If the thing exists, then a Perception check can find it. If the thing doesn't exist, then no amount of success on a Perception check can possibly find it. Whether that thing exists (and can be found) or doesn't exist (and thus cannot be found) should have been determined by the GM prior to the player character searching for that thing; it can't causally depend on whether or not the character is looking for it.
Now we're getting into the philosophical. The old tree falling in the forest when there's no one there to hear it. Nature of the observer & the observed. Even IRL, where most of us, children of a scientific age, have no problem believing there's an objective reality, the power of expectation and perception to shape how we experience that reality is well-known.
For an imaginary world, so much the moreso. If the DM hadn't thought about a detail, and a PC goes looking for it, the DM could decide, on the spot, whether it was there to be found. Or, he could let the PC make the check, and, only if he succeeds, make that decision (because the detail remains irrelevant if the PC doesn't notice or interact with it in any way). To the player, there's no difference between either of those and the DM deciding beforehand.
Once the game starts, the player (in most editions of D&D, and many similar games) has zero authorship power beyond the in-game-world capabilities of the PC.
That's mainly a matter of style, not of the game, itself.
Moreover, from a Sim/RP standpoint, the player doesn't want authorship power over any part of the world's history.
Very much a matter of style & opinion, especially with regard to RP. A player might well want to fill in gaps in his character's backstory long after chargen, for instance, so as to best RP the character.
Asking a strong simulationist player to repeatedly assume the author-stance is effectively asking that player to leave the table, because this game isn't for them.
Frankly, anyone who would leave the table over something like that (whether because they feel 'forced' into the stance, are are whinging over being forced out of it) should leave the table whether they encounter the issue or not - and try an alternative to TT, because they just don't play well with others.
Whether the defender parries is something that happens right now, and is a direct reflection of something that is within the power of the character. The player doesn't author anything. The player makes decisions on behalf of the character, but it's entirely the character who successfully manages to control the outcome of an action.
How a successful parry is described may be well within the realm of 'player authorship,' and even whether it does can depend on the player using a resource the PC may not have direct control over or awareness of (like 'luck'). And, though I'm getting tired of saying it, the character still doesn't exist - at most, you might assert that the resolution mechanics 'control' the outcome of an action.
And if the character fails to hit, then it's the opposing character who succeeded in parrying or dodging or whatever. The fact that the rules (in D&D) give the defender a static AC of 10 + bonuses, rather than asking the defender to roll d20 + bonuses, is irrelevant to the fact that the defending character is actively defending.
I'm pleasantly surprised that you grok that.
To contrast, any detail of the enemy's armor would have been determined long before this combat ever started, and is not something that the character has any control over
First half of that statement is false, the DM may not have made any prior determination of such a detail. Second half, OTOH, sure, the imaginary character has no control over the imagined nature of the imaginary armor.
That fact of the game world is beyond the agency of the character, and is thus something that only the GM can determine, and which should have been determined prior to the check being made.
Again, it depends. In 5e, ultimately, everything is under the DM's control. In other eds & games, maybe not so much. But, even in 5e, the DM can cede some of that control to existing or improvised resolution systems, or, indeed, to the players.
Yeah, that's kind of the point. In a simulationist game, the players have no agency beyond that of their characters. Everything that happened within the game world, prior to the game actually starting, is a matter for the GM to determine. Everything that happens after the game starts is determined by the GM and the player characters.
Simulationism is an attitude toward games. Simulationist may like games that sacrifice playability the way actual simulations do, but that doesn't make the game 'simulationist,' and no game can be held to that philosophical extreme. Certainly not a game as far pre-dating the concept, and as murky in it's beginnings, and as intentionally inclusive to a wide range of styles in it's current incarnation, as D&D.
And it's not entirely a binary state, either. Games can be more or less simulationist, but the more you slide along the spectrum toward narrativism, the further away from simulation you end up.
Narrativism and Simulationism can certainly both ruin a game if taken too far, because they can go and sacrifice the things that make a game at all playable ('gamist' considerations). But, that's more a matter of how a game is played, than what the game is. A game is a game, and RPG is a specific kind of game, but a simulation and a narrative are not games, at all, nor or they mutually exclusive. An historical re-enactment, for instance, is both a simulation and a narrative.
