…Must…Use…Available…Polygons…

Ha, way to go making me think that the UT2003 Rocket launcher looks muddy and under-detailed now .

I think this boils down to aesthetic taste in the end. I personnaly like the newer gun models for UT2007. Unreal has always been about ridiculous excess. More polys, more bloody chunks, more explosions, bigger textures, more guns, more firemodes. After seeing the new shock rifle in action I gotta say, its pretty cool.

It also goes worth saying that UT2007 is as much of a game as it is a tech demo. Epic one ups id in actually having a pretty awesome game to flagship their graphics engine. But UT2007 has to display as much about the incredible Unreal 3 engine as it can. Has it worked? Check out http://www.epicgames.com/ and see just how many people have licensed it already.

Damn that rainbow six screen looks hot by the way. Is that Unreal Engine 3 as well?

Yes, reading the post last night (but not commenting as I went to bed), the exact same thought as Holliday’s came to my mind - that UT games have always been not only over-the-top romps with big guns, but demonstrations of what their engines can do for potential licensees. That of course made me wonder why id didn’t go that same route with Doom 3 (just all dark corridors), but I guess they were more concerned with keeping up with tradition, seeing as they already knew everyone would think their engine looked good.

I have thought many times before though - once graphics are photorealistic, full transparent anti-aliasing is as standard as dynamic T & L, and games really do look like playable movies, what are the graphics card companies gonna do to convince people to upgrade? And what are the engine developers gonna do to make new games that look… more real than their already completely-real-looking engines?

If they are going to spoil us with all these superfluous graphical goodies now like we see in UT2007, I can’t really think how I’m going to be amazed once it all plateaus. Seems to me they’ve already pulled all the tricks out of their sleeves. I mean, yeah, they can increase the scale dramatically and improve the physics… but that’s already being done.

For that matter, isn’t it true the more “realistic” a game is the more drastic a comparison we make between it and the physical world in our mind? The result being it pales *all the more* in the face of the physical world? Gran Turismo is held up much closer to reality than, say, the exaggerated F-Zero X (despite the crafts behaving as they “ought”).

With all the wondrous graphical advances being made on a year to year basis, I too wish developers would cease being hung up on the latest shadow effect or, to sum up the article, piling up polygons for polygons’ sake. Imagine how this creative technology could be invested in crafting artistry — this last aspect having the potential to be more effective the less “realistic” it is.

As I see it, art follows the tech push. With each jump you get a huge push towards hyper-realistic graphics and ridiculously detailed effects. Then, as the technology used to make these visualizations gets widely understood and used, the art really starts to blossom out of it. Shadow of the Colossus could not have been possible without many less-than-inspired games before it.

Its a cycle. It always goes on and I remember countless times saying “How much further do we really have to go?”. But each time the new games make the old ones look so primative. Then the artistry gets even more choices and ways to display itself.

In Reply to #3: You know, it’s funny that you should mention these things. I seem to recall sometime last year watching a documentary on art in the ancient world (No! Don’t stop reading now, I have a point here somewhere…). The gist of the episode was that the Greek sculptors spent years trying to carve the perfect, anatomically correct, human form believing that being able to do this would produce the perfect sculpture. The conclusion, however, was that they were wrong. At some point the Greeks achieved their idea of perfection, discovered that it wasn’t “pleasing” and they went back to exaggerating/misrepresenting the human form because they actually preferred this more.

Given the above, perhaps it is reasonable to infer that the achievement of photorealism in games will not achieve graphical perfection. It could be, for example, that what’s available today is as good as it is going to get, or even that what was produced before was more pleasing. My position remains that I prefer the look of weapons in the previous UT.

Anyway, something to ponder.

I think you are being overly worried here.

I personally like having good graphics, and whilst some games may go OTT, this may be just part of the game as has been mentioned. Even if you don’t really have time to appreciate the finer nuiances, they look good as an overall package and make the game look better as a whole. There will always be developers that keep things ‘real’ in war games for example.

Then there will also be the games that have basic graphics despite the potential to do better, sheerly because the game looks best with simple graphics (I’m thinking mainly about cartoony games here, like Katamari Demarcy).

Good article. And seriously, I know Epic make Unreal to flesh out their engine and pimp their wares… but that rocket launcher is fucking ridiculous. I thought it was a starship or something when I first saw a pic for the game. Way over the top.

I agree with you, all in all. But then, I’m not really the market for this sort of thing. I don’t upgrade parts regularly, if at all. I don’t stay up to date with all the latest news. If I were to play a good game, that looks nice, then great. The rest is bullshit that keeps it from running properly on my PC.

Normally I’d chime right in and tell you how alarmist you’re being but lately I’m not so sure. What bothers me about this is the rising costs that are accompanying this. Both in terms of from a development standpoint–ie you now need to hire x number more artists than you did before. And from a consumer standpoint, you need either an HD TV and pricey console or a pricey high end PC to play these games really exceptionally. And the fact of the matter is sales increases are not matching cost increases.

