Videogame Graphic Advances - Not What They Used To Be?

Thanks to GamesRadar for its PC Gamer-reprinted article discussing why graphics alone aren't enough to sell a game anymore. The author explains: "During the final days of Steam, I found myself playing the original Half-Life. And, frankly, it looked perfectly acceptable. While it clearly lacks the fine polish of modern first-person shooters, the world it presented me with was entirely comparable with anything around. And, being a great game in the first place, it was more enjoyable than - say - Unreal II." He continues: "However, if you went back to 1998 when Valve's masterpiece was released, and attempted to play a game five years older than that, it would be a very different experience. To go back and play System Shock, Doom or Wolfenstein requires a whole re-arrangement of your thought processes to accept the difference in graphics quality." Do you agree that "...the days when graphics ruled videogames are rapidly drawing to a close"?

Comments

but then they'd have to rely on content/gameplay

[Anonymous Coward]

which is very very hard for most game companies.

Re:but then they'd have to rely on content/gamepla

[El_Muerte_TDS]

people want as much eyecandy as possible before the game ships.
but they want gameplay when they bought the game.

Good Riddance

[tid242]

Good riddance, i'm sick of the plethora of shitty games that survive solely because they have 'good graphics.'

Maybe now people will actually develop good games instead of their own graphical egos.

-tid242

All about the hook

[GTarrant]

I remember getting my first 3D card and going from playing, say, Quake, to playing GLQuake, and practically falling off my chair with how GL looked compared to software.

But did Call of Duty do that to me, compared to some FPS of two-three years ago? Not because of graphics.

But this happens all the time. You need a hook, to sell. Graphics aren't the big thing now. But back when the PlayStation came out, or when 3D cards were becoming more common in PCs, did you get something that advertised in big letters "3D!" on it. A developer would take anything, stick it in the box, and if it was 3D, it was 'cool' and people actually bought it, even if it was absolute crap. Games that were good, and 2D, didn't sell, and games that were lousy, but 3D, sold. Go back and read some game reviews from the period, and you see all sorts of reviews like "This was a great game, but with the '3D revolution' we're in now, it just doesn't cut it." Then a crappy 3D game gets a 8/10 because it's 3D. It's a hook. They're always looking for a hook.

Graphics aren't a hook anymore. How often now do you look at screenshots on a box and go "Wow"? Not nearly as often. So they find a different one. If I had to pick one, I'd say right now it's "Online play!" Games with online play mention it about 14 times all over the box. Great games get some crappy netcode slapped onto them just so they can be "online!" Otherwise good games get hurt in reviews, even if they're single-player titles, because they don't have online play.

What will the next hook be, when almost everything's online and "it's online!" is no longer something that reviewers will give bonus points for? That's the real question.

Re:All about the hook

[neostorm]

The future "hook" is likely total immersion. When I think of something that will blow me away today, on the same level as, say, the GLQuake example blew us away when we first began utilizing 3D cards, it's total immersion.
Graphics today have a long way to go before they peak, there are still many things we can do that we just haven't had the time or power to do yet. However I don't forsee any of these things being revolutionary on that level until we are *in* the game.

Now I feel like I'm in the mid-90's saying "Virtual Reality is the Next-Big-Thing" all over again, but I think that was the right attitude all along, just far, far too early to be realized.

Re:All about the hook

[aanand]

The new hook? It's physics, man. Physics is the new cel-shading.

Back on topic, though: I think one pivotal reason in the levelling-off of graphical advances is the sheer unfeasibility of building such densely modelled environments. How long does it take to model a skyscraper right down to the screw threads?

Re:All about the hook

[kisrael]

Real physics for every damn thing. People that don't walk a scripted motion captured "walk" animation, but because of the the way their muscles and mass are working together. Rag doll physics is just the begining.

Gah, I hope so...

[Gleng]

I would trade in 100 games with good graphics for one game with great playability.

Some of my favourite games over the last five years or so have been things like Baldur's Gate 2, Civ 3 and Sim City 4. None of those can claim to have great flashy graphics (although the artwork in BG2 is fantastic), but they offer an unparalleled level of depth and gameplay.

