I've read a lot of posts about making guitars that last a long time, 50 more years, that sound like a Martin or Gibson. Then in this same posts I've seen arguments as to why that approach was right or wrong, even though the original poster was simply expanding on the basic designs of the manufacturer they were studying( i.e. tone Vs. warranty ). But, I'm curious about ideas beyond those boundaries.

What if there were one instrument you could make, and it were played one time and it sounded the best it could, given that "best" fits within the standard definition of western music.

What would the instrument be? And what would be the basic construction?(i.e. flat top, 6 string tuned E -E , fan bracing with steel stings...) Forget about 50 years from now, we're all dead, you cant hear it anyway s, what about now?

Be aware that these things are systems: often it happens that things work in ways that are hard to predict, and changes that might seem innocuous could jump up and bite you.

Look at any of the instruments people show on here. Everyone's trying to do that on every guitar they build...

I'm not looking to be contrary, but you sound like you are really asking what is the best way to construct a guitar for the best sound and defining absolutely nothing.
(I'll take it you are talking SS, not classicals)
For a start, "best" is subjective. e.g. Martin/Gibson/Taylor/Lowden etc (just some "big" names, use Olson/Somogyi/Kragenbrink if you prefer)
Sound? - to play what type of music? Fingerstyle/Blues/rockabilly/Jazz/Celtic.
Best - Responsive? - see Gore/Somogyi books (possibly)
Instrument - parlour/dread/OM/archtop, etc?
Played one time? - no break in time? - - Kind of rules out spruce for a top.
Played one time -then it collapses? - Really, no-one tries for this (as far as I know)
I've sometimes heard that the "best sounding" of any type of guitar is built on the point of collapse. Doesn't mean it's the "best" guitar.
Get my drift?
BTW - also see above post.

No clue. But I am pretty certain that there are no shortcuts to developing your own answer to that question.

The best guitars I've built were built very light. I keep going in that direction pushing the limits. I highly doubt they will last 50 years without at least a neck reset but who cares? Maybe some people do but then they won't be playing one of my guitars. Having done repairs for 20 years has been a great help in seeing how old guitars were built and what they turned out to be. An old guitar can still be non responsive, the age has little to do with it if it's over built in the first place.

Yep... The real devil is... What's optimum?

Then - if you poke around, you will find that there are infinite variations on bracing schemes.. Some are wild, some are just slight modifications of standard well understood patterns... and yet - the guitars always sound like guitars.... There are certainly differences within the guitar sounds... but they don't sound like Pianos or Trumpets....

THEN - if you want to play amplified - you don't want your instrument feeding back all the time... so you would like it designed for such.... and there are trade off's there too.... One of the beauties of normal, super duper overbuilt Factory guitars is that they are generally well behaved on stage....

I guess - what's YOUR optimum?

One definition of 'optimum' is: "the greatest degree attained under implied or specified conditions". You have not, as others have pointed out, specified very much, so it's hard to guess what the 'optimum' might be.

What makes you think that it has not been reached? After all, generations of makers have been trying for it. To think that they may not have achieved something close to 'optimum' design implies, as David Pye said, that they were all fools. All bets are, of course, off, if you introduce a new material, such as carbon fiber composite; then a new, better 'optimum' might be said to exist, and you
ll move toward that.

But note that 'optimum' is not the same thing as 'perfect'. The latter implies that there is some level of performance or state of being that cannot be improved upon, whether or not is has been, or can be, realized. 'Optimum' is simply 'as good as anybody's been able to do under the circumstances'.

It would seem to me that the way to go about this is to first look at the existing designs that use technology you're comfortable with. Decide for yourself how those depart from 'optimum' as you see it: perhaps, as you imply, they are over built, since in your world longevity (and the reliability that goes with it) are not as important. This will lead you to design changes that might 'improve' the function as you see it, without incurring unacceptable costs. Be aware that these things are systems: often it happens that things work in ways that are hard to predict, and changes that might seem innocuous could jump up and bite you.

The important thing to see is that existing designs are probably nearly 'optimum' in some circumstances. What you need to do is define the differences in circumstances that interest you, and figure out how to change the designs in response to achieve a similar degree of optimization in the new conditions.

One aspect of systems engineering that is counterintuitive is that optimizing all the parts of a system does NOT necessarily produce an optimum system. That plus the difficulty of defining what the optimum system is makes this a retorical discussion.

This may interest you http://www.brunner-guitars.com/?page_id=387