Are FPS games progressing?

Back when Quake first came out, I was very excited about the ability to walk around, and almost exist, in a virtual 3D world that I could create and remodel as I pleased. The thought of what the future might hold, when the technology caught up with our ideas, seemed really amazing. It's why I got into computer graphics in the first place.

It seems like FPS games have kind of hit a wall. All the focus now is on subtle surface-rendering effects. I don't want to play a game with a bazooka so that I can notice "the subtle interplay of normal mapped diffuse and specular lighting...".

I submit that FPS technology is not actually progressing much, even in graphics. Shaders were supposed to be the end-all, be-all technology upgrade, but about the only thing they turned out to be useful for was making pretty water. Graphics are getting more stylized, like a comic book, but I wouldn't say they are getting more realistic. Add some stencil shadows and a few shaders to Unreal Engine 2, and you have a next-gen engine.

An exquisite rendering of a plastic-molded action figure. Notice the seam running down the middle, from the injection molding process:


I think that the technology is stabilizing. The engines you use now will be good for another five years. The next graphical breakthrough isn't going to happen for a long time. I think the next breakthrough is going to be in the level of interaction. We're still moving around in the world, unable to touch or effect anything except my shooting it, or more recently, ramming our faces up against it.

Why are FPS games still so linear? HL2 promised an entire city to explore, but delivered only linear corridors. The point has been made that these kind of games might not even be considered true 3D, due to their linear nature. Why not create a world where everything can be interacted with, according to certain rules? Why not create a world where you can leave your virtual home, built on your virtual property, and travel in your car to a movie theater, buy a ticket, and watch a real movie? Why can't I just get in my virtual car and drive anywhere? Why can't I buy a gun and go kill another player, and steal their car, then get pulled over by another player working as a virtual cop? Why can't my AI neighbor stop by and ask how my day in the real world went? What's the difference between working on your computer, and working on a virtual computer in a virtual world? If money starts changing hands, and virtual real estate is being sold along Highway 17, where do you draw the lines of reality?

These examples are just ideas, but my point is, we have the ability to create a "matrix" virtual environment, yet we are just creating hallways with ever-shinier wallpaper.

Just some thoughts.

I think that games are just supposed to be for entertainment. Why have a virtual neighbor ask me about my day when I have real neighbors etc...

HL2 provided a great level of interaction. Just about everything could be touched and moved. I don't think that I (or most) gamers are looking for a world they can just stay in forever. There are some people sure ... but overall it's nice to have an adventure for a few days or weeks outside your normal life and then get back to living.

Doom 3 and Half-Life 2 were the major FPS's last year and I finished them both. Both were enjoyable even though slightly linear. But really ... what isn't linear. Even your life is linear! You are provided with a few options of how to live it ... just as you are provided a few options with how to shoot things in FPS's.

So to that end I personally would like to see maybe some cooler multiplayer aspects of these games and probably some new play mechanics. I'm working on an FPS/TPS right now that I think doesn't stray too too far from the shooter aspect but adds a new little mechanic that makes it a bit more interesting.

It reminds me GTA... well I'm playing it, although not fully "insteractive" as I expected, but still quite good.

For FPS... uh.... I still think Quake3 Arena is quite fun and exciting, however newer FPS titles looks very similar (must admit the technology made the scene/FX/enviromnent even better, but I prefer GamePlay more than screens).

Nice topic and made me thought a lot...

Yeah but years ago I talked to my colleagues about one day there will be more realistic physics on the enemies when they die and hit locations (e.g. soldier of fortune), and interative scenery (SIN sort first did this, actually so did Duke3D) and weapons blowing up scenery (duke sort did this but it was done more in Red Faction (or was it RF2?). Thing is various games have done these things, and different things on their own but it's rare to see them all in one game.

Three words: Physics Processing Unit.

They're here now.

However there is still nothing on the horizon to fix level design & gameplay defects which aren't visible on screenshots.

An article I read recently on the future of MMORPG's has got me thinking.

The UAC research base on Mars is almost a real place in time and space. Can the UAC research facility and highway 17 exist in the same universe, at different times, in different places?

I'm thinking about an integrated universe, in which a coke can on Mars is the same as a coke can in Black Mesa. If virtual space and materials were finite, they would be worth real money. Our money exceeds the gold and silver that backs it, anyways, so it can't really be considered "real".