I'm certainly looking forward to whatever the "new Black Isle Studios", Obsidian Entertainment can come up with.

None of today's games have "good" graphics

[Kris_J]

The games of yesteryear that were considered to have outstanding graphics really pushed the hardware, I mean that they required real talent in programming and required every CPU cycle you could devote to them. None of today's games really push the envelope, just your wallet. Wake me when someone writes a game in assembler and it still requires a 2GHz machine and the latest DX9 video card.

Re:None of today's games have "good" graphics

[P-Nuts]

I guess you consider '$' as a good representation of a monster?

But '$' would be an an awful representation of a monster. Monsters should be shown as alphabetic characters.

Re:None of today's games have "good" graphics

[GTarrant]

There's also the fact that there was a point where if you had a 3D card, the developers pretty much KNEW which one you had. When GLquake came out, everyone pretty much had a 3dfx-brand card, if they had one at all. Then it's easy.

Now, everyone has something different, and all the cards have different capabilities. Worse yet, some manufacturers are shipping their newest computers with some of the oldest 3D cards (Sony, for example, ships some otherwise top-of-the-line systems with TNT2 cards). Some manufacturers (especially the video card Big Two) pack a lot of neat features into their cards that allow developers to push the limits, others don't.

As someone who has worked in gaming tech support, you have no idea how often you get a question like "I have this BRAND NEW SYSTEM and it came with a TNT2 and my game doesn't work." And you say "I'm sorry, sir, but the requirements listed on the box state you need an X card or better - it even names manufacturers and chipsets." And they say "But this computer is BRAND NEW, I'm never buying your games again!"

Before, a game could say "3d accellerator required" or something, and you'd get _fewer_ people buying it that didn't have one. Now, EVERYONE has a 3d card - but when the devs try to push the envelope, some cards get left behind.

"Surface" technical feasability is here

[ReyTFox]

But there's plenty of things one could add to a game that can't be done quite yet; some of them are programming questions like AI or systems modelling; others simply require more hardware, like modelling worlds with totally deformable terrain(Red Faction, if you even remember it now, felt like a hack because only certain parts could be destroyed) or bringing the detail of our CG characters and environments in line with pre-rendered work.

The real issue is how we, so to speak, "start over" now that games can do everything we can imagine, when a big enough budget and schedule is allowed. Lots of people want to do virtual realities, in an online or single player form, and over time the distinction between that(when it was still theoretical) and a game that constantly tests our abilities or acts as social glue in the way that sports or board or card games might has gotten muddied; games today are often made heavy and slow-paced by feeling the compulsion to satisfy both the requirements of skill/excitement and of VR. Games rarely ever have a continuous stream of challenge thrown at you anymore; instead, it's broken up into little chunks wherin you explore a little, and then you fight or solve puzzles or whatever, and then you go back to exploring...

That said, I have great hopes that the market will reinvigorate itself with a whole new set of ideas; there's plenty of untapped potential floating around that is likely to unleash great stuff over the next few years, games that try to do things "new and different" like any art should.

Re:Funny that this should be posted tonight

[bugbread]

No, you aren't the only one on the planet. However, a lot of people like it too.

It's like chocolate cake. Chocolate cake has been done to death. Everyone's eaten it before; every variation has been done. That doesn't keep good chocolate cake from tasting good (use some other food example if you hate chocolate cake). Sure, chefs are experimenting with new non-chocolate cakes, and making almond crumble chiffon turnovers and other brand-new pastries, but that doesn't mean that chocolate cake isn't still good. Halo isn't popular because it is original, or innovative, or all of the other catchwords that get bandied around. It has a lot of fans because...it's fun! It may not be your cup of tea, but like it or not, it's popular because people enjoy playing it.

