[1] https://www.disneyanimation.com/technology/innovations/matte...
I worked on a GPU-based real-time path tracer prototype at Pixar, which became the RTP renderer used for their in-house lookdev tools [1].
[1] http://on-demand.gputechconf.com/siggraph/2017/video/sig1738...
I played FPS games before HL, and I played all the original HL games including Blue Shift and Opposing Force, plus a bunch of single-player fan-produced games and even Gunman Chronicles, all using the engine, and I don't follow what you mean here.
When an HL1 object is “on fire”, for example, it’s emitting “infrared-radiation particles” as hitscan shots off of a point-source at its center. If these hit something, the physics sim checks the “on fire” object’s “hotness” attribute, and its distance from the target it hit, and from these computes a “heat impulse” that should be given to the target, which may or may not be enough to set the target “on fire.” Different things burn with more or less “hotness”, and for more or less time before burning out, depending on the material they’re made of. Explosive barrels don’t have any special code—they’re just “on fire” for a very hot, very short time (but they’re also “concussive”, which is a separate instantiation of this same logic.) The Crossbow’s alternate fire just slides a physical object that’s intensely “on fire” through the scene at high speed. Etc.
This system continues into HL2 as well. Pheropod/bugbait objects are just a copy of this system, where “on fire” is instead “acts as ally”, and antlions can be temporarily or permanently caught “acts as ally” by getting hit by a strong enough dose of pheropod radiative signalling.
This system is basically why HL1 and HL2 play the way they do: rather than only being able to kill things with weapon-objects programmed to emit “kill” signals toward enemies, you can kill things with pretty much any object, in a variety of ways, as enemies are physical-sim objects that can die for physical-sim reasons. And it’s very easy to make a useful new object in these engines, as you can just model something, give it some interesting physical attributes, and plop it into a level. The sim will then take care of making it useful.
Whereas Half Life 2 was definitely a big ol tech demo of the source engine.
[1] https://www.ign.com/articles/1998/11/26/half-life-5
[2] https://www.gamespot.com/reviews/half-life-review/1900-25373...
Their first product was a game based on technology licensed from Id. It's a fairly different development path from Id and Epic.
https://en.wikipedia.org/wiki/Pathways_into_Darkness
https://en.wikipedia.org/wiki/Marathon_Trilogy
The stories of these games have been the subject of discussions for many years.
http://marathon.bungie.org/story/
You may know Bungie as the studio behind Halo: Combat Evolved (2001) and subsequent games.
Before Microsoft bought Bungie, They open sourced the Marathon code and gave the game data for free. Luckily people have been working on bunch ports for modern OSes:
Idk about that. I think Valve has simply changed their identity from "Game Developer" to "Gaming Platform Holder".
They have continued to experiment with new tech like they always have. They've made significant contributions to VR, streaming, and (for better or worse) monitization - technologies that are in service to their platform.
Valve as a game developer isn't dead. It just takes the back seat to Valve as a platform holder.
I’ve given some thought to this, and I invite you to do the same: if there was an HL3 that continued in the tradition of serving as a tech-demo for a new kind of realtime physics-sim tech, what are the possibilities for the tech it could be showcasing?
I’m not too deep into physical-simulation research, so I can’t say whether there’s something just sitting around as a journal paper. What I can say is that I haven’t seen anything in a modern title (even an artsy indie title) that would have made for a good basis for an HL game. Have you?
HL2 had some interesting mostly-environmental exposition intermixed with an extremely lame narrative arc. You are basically just a guy with a gun who spends most of your time in transit. If you swapped places with Barney or a random citizen in an early scene, nothing would have changed. Your character is sterile. The main character was the environments, some of which were very good, but it felt more like a very long "demo" than a game. Everyone expressing surprise at how you are in fact Gordon Freeman and it just doesn't matter at all. You're a lunatic with bullets to spend, plus a couple science fair level physics ramps to complete. Tomb raider, every single one of them, did physics puzzles better, and they didn't have a physics engine!
I think everyone was wowed at the tech at the time and didn't realize just how bad the rest of the experience was. Replaying it, its very easy to feel (compared to replaying Deus Ex or Thief or HL1)
There was no emotive force in HL2. The story become uncompelling. I think they realzied HL3 would fall super flat without some new technology to showcase, because they made the Gordon story so damn boring. (You'll notice in a lot of fanfics, people actually talk, or they aren't about Gordon at all.)