I'm not suggesting a virtual world as an alternative to the real world. I am suggesting a virtual world that you can spend a short amount of time in, and accomplish real things. Video conferencing, for example. Why not meet in a bar in the virtual world, and have a "face-to-face" discussion?

No... they aren't... do they really have to? doubtful. Would it be nice? Maybe.

Physics and time manipulation is the only new tricks for FPSs... And they aren't really well used.

Chronicles of Rid@#!* was the most evolved FPS i've played in years.

Halo....

I agree with you entirely. Up until about 5 years ago there seemed to be really big leaps in 3D engine technology but now people are just adding extra bits to them (shadows/shaders etc).

I reckon the current top 3D engines do have the potential to create something really impressive but at the moment developers are just really trying to show off certain aspects of it. Doom 3 and Quake 4 are examples of that. I have yet to see the new Unreal 3 engine working so maybe this might be an exception, I don't know.

If a little more time and care were used in creating the worlds instead of trying to meet deadlines then I reckon we could see something really special arive soon. I'll place my bets of the next Wolfenstein game for starters :)

I realised the other day that I don't typicaly like FPS. Allthough I do enjoy the AI in fear and the particles, that does seem to be a step forward.

I don't feel realism is important at all. But interaction does play a big part in making things interesting. When I made motorsiege so many things were destructible, about 400-600 meshes on average had destructible states and pretty fancy expolsions.

The Oust FPS game that I started recently with Jeremy seems to be quite good and I think that would have a lot of potential if we can get the character interaction in, and having looked at the advanced game programming book which covers destructible objects in the book, I can see adding a lot of motorsiege type destruction being possible with our new art pipeline.

This includes non linear level design as with our current system swapping out meshes with collision for broken meshes with different collision should be pretty trivial.

Just need to work out a way to get maxscript to spew out .mis position and rotation data with a b3d file.

There's so much potential to make something different. Doom and quake just seem really empty and sterile despite all the fancy rendering tech. I think an indie team with the right tools can outdo a lot of AAA FPS titles in the gameplay department, if they can find the time and motivation. FPS in commercial games has hit a wall in part due to the narrow vision of publishers, and the time demands required to get 10+ hours original scenery to get gameplay interesting.

Starting to ramble so I'll shut up now. C ya

HL2 doesnt have the plastic feel that Doom 3 has, at least not as much. Doom 3 looks better over all, but HL2's physics interaction makes up for that.

My brother was talking to me about the next Metal Gear Solid for PS3, and he said that Hideo was wanting, well, non-linear world. That which you choose when to do you mission, how, etc. But havn't we been hearing that for the last few years?

Well, the fate of any successful indy team is to become professional. I like working with professionals better than indies because they have money, and don't mind parting with it.

I like working with professionals too, but generaly the publisher gets in the way of innovation by enforcing marketings idea of what gamers want. Problem there is that most research is based on games that are already exist determining whats doing well and is going to sell. So more often than not you end up with more of the same old games. Kind of like your generic hollywood action movie, or the 40 almost identical FPS that come out each year.

At the end of the day a successful indie will often end up going where the money is, kind of a shame really :(

I thought HL2 looked better than Doom3 despite not having that much in the way of fancy lighting tech. What it did have was some nice GI type static lightmapping that added a lot more subtle depth to the scenery, rather than the harsh high contrast look of the d3 engine. It's the harsh contrast lighting that gives the ID games a plastic amature animation studio lighting look. If like me you spent a fair ammount of time at university in the film studio and played around with claymation, you will probably know what I mean.

With doom 3 I found that turning up the ambient light in the console to as much as 1.4 or 1.6 gave you far more believable realistic lighting.

I think Doom 3 looked kind of plasticy but Half-Life 2 just looked like a standard video game most of the time ... pretty flat really. Doom 3 was far more tangible looking ... so the experience was more vibrant for me.

Both of your observations on indies vs. professionals make no sense. Independent's are professionals, it just so happens that they put their own money where their mouths are instead of playing pimp and ho with the publisher. Additionally, they make games because they are truly inspired to do so ... instead of shoveling something out just because they need another pay day.

You should get paid for making a game, but you shouldn't make a game to get paid.

