The dialogue thing is really complicated especially if you think that the have to do it in at least 5 or 6 different languages...
That's true, I hadn't considered that. You'd basically have to write a whole new rule set for each languages grammar rules. Though that seems like it'd be a more time/resources problem than a "we can't do this yet" problem. If there's anything Valve has, it's time and resources.
Sounds like massive wish fulfillment, really. I don't think Half Life 3, if it ever surfaces, will ever successfully solve every creative resource issue ever suffered by an RPG title, for the simple fact that Half Life is not an RPG series.
Half Life has always been very player experience focused with an emphasis on immersion. It's also always been an engine showcase. The more engine features they show off, the more people use it.
You could almost say it was similar to a jrpg. A very linear RPG game. Now one thing that makes an RPG and rpg is the fact that rather than you improving at the game your _character_ improves in the game. Which you could very well argue that Half-Life does. But instead of through things like stats and etc it's by finding weapons and secrets. I personally wouldn't see half-life and rpg at all, even a linear one. But I guess I can understand some arguments for it. One big issue I see in the idea is that while your character _does_ improve you have little control over his improvements, outside not finding/skipping certain weapons. There's no certain skills to deal with. Half-Life is its own experience altogether. A very linear game without pre-rendered cutscenes, etc- Idk it's a lot of things. =)
If finding weapons and secrets makes an RPG (which as a genre term is very overused, simply by dint of how easy it is to insert RPG-esque mechanics into other games), then DOOM, perhaps? The point of an RPG is to roleplay. Interactive NPCs, crafting your own adventure, making a non-linear experience. Half Life has always been linear. The maps in Half Life 2 alone are the pinnacle of linear design (there may be the odd cache off to a side, but the main progression is done in a wonderfully subtle way in that you're always moving towards the map exit. Even Ravenholme doesn't feature much / any backtracking). The ideas on display here are very RPG-esque. Pinpointing the specific flaws of character-driven narratives in existing open-world games. The whole bit about destructable, responsive terrain, intelligent AI, different biomes . . . it screams RPG-based technology. If Valve were truly willing to invest this much R&D on that front, they're make an RPG. They're incredibly smart businessfolk. Immersion is a poor argument because it differs for everyone. I mean, sure, there's an art to maximising that baseline of belief across your consumer base, and Half Life as a setting kinda succeeds and kinda fails at that in equal (almost spectacular) measures. I consider Portal more immersive than Half Life. I consider Portal more of a tech demo than Half Life, to be fair. Half Life 2 was the poster child for the Source engine, but Half Life wasn't originally even wholly Valve tech. The GoldSrc engine, sure. Modified from something else. Any one of us, I think, is very well-equipped to write an essay on the merits of Half Life and how Valve have affected modern gaming with their games. But that doesn't mean that these suggestions are anything more than simple wish fulfillment. If that's truly, really, what they're going for? Then more power to them. But it doesn't sound like the kind of Half Life game I've played before, and I probably won't play it. And it doesn't seem like the kind of third iteration in an established series Valve would produce. They're not known for taking risks, and they are known for making sound business decisions at the expense of their playerbase (see: L4D, TF2). They wouldn't change it all up just for the sake of Half Life 3 being a tech demo. They've got DotA 2, that's already their Source2 tech demo. Features the Workshop v2, to boot.
Dota 2 doesn't feature any kind of physics simulation. It showcases the visual capabilities and put in a framework to make source engine RTS's, a genre previously untouched by Source. Half Life 2 was a physics demo. I expect to see Half Life 3 taking that same route even if this leak false. Only now we have thermodynamics and destructible objects in addition to gravity. And an attempt at believable NPCs has been a part of Half Life since the beginning. The way scripted events and NPC chatter worked is clear evidence of that. The procedurally generated features they're talking about could be used to make a linear game *seem* open world. No matter which way you turn, the next encounter is x feet away outside the draw distance. Turn left or right? Doesn't matter, but it seems like it does. And when you go back and go the other way it will generate the next appropriate encounter while making you think you just took the wrong path initially. Valve doesn't ave a physics showcase for their new engine yet. They need one, and HL3 will get attention drawn to every new feature they pack into it no matter what they do.
