Well, I suppose I'm not the assumed target audience. Back when I was younger I had the time to tweak everything on my computer: my Linux distro, my games (dualbooting windows just for that), and looking things up for that reason. I also could play long gaming sessions.

Nowadays I'm a dad with practically no spare time between raising a toddler and work. Suddenly the question of "does the game overstay its welcome?" has become a thing; I almost exclusively play games that can be played in short bursts of time, and that deliver a great experience as a whole that can be completed in a relatively short playtime. I got a Steam Deck a few years ago for the specific purpose of separating my work computer from my gaming platform, and being able to pick up and play a game and pausing it without problems.

Even with the built-in performance overlay of the Steam Deck (which is very nice) it takes time to assess the quality-vs-performance results of every possible combination of settings. Often more than I would spend on playing a game.

I suspect that people like me either already are or soon will be the bigger segment of the (paying) customers though, so that is something to consider for developers.

And some games do give short explanations of what each type of technique does and how they compare, along with statements like "this usually has a small impact on performance" or "this has a large impact on performance" to guide me, which is already a great help.

Not a parent but I still agree that short games are great, 2-3 hours are great (or even a bit less, there is a reason for the standard 90 minute movie). 4-5 hours can be nice too ("chapter" divisions are helpful). Games are inexpensive and plentiful these days so a nice short game is great for everyone not just people with little time for playing games. (I guess I should say the flip side is that time spent with characters is one of the interesting things that games can use to good effect and bonus objectives that encourage you to explore details of the world can be nice too depending on the type of game and are easily skipped).

The best game settings have a scene to illustrate the effect of choices along with estimated(?) performance. Unfortunately I haven't seen that too often (mostly Falcom games that PH3 worked on). I agree that stating the impact on performance is quite helpful when settings need to be lowered. Usually they are ordered "best performance" to "nicest looking" so I set the last one and only fiddle with them if necessary (a smooth 12fps works great for me so luckily that isn't often).