage being put in front of you with a real phantom center, feeling the bass with your body, ect is just two different experience and cannot be replicated with software + headphones.

Well the first phrase holds true for only some loudspeakers, but not all!
 
 

 
Besides that I agree.
From my point of view, it depends very much on what kind of music you are listening to.
Classical music recorded in a concert hall, with only a couple of microphones, does not do it for me on headphones as some of the experience in listening is the sense of having the orchestra in front of you and recreating the acoustic of the concert hall. One can argue, that only a multi channel setup can do that.I have heard binaural recordings over headphones that comes some of the way, but for orchestral works only loudspeakers work for me. Quite another story is music made in a studio. Here the experience from listening through headphones can be very exciting althoug quite different from loudspeakers.
 

 

[ToroFiestaSol]

In response to OP:
 
Assuming that one has a perfect room (like 150k spend on a custom room, and specially tuned for the speakers)
 
It depends on which headphones and which speakers, for example:
 
Real man Stax setup (007 MK1 or modded MK2 & Kevin Gilmore amp)...something like a YG Sonja or Rockport Altair 2 , with a really nice amp (think Acupphase class A or Gryphon with matching preamp) sounds at the same level of purity, tone, timbre and detail. Of course, soundstage, bass and imaging is another world (main advantages of loudspeaker design).
 
Don't forget the "perfect room", without it headphones will ALWAYS have more detail and better tonality than speakers (no room interferences with headphones).

 

[dguitarnut]

My main caveat about headphones is that I tend to turn the volume up much louder and this has done a number on increasing my tinnitus. This was the reason I returned back to speakers. By the time I crank them up I am being yelled at by the wife and neighbors.
I do prefer speakers because of the stage imaging. My speaker system is as accurate as my Hp system but I invested twice as much to achieve it.
Larry

 

[mulveling]

My main caveat about headphones is that I tend to turn the volume up much louder and this has done a number on increasing my tinnitus. This was the reason I returned back to speakers. By the time I crank them up I am being yelled at by the wife and neighbors.
I do prefer speakers because of the stage imaging. My speaker system is as accurate as my Hp system but I invested twice as much to achieve it.
Larry

I absolutely had that problem with headphones being much worse about eliciting listening fatigue, and having the tendency to make you push the volume up to even approach satisfying dynamics and bass impact (a bad combination). Even with an "easy listener" dynamic like the HD650.
 
This advantage of speakers almost completely disappeared when I switched to Stax headphones. It fully disappeared once I upgraded my source & amp to better match the SR-009's quality. Of course, good speakers will always clobber headphones in staging and visceral impact.

 

[dguitarnut]

I absolutely had that problem with headphones being much worse about eliciting listening fatigue, and having the tendency to make you push the volume up to even approach satisfying dynamics and bass impact (a bad combination). Even with an "easy listener" dynamic like the HD650.

This advantage of speakers almost completely disappeared when I switched to Stax headphones. It fully disappeared once I upgraded my source & amp to better match the SR-009's quality. Of course, good speakers will always clobber headphones in staging and visceral impact.

How did the Stax hp help prevent you cranking them up to harmful levels? Are they not as loud or as bassy?

 

[mulveling]

How did the Stax hp help prevent you cranking them up to harmful levels? Are they not as loud or as bassy?

They have lower distortion (especially up top) and are less fatiguing at any given volume level. But also with lots more detail overall, I seem to hit a "satisfying" listening level at a safer point, and don't feel so much of a need to push it. 

 

[nvfan]

  Arguably, a very small single driver making all the sound (headphones) and without a room to colour the sound is more accurate by default then a speaker with many drivers, a XO,, the crossover components, the room, the off axis response, ect.
 
However, the sound pressure that speakers do in a room, the soundstage being put in front of you with a real phantom center, feeling the bass with your body, ect is just two different experience and cannot be replicated with software + headphones.

 
In regards to your comment on single drivers, by that logic aren't IEMs going the wrong way with multiple driver designs, and dynamic-balanced armature hybrids? I don't think anyone is arguing that that best dynamic IEMs are better than the best balanced armature IEMs?
 
Also I'd argue a well-made 2-way speaker with shallow crossover slopes (like the Dyanudio C1 Plantinum) is almost indistinguishable from a point source. And there are excellent coaxial designs that can mimic a point source, like the Pioneer/TAD Evolution One.
 
It's true speakers will never reach the resolution of headphones due to the room, but then again headphones will never reach the resolution of IEMs due to the crevasses and valleys that impede sound due the shape of the ears, the flip side of each type being increasing soundstage which is also a huge part of music. Everything in a way is tradeoffs, but speakers can chase the "live venue" sound better than any headphone or IEM.

 

[mulveling]

 
Also I'd argue a well-made 2-way speaker with shallow crossover slopes (like the Dyanudio C1 Plantinum) is almost indistinguishable from a point source. And there are excellent coaxial designs that can mimic a point source, like the Pioneer/TAD Evolution One.

Since I started in head-fi rather than 2ch, coherence (or lack thereof) of multi-driver systems was definitely an issue when getting into speakers. I found my happy place with the Tannoy dual-concentrics -- though the tonal balance of their various models tends to be a bit over the map, they get the coherence right, and the better sounding models (usually just a very big 12" or 15" 2-way coaxial) have it all. And you can seat yourself midfield/near-field without issue -- thus reducing room interaction, which their controlled (and symmetric) dispersion also helps with.

 

[jibzilla]

Maybe I missed it but I think it should be pointed out that the sourcing is usually quite different. At least from my perspective speakerland is filled with allot of audiophile cd's and direct to disc records where headphoneland is allot of flac and binaural. Not to say you can't listen to flac from speakers, just do not expect an awesome soundstage. Not to say you can't listen to records from your headphones, just expect panning to completely ruin some songs.