People need to move away, not only from the idea that "Good Graphics = Good Game", but that "Innovation = Good Game", or "Realism = Good Game", or "Good Storyline = Good Game", and remember that the key is "Fun = Good Game". If good graphics, story, ideas, originality, etc. help make a game fun, then that's an added bonus, but even a game with trite, rehashed ideas, bad graphics, and a laughable story is awesome if it's fun.

And if you haven't played Halo multiplayer, or you've only played on PC, you missed all the fun bits, so I wouldn't be surprised if you find it un-fun as well. The XBox multiplayer part is where it really shined.

Photo-realism

[limekiller4]

Game graphics are asymptotically approaching photo-realism.

Is it slowing? Yeah, because as you get closer to this holy grail, you spend more and more time/years getting less and less return for the effort. But are all the nails in this coffin? No, not even close. What we have now looks good but isn't going to convince anyone they're looking at footage.

One More Graphics Advance

[Babbster]

There's one more graphics advance that will stun people and that's when material/environmental science is applied to every set of polygons in a game. For example, not only will you see a newspaper blowing along the post-apocalyptic street you're walking down in the latest FPS, but the newspaper will realistically alter in shape according to how the wind blows and how it hits the ground or other objects. Instead of blowing up a brick wall, breaking it into predetermined bits; the brick wall will break apart differently - dictated by realistic physics - depending on what was used to blow it up, the size of the bricks used in construction and the mortar holding it together.

I consider this an advance in graphics, in addition to an advance in gameplay and game physics, because it enhances the visual realism of the gaming experience. Shooting a chandelier and making it fall onto enemies is cool. Shooting down that same chandelier and watching it hit the enemy and ground, breaking into realistic pieces flying in realistic patterns would be awesome.

So many times I've been wrong...

[BTWR]

OK, I always knew Atari, even in its day, had awful graphics. I remember playing the NES game "Total Recall" and thinking (as an 8 year-old) "wow, what a crappy attempt to recreate Arnold."

Ninja Gaiden, the cut scenes at least, had "awesome" graphics at the time but I knew they were just really good cartoons. But when I played Wheel of Fortune on the SNES one time, and it had a near-photo quality still-picture of Vanna White on the title screen, I thought that "well, this is the best it can be... because you can't do better than photos!"

Flash forward to 1997 when I first saw Mario 64. I walked into my friend's house and I seriously did not even take off my jacket. I was standing there for like 10 minutes just marveling it. Amazing, I thought. They did it. They peaked. Can't get any better.

Once again, I was wrong. Super Mario Sunshine is much prettier.

Stupid me, here I go again. Just 5 minutes ago I finished watching a preview for EA mvp baseball 2004 for gamecube and I thought again "Wow, this is a looooong way from Bases Loaded on NES. look how awesome this looks! Seriously, how get much better than this!?!?"

Something tells me that I'll laugh at that statement once again in 2008...

Re:Funny that this should be posted tonight

[a whoabot]

I agree. Fun is important. And it may, or may not, be solely what makes a game good, but it's obviously not what makes games sold and used solely. Unlike, a chocolate cake, a game can be "enjoyed" over and over again. They're not selling very many copies of Super Mario Bros. 3 anymore, despite the fact that it was, and still is, fun. Actually, they sell so few, that Nintendo doesn't even sell it anymore. They sell new games as "replacements" for the old games. These new games may be fun, but, are they anymore, purely, fun than games of past? This is doubtful, no doubt.

Then why do people still buy, and enjoy, new games, if there are already thousands of extremely purely fun games out? There's obviously more to games then just pure gameplay fun. New stories, new graphics and who knows what else, also effect what games people get and use.

So yes, perhaps if you define "good" as "pure fun", then, tautologically, you're right, "fun = good game", but if you consider what people buy and use in the definition of fun, then there must be more to it than "pure fun".

spatial immersion is more key

[ghostlibrary]

"a whole re-arrangement of your thought processes to accept the difference in graphics quality."

I disagree. Doom is still quite playable. People don't 'rearrange their thought processes' to play cartoonish games like Jet Grind Radio, lower graphics quality doesn't require a shift. What does require a shift is 'how do they model the space'.

