With respect, the basic question is nearly meaningless, and the conditions are mostly irrelevant.
I cannot think of any way of establishing or measuring "peak performance" for a guitar, since "performance" is not a single characteristic, nor can it be defined satisfactorily for more than one person--nor will even a single person's notion of "performance" necessarily remain static over the long term. I am the original owner of a 1965 D-40 (purchased in 1967), and while I value it highly, my attitude toward it has not always been exactly the same. I am pretty sure that had I made careful recordings of the guitar over the last 40+ years, I would detect changes in its sound, and that I would think them changes for the better. But I didn't, so that's just conjecture. I do know that I currently find it a very pleasing guitar, particularly for slack key.
The only useful notion of "lifespan" I can imagine for a guitar has to do with physical structure, and the violin family demonstrates how the components of an instrument can be maintained and repaired over centuries as glue joints deteriorate or cracks develop or tuning pegs wear out. The modern guitar hasn't been around as long as the fiddle, but I own two Washburns from around 1900, and they have had neck resets, cracks cleated, fingerboards and bridges and frets repaired or replaced. I have a 1920 Martin that is all original except for a replaced bridge. I suspect that their characters were established long before I bought them, and I see no reason that they should not remain not just playable but very satisfying for the player who acquires them from my estate later in this century.