Now combine that with what the GP said about HL1, some A-tier object manipulation and the impressiveness of VR and you have a game that's a lot of fun to play even if you're just along for the ride (in terms of plot). Throw in a decent story arc, hook people with letting them actually make some tough decisions (earth is descending into a dystopia in the time it's set so plenty of opportunities to force the player to choose between a bunch of bad options and feel guilty about whatever they choose) and you have a highly immersive experience.
I don't mean to make it sound trivial, each one of those things is hard to do right and they all have to be done right but it should not be rocket science. Valve just needs a Decent(TM) game using their existing principals. The hard part is going to coping with the constraints of VR without neutering player freedom within the story or botching the VR part.
https://www.metacritic.com/browse/games/score/metascore/all/...
A good analogy in other media would be Citizen Kane (movies), NWA (music), or Akira (japanese animation). Neither of those would garner mass praise if they were released today, but they were first of a kind and paved the way for moving their respective medium forward.
But I think their goal with HL2 wasn’t really to deliver a compelling narrative, any more than the point of a modern Mario title is to give a compelling narrative. It’s a sandbox with a goal condition. (Sort of like Scribblenauts, now that I think of it.)
> Products are usually the result of an intersection of technology that we think has traction, a group of people who want to work on that, and one of the game properties that feels like a natural playground for that set of technology and design challenges.
> When we decided we needed to work on markets, free to play, and user generated content, Team Fortress seemed like the right place to do that. That work ended up informing everything we did in the multiplayer space.
> Left 4 Dead is a good place for creating shared narratives.
[0]: https://www.reddit.com/r/The_Gaben/comments/5olhj4/hi_im_gab...
Half-life basically helped kick start story-driven FPS games (Bioshock, etc)
Half-Life 2 helped kick off physics based FPS games.
Portal basically kicked off the FPS puzzle genre (Talos Principle, QUBE, Quantum Conundrum, etc).
L4D basically kicked off FPS co-op games (Borderlands, Killing Floor, PayDay), as well as asymmetric co-op.
For better or worse, TF2 with hats and later CSGO basically started the microtransaction/crate system.
Lastly, Dota 2 did a lot to push e-sports forward too.
This is the kind of advances Valve is interested in. It's hard to tell now, but this game could easily have a huge impact on how VR games are made in the future.
This isn't a technology, rather it's a set of game mechanics.
Both Portal games are cool though.
HL:A possibly represents a watershed. The 40-hour, single-player narrative action VR experience (not to mention a Counterstrike VR mod). What recently released Asgard's Wrath referred to as "God-scale" gaming
But I would say that, most of the time, Nintendo (and similar allied shops like HAL and Intelligent Systems) just invent novel game mechanics out of thin air, then ask how the player should interact with those mechanics; they make the interaction with the mechanic as a prototype, and playtest it on its own; and then, if that turns out to be fun, they wrap a game around that.
This is why, for example, the Paper Mario series looks the way it does. Nintendo + Intelligent Systems both agree that you can’t just take the same mechanics and make a new game, even if the mechanics you came up with the first time were really good. They already explored the possibility-space of those mechanics as well as they could manage in the first game!
VR could play a part in a physics simulation.
• HL1 was an exploration of the possibility-space of having objects interact through emitted “energy” (i.e. by broadcasting messages that get attenuated by distance.)
• HL2 was an exploration of the possibility-space of putting a dynamics simulation engine in a game (and also spawned Gary’s Mod to continue said exploration.)
• The major offshoots of the series, Portal and its sequel, were explorations of the possibility-space created by combining “beams” of matter or energy (lasers, hitscan gunfire, tractor beams, paint), gravity, and wormholes.
In other words, each major new entry in the Half-Life universe was essentially a narrative wrapped around the results of playing around with implementing some journal paper (or reimplementing some prototype) of a novel type of realtime physical simulation, that had obvious game-mechanical implications.
I’m not surprised that we never got an HL3, because for the past 10-odd years, there hasn’t really been a big new obvious realtime physical-simulation tech to turn into game mechanics.
I’m also not surprised that Valve would return to making games under the Half-Life brand in order to explore the implications of a new technology (even if it’s not a new realtime physical-simulation technology. Or maybe they’ve got one of those up their sleeves as well...)