First, that games have become more buggy over the years due to increasing complexity.

And two, that games have not become better than past game and, in fact, have become worse.

Because that’s the type of person I am, let’s start with the second point first:
I heartily disagree. I’d say that it’s plain confirmation bias and availability bias. That’s to say, that first you only remember good games from old systems. After all, they are the only ones that were able to stand the test of time. Sucky games came and went. Everybody remembers Pac-Man, but nobody remembers Pac-Man for the Atari 2600 (or, at least, they don’t want to).

So, we’re only remembering the good games (or mostly good games) and that list is static. They’re not making new games for old systems, after all. But they are making new games for new systems. And, since it’s in such recent memory we can much more clearly remember bad games when they come. They stand out since they just happened.

In other words, we only remember the good games from old systems and since new systems have both good and bad games, we have a much more negative view of new games.

And now for the first part:
This I agree with. And, yes, it has to do with the fact that everything is getting a lot more complicated. There’s a reason why it costs millions of dollars to make a triple-A title and most of it is due to just getting all the basic stuff setup. The game industry is horrible when it comes to re-using code. There isn’t really any big open-source game engines, at least none that are equal in quality to say the Unreal engine.

As we get more and more complicated computers, it’s a combinatorial explosion of bugs.

Now, is this increased complexity worth it? On a game by game basis, it depends on the game. For something like Mirror’s Edge (which I really enjoyed) the game wouldn’t be the same without the high-powered graphics and physics. It would totally break immersion and be an entirely different beast. On the other hand, I’ve run into a couple games where I said to myself “This game would be a lot better in 2D”.

But is it worth it on an industry wide basis? I really don’t know and that’s because I really don’t know where the industry is heading. It’s liking asking is it really worth it to build huge particle accelerators to search for confirmation of the Standard Model. We’ll only know when the answer is staring us right in the face.

And yet, I think that game development has become a cycle that obviously produces buggy games. Ever since the mainstream had a whiff of Halo and GTA, and subsequently sunk it’s teeth into the gaming industry, investor dollars have flooded in like no other time in history. It stands to reason that when the mainstream pushes this industry to be more like the movie or music industries, we get games that are rushed out the door so they can coincide with pricey advertising campaigns.

What I’m trying to get at is that while I would like the adult median of games to come together with fists upraised demanding that developers focus more on quality, I doubt it would happen. That this is a mainstream process now, we can assume that the larger companies like Microsoft and Sony want developers to accommodate to the lowest common denominator. This LCD, of course, being people that are captivated by the advertising campaigns and hype. And that is where the loop appears and transforms the gaming industry into the controlled mess it currently is (and forces me to loose faith in my former favorites.)

Maybe we just need another E.T. to screw this whole thing up so we can start from scratch again.