This thread is like polar opposites unwilling to give a middle ground and overstatements abound. The leaked footage looks like garbage. The documentary looked like garbage (and grey scale for the most part haha). The promo SP shot looks average at best (normal map attack!) The EGM footage looked like garbage for the most part. The E3 2006 realtime trailer looked awesome (I thought it was CGI the first time I saw it).
What will we get in the final version? Who knows! SP/MP could be totally divergent, they may have cut dozens of features, they could be running builds with placeholders everywhere for gameplay purposes... who knows.
Divergent? Honestly, I'd say count on it. Their entire engine is geared towards a totally different goal than multiplayer. 13
+ mile draw distance? That's not exactly useful for most MP matches. Atmospherics simulation (Rayleigh/Mie scattering)? It'll certainly have uses, but as they've said, it's not the same. An average encounter described to us included 25 on-foot enemies, 14 vehicles, 6 marines, plus more enemies streaming into the area. At other times, there'll be enemies miles away (such as Banshees), which you could kill if you had the right weapon. The scale is totally different.
On the last point... gameplay... what has me curious is how are they progressing gameplay. I thought Halo1 for its time did a really good job on taking elements tested in other FPS games but not perfected or popularized (vehicles, weapon # limit, grenades, melee) and did a great job of balancing them out as well as developing a rewarding health system and quality AI and hit the "jackpot" of right time/place with system link/Halo parties. Halo2 was shorter but really hit the jackpot again with production quality and Live.
Halo 3's been handled much differently from Halo 2 as a whole. At e3 2002 we got a teaser; at e3 2003 we had the 9 minute gameplay demonstration (but it was essentially little more than an engine demo, then they had to rework their renderer and the level wasn't usable in of itself); finally e3 2004 brought us just a taste of the MP side of it. At the equivalent of 2002, we got nothing; 2003, an announce trailer, and at the same time as e3 2004, we're getting a public beta of MP. For one, I think we'll see Single player in July this year. Two, they got burned by that '03 demo they did. Which leads to...
Everything they've talked about has been handled differently from Halo 2. Going through Halo 2's development, I can remember a ton of vague outlines regarding how ambitious they were going to be. Tons of this stuff never made it to the final game. Total opposite with Halo 3. Chock full of comments regarding the certainty it'll be in the final game, actually--and then most of the time we get only the vaguest of hints of what it might be. But we haven't heard of anything that won't absolutely be in the game, with the exception of minor details--flocking birds, for example, where the caveat was "if it so much as twitches, it gets cut." Jaime Greisemer also made note of how much easier it is for halo 3 to experiment with something--that it's built into the schedule, whereas the same wasn't true with 2.
The point being: If they can evolve the gameplay in a way that makes sense, I think they will. For halo 2, they had time to create better level designs (unfortunately, more linear and "best path only" too), added vehicle boarding for you and AI opponents, marines could drive, and dual wielding. This time, what we know is that we'll get far larger environments (regardless of how much is playable); there's the potential to have a more sandbox style to the game, like Halo 1; New equipment deployables
will change combat significantly (portable grav platforms, the bubble shield, possibly a sheild-draining device, tripmine, more that we haven't heard of yet, etc.); and we may get a new class of weapons, heavies, which behave differently from the standard weapons (rumor: you can't crouch, walk slower, can't switch weapons, etc.). Of course, everything in there has been done, or probably has been done, before. Nevertheless, the game's features are being expanded, and if what Halo 1 and 2 did was good enough, I think the same will be the case here. The sense of scale will be the biggest one, though: A truly large scale, whereas the first two games either used, or failed when using, tricks to make you think it was bigger than it was. Here, the levels will be much larger, the enemies in combat much larger, the allies you have more plentiful, and so on.
Yet thus far we haven't seen much from the gameplay. I know a lot of fans want Halo supersized... but I would prefer a true next gen effort that evolved the gameplay. Halo 3 will be a solid game if it is pretty graphics and Halo gameplay... but I think it will ultimately leave a lot of us feeling flat. Sequals tend to do that as devs get shoehorned. The fact MP is still 16 players is a concern to me, as well as no bots. (Halo fans will disagree... change is uncomfortable... more of the same!)
What ways, specifically, would you like to see the gameplay evolve? If I were to get a vote for one thing I'd want them to experiment with... I suppose it might be more dynamic levels--more destructible levels, especially with human structures, where it makes sense for a rocket to blow bits and pieces of them apart.
Personally a lot of sequals (MGS4, Halo 3, GTAIV) are not doing much to pique my interest at this point and are instead living on their pedigree. While you would hope they would evolve and live up to expectations I have some doubts and think a number of big name sequals will falter -- only for this generations "Halo and GTA" to take their place. Games like Mass Effect, Assassin's Creed, Battlefield Bad Company all come to mind (quality devs, proven track records, forward thinking designs that break the mold).
Points 1 and 2 go for all the games' developers you mentioned. Point 3, I think that certainly applied to those games when they were new IPs. You also mention another sequel in there. While we might set such low expectations internally, in the hope they're surpassed, I don't think that metaphorically placing bets on those low expectations would be a good idea. And of course they're not going to pique much interest when we haven't seen them yet. But that's a luxury afforded by having a "guaranteed thing" such as those 3 practically have. I'm not going to knock them for not spending time doing special work just for the sake of showing us stuff earlier, when that time could be spent on the actual game. See the e3 2003 demo, all of which had to be scrapped because it was nothing but an exhibition demo.
This has been a strong theme with 3's development. We got a teaser with tons of placeholder stuff in it (textures, the billboard plants rather than the 3D, properly lit and shadow-casting, physics-driven (wind, objects/characters), destructible foliage and other decorators. Certain things were polished to make it more presentable, but they didn't do a bunch of work that would be scrapped--that's a location and event from the game. I imagine they even chose that scene partially because they only wanted to do a teaser, and not reveal too much, but also due to the engine not having gone through final optimizations. Which isn't necessarily a bad thing, as Halo 1 and Gears go to show. And the same thing is true with the MP, where it's been pointed out that certain things are the way they are because they're not getting out of line with their schedule for the sake of making one way or another for a beta test, whose purpose isn't to show off graphics.
We can all hope, graphically, that Halo 3 is more This:
than the "stuff" in the leaked footage. It seems based on the beta we may be waiting until late fall 2007 to know one way or the other. But as much as graphics may be a point of discussion, I am more concerned about gameplay. A lot of games just are not moving the yard stick forward. Halo 3 may be too important to take risks and experiment.
They're not going to do something that doesn't make sense. But, if anything, they're experimenting more than they did for Halo 2. That's a good direction, IMO. And as for the graphics, they've said the engine itself has only improved, and hey, the art has improved as well. They're certainly more ballsy/confident about their technology and art than I remember them being for Halo 2 (not that BS about being better than anyone else, but rather that they're not behind anyone). Not undoubtedly a good sign, but also not a bad sign either. Which is a good sign.
to guess that with a year and a half on actual Xenos/Xenon hardware now that no matter what screen shot critics may say, Bungie is going to release MS's generational FLAGSHIP title as a technically amazing game (graphics too) with HDR and AA even. 