What 3D games is a common library of ready made objects e.g. sinks, cupboards, loos, cars, food everything and then they can implement the object in their game if they want and all the physics and anim is done they just need to insert graphics and sound. This would save loads of programming or scripting as you knew that you could include a flushing loo in your FPS really easily and remodel it to suit your epoch e.g. roman or space loo.

I'd suggest that the FPS genre is not where the big advances are being made anymore - it's in games that are defined as being 'open-ended' (single player). They're the ones where the developers are being pushed to create complex AI controlled entities who 'do their own thing' and worlds which have their own rhythms.

Jeremy said;'I don't think that I (or most) gamers are looking for a world they can just stay in forever'. I don't agree with that statement at all - witness the buzz for titles like Oblivion and X3.

The other point I'd make is that there's been criticism lately about how 90% of games are designed by 20 something males for 20 something males. Which is true. But I'd argue that the games with more open ended worlds usually allow for more creative (peaceful?) approaches to game challenges, don't rely on headshots as being the only social interaction available to the player and so can appeal to a greater demographic?

and yeah, I mean female gamers and older gamers.

I'd like to see you lot do Doom3's graphics.
I've managed it recently. But have you?

The point here is, it's easy to critictise...

Non-linear progression through a gameworld is not something you're likely to see nurtured outside of text adventures any time soon. It balloons the cost of asset generation while simultaneously decreasing its point (because a healthy proportion of whatever you generate won't be seen) - that's a lose/lose situation to any publisher thinking of investing millions.

>The point here is, it's easy to critictise...

BS, If you are only allowed to critisize if you can do a better job, then there would be no creative thinking at all and we'd all be living in the stone age...

As for the topic at hand, I'm growing tired of playing what is essentially Wolfenstein again and again. And most FPS games today doesn't even have the freedom that the Mercenary series of FP games had.


Andy

Incidentally, I prefer this approach as it helps avoid the so-called Uncanny Valley. I hated the creepy renders that Core started doing around the time of Blam! Machine Head.

'Non-linear progression through a gameworld is not something you're likely to see nurtured outside of text adventures any time soon'

But we are seeing it, or at least a simulcrum of it - and have done so starting with Elite.

'because a healthy proportion of whatever you generate won't be seen'

I agree that open ended games usually require more assets than a linear FPS, but I think the major drawback to developers is the effort required to balance and play test an open-ended game - there's many more variables involved than a linear FPS.

Also there's the whole idea of content being created, organised and combined through mathematical functions. I don't think that's ever been fully exploited. I'd pick that that's going to be an emerging 'hot thing' in games over the next 5 years. Spore will possibly be the start of a push in that direction.

Grey Alien and Sledge make a good point.





I think the problem is that modeling and texturing are simply too much work. You spend most of your resources creating a bunch of real-world items that are going to be used once, and never seen again. It's no wonder environments are so limited and repetitive.

I've read a lot of articles that say pretty much the same thing. The costs of production are so high, it takes so much work, and at the end, you just throw it all away when you are done.

A virtual VW bug costs about as much as a real one. If you go to the store and buy a hammer, do you pay $80 for a hand-crafted one, or do you buy one from a company that mass-produces them, so that you can pay $6? Virtual "goods" cost more than real items, because each one is "hand-crafted". Are they still virtual, or are they real objects that can be cloned instantaneously?

I wish there were more standardized techniques and content. I think eventually you will be able to just buy or download an object and place it in the virtual world, and have it follow the world's rules, but for now, every game company is going to keep building everything from the ground up, for each title.

Most indies aren't professional at all, thats where the stigma comes from. So I completely disagree with indies being professional. Probably less than 3% of indies would be what I'd consider professional, competent game developers.

I felt that doom 3 had probably the least tangible environments of any of th enewer FPS games that came out in the last couple of years. To me the entire game looks like a poorly lit plastic model kit where nothing can be interacted with. You just run around shooting at anything that moves, but the environment doesn't react.

The Art direction is poor in the lighting department, the textures are extremely low resolution even when displayed at 1024x768 32bit. And if they used specular maps they weren't very effective as everything has a fairly uniform plastic sheen where the specular highlights have no variation, and because of the low resolution normal maps, there are very few edges unless they are derived from the extreme low poly models.