I could be horrendously wrong, but you need extensive physics tech to handle any genre with any degree of predetermined accuracy. Dawn of War, an RTS which I'm intimately familiar with, had physics in it back in 2004. The sequel, DoW II, featured Havok engine integration with obviously handled a lot of the kinematics and so on. I'm pretty sure the Source engine licenses Havok, too (EDIT: upon Googling, Source v1 uses something called VPhysics which is a modified Havok engine). You're getting too hung up on physics alone, I feel. When Valve re-released Half-Life (classic) in Source it was in my (and other anecdotal) opinion(s) a mess because the game wasn't designed to account for actual velocity, acceleration, decceleration, and so on (Xen in particular is a nightmare because you lack pixel-perfect jump capabilities). Source was used as a tech demo for a long time, throughout its games (Left 4 Dead being another, and Portal being the one I've already mentioned). Source2 will already have physics. It doesn't need to demonstrate something like that because the age of demonstrating fancy physics engines was a thing of the past. Battlefield did real-time destruction and advanced megascapes already. id's Rage did a bunch of other things around physics and open-world environments, or maybe I'm confusing that with megatextures. DotA 2 has physics. Any need for Valve to prove they can handle wacky, wacky physics integrations they already proved with Portal. They don't need to demonstrate that twice, even with a new engine (and arguably they have done with their mesh stuff, hair and cloth, etc . . . also on display in the DotA 2 cosmetic content). I mean the whole point I was making was that that leak was hokum. I'm not saying Half-Life 3 will never exist (though I sincerely doubt it at this point). I was just saying that that "leak" is not a leak, and Valve wouldn't do half the stuff listed there because it doesn't fit with their established business trend of taking a single aspect of a game and polishing it, sometimes throwing in something quirky to mix it up with. For Half-Life, this was level design (and arguably still was for Counter-Strike, even adapted from a mod). For Portal, this was physics and natural story progression (as Half-Life was critiqued for a rather linear plot of vague science-fiction doomsday-ness). DotA 2 was e-Sports, pushing the Steam Community integrations (at the cost of, well, everything else on the client, but more on that another time). And so on, and so forth. Left 4 Dead was teambuilding and actual PvP interactions, something Valve typically didn't do much of (again, Counter-Strike's gameplay elements were finalised within a mod, not by Valve). Valve aren't going to combine every single genre in an attempt to make Half-Life 3. They're not John Romero.
The worst bit about it is- I already have trouble decorating maps in Source... Looks like I'm going to be severely outmatched whenever the new Source 2 SDK is more widely available! At least the newer source SDK is set up very similarly to Source 1's. I just love how hammer looks.
They're actually made Hammer not a terrible piece of unholy Satanism, too. I remember looking into Hammer when I was in my L4D phase. Yeesh.
It's my favourite engine editor to work with. Brushes are my life. <3 An editor not having brushes baffles me. *cough* Unity.
Brushes are an easy way to snap / edit existing geometry as far as I'm aware. Unity is a bit low-level for that (as well as Unreal), though it's incredibly likely there are third-party addons that provide similar functionality. Hammer is more of a content generation machine than a true engine editor / SDK, which is why brushes make a lot of sense.
Unity has inbuilt map / geometry tools, but they're a bit limited (or were, when I worked with Unity). It supports importing heightmap data, however, and obviously you can import 3D geometry from a range of programs (I think). Hammer lets you design content specifically within Source, you see; everything's locked-down. It affords Valve flexibility in making these geometry tools in the first place.
Oh no I know, I love how Hammer handled it, Ive been working with hammer for years now, back when I wanted to make silly Half-Life 2 mods. Now I'm making CSGO maps. =)
So, who's been playing the Dragon's Dogma PC Port? I was pissed that leak players got the first Ur-Kill, so I'm taking it easy now and going for a 100%ish run. I had to kill Lennard's pawn to finish a quest.
I wasn't too interested until I found out you can pick people up and carry them around like a big sack of flower. And throw them off cliffs. It's on my wishlist now. XCOM 2 first though. XCOM HYPE.
