How do you know what a good speaker is?

Yep you only know a good speaker for sure when you hear it and each speaker has to be setup properly (including a proper amp) to enable that. Speaker A might handily sound good and Speaker B not on Amp A but swap in amp B and the reverse is very possible.

Also it is true that if the speaker is an easy load to drive, chances are it will sound good off a wider variety of amps than otherwise.

A good example is KEF ls50. State of the art within its limits in many ways, but will not sound good if amp is not up to the task of driving them ie lots of clean not just power but also current. I’ve heard both cases in my house.

Compare to Fritz Carrera, designed to provide a very easy load to drive. . I heard these sound very good indeed off a 7 watt headphone amp recently and have no doubt they would excel with most any amp used.

Regardless of price, every speaker will be a compromise in one way or another. Basically it sounds perfect to the designer given the constraints (price, resources, materials, drivers, components,etc).

what it it comes down to fir me is can I live with those compromises and do I “feel” the music? If I get goosebumps listening to certain tracks then I know they are good for me.

@audiojan I have met a few speaker designers in my time, typically during the introduction of a new " ready to sell " set off speakers. This has occurred in retail outlets, hotel rooms, shows, my home, as well as others. The fact is, every designer I have met, never had the " my design is perfect " attitude. That " attitude " was handled by a hired marketing rep or sales agent ( I did that for 2 companies for a short time ). Not that my conversations with any of them were very long, but it was always as if what was going through the designers mind was I could have / should have done this. None of them were " happy " about the design, probably because their minds were thinking of the next improvement or upgrade. My take from experience. ( BTW, same for any designer I have met. The sales and marketing reps, they were the promoters. Enjoy ! MrD.

@mrdecibel I think we agree. :-) Hence the comment "given the constraints". I've found most designers I have spoke to are reasonably happy with their design (and yes, they are all thinking about what they could've done differently and how they are going to improve them).

The point I was trying to make that a "good speaker" is highly subjective and in reality doesn't exist. It's only "good enough"

n80 wrote:  I don't see how measuring a speaker against what a live guitar or live piano sounds like is practical. Even in a live situation a mic'd acoustic guitar will sound different from an un-mic'd guitar.

Agreed there are problems involved in comparing live to reproduced sound.

Yet I have to say I've found such comparisons quite useful and enlightening.

I've been obsessed with live vs reproduced sound as long as I can remember.  So when I became more fervent about trying many different speakers and systems, in the field but especially in my own home, I made decent quality recordings of familiar acoustic sounds - my wife's voice, sons, my acoustic guitar, my sons playing their school instruments - trombone, sax, etc.

Being extremely familiar with those sounds, they really illuminate how closely a speaker in question can reproduce them.  And sometimes I would do direct live vs reproduced comparisons (typically with speakers in my home).

Lots of speakers fail the comparisons, but some are surprising. 
My Thiel 3.7 speakers, for instance, reproduced the sound of my playing my acoustic guitar with fairly astonishing accuracy - timbre accuracy, clarity, etc.

My MBL 121 omnis can reproduce the sound of my son playing sax with amazing verisimilitude. 


But I don't always need those recordings on hand.  When I go to audio shows, or high end stores listening to systems, it's often the case there are real voices nearby (obviously at audio shows, and often at a high end store the salesman may be talking to me or chatting with someone else).

So I often stop and take stock: if I'm playing, say, a good recording of Johnny Cash or some other simply mic'd male voice, I close my eyes and listen to the real voices present in the room and compare it to the reproduced voice.  What IS IT that the real voices have that distinguish them from the voice coming through a sound system?  It's always very telling, and the systems that actually have the least obvious departure from the real voices are the ones that  I inevitably find the most mesmerizing and satisfying for long listening sessions.  (Naturally the more complex and demanding the music source may be, the less able a modest system will be able to keep up.  But, in general, a system that captures certain essences of live sound, to my ears, tends to predict longer satisfaction FOR ME).