I don't think we will see decent graphics till geforce 7 and shader 3 are mainstream minimum requirements to run modern games.

The key word was "nurtured". I suspect the sort of cookie-cutter sandbox paradigm showcased in GTA3 and its sequels is as adventurous as things will get for a while. As long as developers can stay one step ahead of the players "I wonder if I can pick up a whore... hey, I can! I wonder if I can shoot the tyres out now... hey, they added that! etc" then the illusion should suffice.

To be fair, most of the "game" in Elite happens in your head, prompted by fairly barren visual cues - it's a canny product. Frontier, by comparison, added just enough detail to the universe to bring into stark contrast how much freedom you didn't have and was, for my money, much less convincing as a whole. All of this stuff is a balancing act between the context you choose and the level of freedom you're realistically able to implement during development: That's GTA3's major success... you don't actually have that much freedom, but the developers spent so much time pre-empting everything you might try with the available props that you feel as if anything's possible.

Leaping from subject to subject, I agree that a common library of ready made objects would help. Indeed I was talking to Kuju a few years back and they were adamant that modelling houses dedicated to "prop" generation were the only viable long-term way forward. The trouble, of course, is that graphics sell games so everybody wants better looking crates than everyone else and the whole idea starts to fall apart. When every developer is owned by EA none of this will matter, though. :P

Then they'd be amateurs not indies. It's really not fair to lump them together.

As for Doom 3 ... I meant it was the most tangible looking. Everything had a lot of depth and it looked as if you could just reach out and touch it.

Half-Life 2 by comparison looked like a more typcial old school game ... nice images on flat paper. Doom 3 just looked a lot more 'real' whereas Half-Life 2 among others still looked like cartoons or something. It looked real in a few spots but if I wanted to show of graphics to someone I'd show them Doom 3, not Half-Life 2. The fidelity is just so much more in your face.

Comparing it to a plastic model is actually very cool because it's something you can touch and that's exactly how I felt about it. To me it didn't look too far off from CG rendered movies like Toy Story etc... Of course both still look like plastic models.

Doom3 sucked for me on PC. It ran fine, but the models were well bland, nice textures, but they looked like plastic dolls.

Doom3 Xbox was more fun for some reason.

Far Cry and Far Cry Instincts were great.

But now... F.E.A.R. is the greatest FPS for single player, havnt really tried multiplayer yet.

Games need to evolve more, im tired of oohhh look at our 4 pass rendering... who cares give me gameplay. the physics in HL2 were great, but over used, we need more gameplay choices.

HL2 physics were good, just not what I expected. It was either a) throw that saw blade at those zombies, b) throw that explosive barrel at those zombies, c) shoot them, d) avoid them

The actual physics was great, just didnt bring in enough game play options due to the mapping.

I know what your saying, doom 3 does look like you can touch it, but it still doesn't look as real as most other games mostly due to the lack of substance in the lighting. To me it feels like a film noire set in an animation studio. the problem with this kind of primitive shader technology we have now, compared to pre rendered stuff is that the more real something is, the more pronounced flaws are.

Mainly because were looking at things we know. Look at the criticism on TigerZ's realistic old man on another thread. It's a great exercise in realism, but anyone can be a critic because we look at people every day and arfe expert at spotting inacuracies.

As a game I don't like HL much either, but the lighting is far more realistic to doom 3's because it has depth and ambience reflected in shadow areas, rather than inked in black shadows, which for me is more cartoony. I know what your saying though, and despite the subject matter, the latest GF7 shader demo's show what we can expect in the future.

Fear despite being over dark, is a more enjoyable game as the enemy don't have pre defined paths and ovely simple behavioural triggers. I don't like the fact that most of the enemy are clones and all look alike. But it's nice to be chalenged by the AI.

Far Cry was great, right up until the point that the monsters took over being the enemy. After that I lost interest as I'm much more into sneaking around in the open and arranging plans. For me half of the fun is snealing around reconisance, planning actions and escape routes and luring those bad guy mercenaries to carefully laid out traps, controling the situation. If you mixed the best elements of Fear and Far Cry you would have the best FPS game ever.

Other than that, the most enjoyable singleplayer FPS for me was Rid@#!*, allthough once completed I get tired of his attitude the second time around, and don't feel inclined to play it through again.

