Videogames that are predicated on the novelty of their technology may be seen as inferior to those predicated on the novelty of their gameplay. A natural prejudice, given the core of gaming - it's the gameplay, stupid!
Gameplay trumps all, however, seems an incomplete way, to me, of looking at each and every game (even though I've probably said that damn mantra before). There are many other things which I take into consideration that might seem shallow to other gamers, but which I may place on equal terms (depending on the game).
When I look at graphics I consider them as providing clarity of vision. In a medium which engages me first and foremost visually*, the way those graphics are presented can be very important to me.
I don't like aliasing. To me, it is a constant, glaring reminder that I'm but a set of eyes gazing at a matrix of intersecting squares. It certainly doesn't detract from gameplay when I go from a higher resolution to a lower one in order to make the game playable, but on an aesthetic level I am disappointed - it is as if I were allowed to see the Mona Lisa for but a second and then told that I can only really look at her if I smear Vaseline over my eyes. Gameplay may not suffer, but the experience drops in clarity.
Clarity does not presuppose any specific technical stats. But it is concerned with giving a designer as many options available with which to create their vision. Textures that go blurry when a player-character gets close are an eyesore to me - it's not simply a question of making things shinier, or prettier (though why should those be such awful goals?), but rather a question of offering more control. I'm not even talking about immersiveness here, but about what I, as a designer, wish to show to the player.
Once clarity is improved, it can always be muddied easily.
I'm waiting for the day when I see a game with jaggies as a tribute, the way some bands will put the skrtch-skrtch of records revolving underneath a song on a CD.
The driving-forward, then, of visual technology will continue to happen. It may be that there is an upper limit to that sort of thing - I won't pretend any knowledge on that matter.
Calls for intelligent agents and complex stories in games tend to ignore something - progress is constantly being made, just not at a rate equal to progress in graphics technology. But those are two very different things, possibly not comparable at all. Shift the entire industry focus to some idealized notion of better gameplay involving smart AI and engaging narratives would not be a guarantee of, well, anything.
This doesn't mean that people should cease to call for such things - only that decrying HD or dynamic shadows or fancy particle effects as a detriment to gaming, or even counterproductive, seems kind of ridiculous.
Yes, I'm saddened that the architecture of the Cell chips seems to prevent complex AI implementations but, well, where is this amazing AI? It must be somewhere, or else it is speculation. Will AI take a step backward because of the chip? That remains to be seen.
Many artistic movements have had debates concerning the evolving science of their art. I'm not worried at all for videogames, nor am I thrown by the crazy notion that games might look better but not really play all that different. It hasn't bothered me yet in my twenty-some-odd years of gaming.
Chess doesn't play all that different hundreds of years later. Well, unless you're on acid.
*And, yes, aurally, emotionally, intellectually, reflexively; But before those things, and as those are engaged, it hits my eyes.