You are referring a quarter of a century back.Graphics enable the gameplay. Tetris works because its graphics work. The falling blocks you've got to clear fit exactly into the graphical capabilities of the first Gameboy.
Compare the gameplay of Path of Exile 2 (2024) with Diablo 1 (1997). It's not just the same type of game with better graphics. The graphics in Diablo didn't allow to change the direction of an animation or for a large monster to push smaller monsters away or to dodge roll between the legs of an area boss. The gameplay is determined by what was graphically possible to be displayed on screen.
What's the appropriate Steve Jobs quote: "Design is not just what it looks like and feels like. Design is how it works."
Today,
a) we have a lithographic slowdown
b) we are discussing what methods to use for ambient occlusion
Not only has lithographic advancement slowed to a crawl up an exponential mountain, (making advancements in computational resources expensive both in terms of cost and power), but the visual result arguably hinges far more on art direction and asset creation than on computational methodology.
We are up against steep walls of diminishing returns.
I’m a computing tech nerd, and have followed both lithographic development and game evolution since the early 1980s, and using examples from that era to predict or motivate today just doesn’t make sense.
(It’s ironic that you bring up Diablo - a friend of mine sits with a RTX4090 and plays Diablo 2 and Grim Dawn, I have an RX7800XT and currently play Mass Effect and a bit of Baldurs Gate 3 (on a Mac).)
These days, gameplay fun is unrelated to graphics technology. How much more fun does a fetch quest get if the lighting is ray traced? The technology development of today is mainly for marketing. The established businesses need to keep the revenue flowing somehow. But the new gear doesn’t make gaming better per se, anymore than exotic materials makes golf a better game.