I think 'Doom' really nailed the spatial immersion aspect (and is still playable now). Quake et al added full 3D movement. It wasn't just the graphics, but the fact that Game Movement was like Real Life Movement.

So it's sort of a tactile thing. Once you were walking seamlessly (not in chunky steps), and could look around, things had 'arrived'. After that, things just got prettier.

And, they got the audio right-- you got spatial information from where the sound came from. (5.1 stuff has really helped boost that, but I can't pick 1 'pivotal' game that advanced it.)

So I think the next big leap isn't going to be graphical, but spatial. Perhaps handling peripherial vision, so you don't get the 'someone is hitting me but where?' effect, and there's more of a sense of placement.

Or some clever way to handle mapping and direction so you don't feel lost-- one can get lost in an FPS mall due to lack of spatial awareness, whereas it's harder to do in real life.

Or perhaps kinetic sense will be the next thing, actually feeling motion. We'll see.

Uncanny Chasm

[Thedalek]

Sometime in the last six months, Popular Science did an article on a robotics hobbyist whose quest was to create an android head which mimicked the movement and form of a human head completely.

Over the course of the article, they discussed something called the "Uncanny Chasm." This chasm was what happened at a point just shy of total realism, at which things look jarring, unnatural, and disturbing.

This is part of what's happening with games right now. We've reached the cusp of the Uncanny Chasm. Some have marched headlong off into the pit: I can't count the number of sports games I've looked at and thought "Wow, that looks totally incredib... Woah, that looked completely wrong."

SquareEnix and Konami have pushed further towards the far edge of the Chasm, but only in cutscenes. The primary reason is, once the character is under the player's control, it is virtually impossible to keep up the convincing level of motion and still have the player be able to control more than just a modern-day Dragon's Lair.

The answer is a resounding NO.

[__aailob1448]

While It is true that graphics have remained a bit stagnant for the last 5 years or so, I believe this is just a phase. Graphics still have a very long way to go before they become truly photorealistic. Toy Story came out in 1994 and 10 years later, a standalone PC still has a long way to go before being able to render it in real time. And toy story can hardly be called photorealistic. Similarly, today's best techniques and the biggest rendering farms can give us something like The fight scene between Neo and the 100 agent smiths in Matrix Reloaded. Barring some revolutionary breakthrough, A standalone PC won't be able to render something like that for another 20 or 25 years at least. And that scene was not quite photorealistic either. Add to that the fact that we need to progress towards a resolution of 4196*2360 and 100FPS as a minimum and my estimate would be 50-100 years before we can look at something that can fool our eyes completely.

Games aren't entirely focused on one thing

[Peter Cooper]

The main story here assumes that pure gameplay is everything in a game. Sure, gameplay is important, but our standards for other things go up over time too.

I agree with what the guy says about Half-Life, and that it's still more playable than many games today.. but that's because today's games don't have as good gameplay as Half-Life. If a modern game with excellent graphics had the same game-play than Half-Life, then it'd be better. PC games have just tended to suck over the past year.

But, no, we haven't gone far enough yet. When you can render something and it looks just like you're 'really there' (i.e. photo quality), with no lines between texture changes, and the like, then we'll be there. Of course, we'll also want excellent AI, and excellent scaling. I mean.. who wants something that looks like real life (Max Payne and Half Life 2 come surprisingly close here!) but which forces you to take a very defined route to the end?

It's all got to scale. Not just the gameplay, not just the graphics, not just the sound.. but everything.

Re:Yes

[Anonymous Coward]

The happening thing in video game graphics seems to be a layer of special effects that get applied on top of the normal 3D output, and there's still a lot of growth possible there. The motion blur in need for speed underground, the haze effects in prince of persia and the post-getting hit by a mortar shellshock thing in Call of Duty are some recent examples of this.

In a way, this is comparable to what happened to movies. Once the transition from black/white to colour was over, realism was no longer a big issue, and it was all about style.

Graphics are always going to be very important to games - but I predict a paradigm shift away from realism, and towards style.