My favourite game with physics in right now is age of empires 3 lol. It's pretty sweet seeing all those things get blown appart and I like the fact that the cannon balls have actual weight and bounce and hit things whilst in rickochet doing damage. The ship combat with bits falling off and splashing in water, plus the ragdoll like physics on the little people is cool too.

I say, bring on realtime radiosity. >_< Then the graphics nuts will stop crying for more, and they can start making games for gameplay again.

+BlackD

heh, I think that half the reasons that monsters are so popular in the latest gen games is in part because people lose some of the familiarity with the subject and don't notice the lack of true realism. Of course, to get true lifelike realism requires much more sophisticated animation methods than are typicaly used in games today. The way things move is as much to do with it as how they move.

To be honest though, I prefer a more stylized look, and where lighting goes, I think it will be next gen shaders that pave the real way. Not the immediate current tech.

things like subsurface scattering shaders that add depth to translucency like lighting your hand with a torch, and lit candles etc. things that add a little more subtlety will get rid of the plastic look. Which currently has as much to do with the crude lighting as it does the surface materials rendered at the moment.

It's cool that realtime graphics is catching up at a more accelerated rate, it wasn't that long ago that these effects were new for pre rendered graphics. I just hope that the realism fad goes away and people start trying to be more creative with it and realise more interesting ways to visualise things.

BTW I think HDR is going to have more of an impact with new games. thats part of what makes the Unreal 3 engine and other next gen game lighting look so good.

I think they're not evolving nor will they on present set-ups. Physics require a shed load of cpu time. But the next generation of cpu/gpu/ppu combos will result in graphics that A)don't look like moulded plastic and B) Physics that invoke the same feelings a movie does in terms of character integeration. We'll care alot more. They'll react alot more.
And we'll see games that utilise this and the massive increase in processing time offered by cell/cpu+ppu combos to create truely next gen works of art.
MGS4 trailer shows how great games will look next gen. Of course metal gear sotr of always sets the system benchmark for playstation so other games will no doubt look worse, but still as a sign of the type of games the next gen will bring, and to answer your question. No, we've only just begun to see what games have to offer.

The best thing that could possibly happen to promote games innovation right now is for manufacturers to stop devloping hardware. Then games designers would be forced to pull some good, old fashioned playability out of the bag to differentiate their offering.

I think I'm going to print this out and tack it on my wall.


That assumes both a brute-force approach to asset creation/utilization as well as branching-tree style non-linearity. The latter has long been dismissed as a bad approach (although ignorant game designers keep trying to do it,) while the former is a problem that the big companies are finally waking up to (or at least Will Wright; everybody applauded Spore, but nobody else seems to get it really.)

this will only happen if every engine out there follows the same standard so that this 3d doll you buy can work in simply by dragging and dropping it in 3d world. Before this standard of rules of worlds are set, this can't happen so easily...

I think this approach to game design is horrid. Part of the fun of video games is that you don't have to follow the worlds rules! Make something that acts wacky and doesn't make sense in the real world ... stop trying to replicate what we've already got!

I know quite a few people who are trying this approach of shared media with coded rules so that anyone can set up there own little virtual world and have it react correctly. For the most part I think this sucks because it means games are converging into one single boring experience.

Of course mixing some standard features with your different ones does make sense on some level ... but you know it's going to end up making even more clones than ever.

I don't think so. These are very basic rules.

Saying that a unified "world" would kill creativity is like saying that all paintings look the same, since they use the same kinds of molecules.

Your analogy is off. The essence of games and paintings is different. Paintings are great only because they look different and obviously as you state it doesn't matter that they use the same base paint.

With games we're talking about making them all play the same via a standard set of rules. The essence of games is how they play ... looks among other things are secondary to the mechanics. Standard mechanics make for boring ... or even identical games.

An accurate analogy would be all paintings share paint just as all games share programming languages.

Another would be different aesthetics make different paintings, while different mechanics make different games.

At least that's the way I see it. I think most people would agree. Even you said Quake 4 is the same game we've been playing for the past 10 years.

I would like to think that proper virual reality will be the next big breakthrough.
VR seems to have gone put of fashion though. Hopefully some big innvoation in this area will take place at some point

*whistles nonchalantly while going back to work*