Navagon: But even Civ4 was four years ago now. It was scaled down because the engine wasn't optimised enough (Total War series proves that more could have been done) and because it looked to me like they valued prettiness over quantity.
That doesn't change the fact that more could be done now, just so long as you've got the right developers behind it.
But that also does not change the fact that they could have done better game continuing to use 2D engine nor is there any guarantee that next Civ or one after that will be better than if they'd continued to make Civ's with 2D engine. The point I'm trying to make is that 3D by and in it self does not provide any benefit if your game does not have any features that could be made only in a 3D engine. Stick figures will remain stick figures no matter if they are made with 2D or 3D. Military unit symbols remain just that if you make them with 3D or 2D. etc.
Same goes with many other things like voice acting. Voice acting can improve a game sure but bad voice acting can ruin it. Also due to cost and time the voice recoding takes ammound and variety of dialog have to be reduced so unless you can hire good enough voice actors and number needed you're probably better off with completely or mostly writen dialog. Voice acting by and itself does not improve the game.
Same with CGI animation over drawn/2D/engine run cutscenes. Sure CGI cutscenes are wonderfull and all but making good ones cost both time and money so what's the point if making all those wonderous CGI cutscenes means cutting back in everything else? You may well end up with completely shit game that has great cutscenes. Why not make CGI movie instead if you care so little about the game it self?
And the graphics whoring these days. It sometimes feels that some teams have 100 artists/modellers for every animator/programmer/AI scripter/writer/worldbuilder etc.. Yeah, sure that way game gets beautiful graphics but game is buggy as hell, engine broken and unoptimised, stick figures have better animation, 6 year old could write better dialog and story on his/her spare time, AI gets lost every few meters and can't find it's way around single rock and world is stale and static as hell.
I probably could go on ranting all day about current state of game development but hopefully I've made my point. In short, no technology or feature should be used if it does not provide clear benefits (no, just being pretty or cool does not cut it even if marketing loves them) and no single area of game development should be overemphasised over others especially if it means other area(s) is negleted.