In Reply to #8: It’s definitely a reasonable point. While it would appear that the Unreal Engine 3 will take some of the pain out of creating (relatively) low-poly versions of complex models, I’m sure that it takes longer to create a complex model than it does a simple one and therefore you are quite right that the costs associated with such development must increase and, mostly likely, the time required as well. I remember knocking out a few levels in the original Unreal Tournament and it was pretty easy to do (although not as easy as Doom was, mind) but that it was much more complex in UT2003, assuming that you were going to be adding meshes created in Maya. Complexity almost always equals additional cost.

Given this I am quite in agreement that costs are increasing but that sales aren’t. I can’t back this up with any statistics but I can say that I spend way less on games these days than I did before and a lot of this is because my current systems can’t run the newest games and developers/publishers have abandoned the older systems. Take for instance the news that Starcraft: Ghost has been put on hold so that it can be released for the new consoles. Blizzard would have got my money for a Gamecube version (which was canned ages ago anyway) or even the PS2 version but I’m not planning on buying any of the new systems so that’s £40, or whatever the cost would have been, of my money that they won’t now see. Am I the only one who represents a lost sale for Blizzard here? Perhaps…

In Reply to #9: Starcraft: Ghost was one of the reasons I bought a Gamecube in the first place. Which is currently a great source of annoyance to me. And no, I’m not going to buy a next-generation X-Box or PS for it either. I would have gotten it if Blizzard had pulled through on a GC release, but right now I’m joining you at the ‘lost sale for Blizzard’ line.

In Reply to #9:

I too, was planning on getting StarCraft: Ghost for the GameCube. The GameCube turned out to be my favorite console of this past generation (though the DS wins if we include portables) and I was quite disheartened to learn the GC version was cancelled.

I’m not sure if I’ll be getting a next-gen console any time soon, but I will most likely NOT be getting StarCraft: Ghost when it is released, mostly because of decisions like these.

Oh, and for a development cycle approaching Duke Nuke’em Forever lengths, the game better be freakin’ awesome, regardless of me buying it or not.

In Reply to #11: I’m definitely echoing the Gamecube sentiments here, and I thought it was going to be a load of rubbish when it was first announced. For a console that really was rather good I’m almost surprised that its support disappeared faster than the Italian army. Then again, my console before it was the Dreamcast and the one before that an N64, so my ability to pick winning consoles is not exactly great…

With regards to Starcraft: Ghost, considering the money that must have been spent on it and time invested, you’ve got to figure that the game must be an absolute stinker if Blizzard isn’t prepared to release it rather than essentially rewrite it for the next-gen machines. The demonstrations shown at E3 and the like looked great so I can only assume that this was not the “real” game that we were seeing (sort of like how that E3 demo of Halo 2 that blew everyone away wasn’t in the game) else you’d have to figure that it would be complete by now such that investment could be recouped.

I am uncertain if Blizzard should even bother trying to develop for consoles. Their development cycle is the entire length of a console’s lifetime (more now that Microsoft has shortened the time). And while when they work for PC’s they can get by with adding some new effects that have been discovered recently to keep the game looking somewhat current (shaders etc.) on consoles it usually just is not possible.

In Reply To +Expense +Time:

I think the industry is going through a transition period right now. They are finding new ways to deal with these things though and I think as they get more of a foothold in this area we’ll see costs level out and the development time shorten again. Look at some of the design techniques for Oblivion. 16 square kilometers of land most of which is covered by forests. Take a look at the trees in these forests and tell me if you can find 2 of the same.

That alone is a daunting task but then throw in the complex radiant AI, making gameplay systems for almost every play style imaginable, countless quests, huge score, massive voice cast etc. and its hard to think how they possibly made this game.

Well if you watch the lovely “Making of” DVD that comes with the collectors edition you find out. Instead of having the artists hand model every square inch of those amazingly detailed forests (they truly are), they developed a software tool that uses “complex forest growth algorithms” ($0.25 phrase of the week) to generate realistic forest ecosystems all with the computer. Then the artists go in and fine tune different elements (adding a path or a stream, swapping a few trees around, adding dungeons and ruins). Now I am no expert but i’ve taken a biology course or two and when looking at the way the lands are set up it really does follow the natural order of things.

Not only that but all the folliage and plantlife that the program uses to grow these forests are made in a similiar way. They have a “tree builder” program which allows artists to create a vast array of different trees with a very simple interface but amazingly life-life results. I don’t see why this kind of software can be applied to other games and we have a sort of “next gen teamwork” going with devs all learning off each other.

In Reply to #12:

Actually, I kind of hated the GameCube when it was first released. I don’t quite remember when it hit its’ stride for me, probably around the time F-Zero GX was released. I figured the PS2 was going to take the cake, though I didn’t expect the XBox to make such a strong showing.

In Reply to +Cost

Apparently Bethesda has figured out a way of recouping their money on the development of Oblivion: Charging for downloads. MadMonk has a thread on GameSpy about it, though I constantly screw up the html tags and I won’t be linking it.

I thought Microsoft was going to limit its “micro-transaction” policy to the XBox360. Apparently at least one developer has decided its a great idea and are charging two bucks for armor skins for the horse, amongst other things.

Doesn’t seem like a great value to me when sixty bucks gets you all the textures in the game.

I personally look forwards to the day when, like the greeks, realism is achieved. Then people will have to go in other directions to set their games apart.

I was thinking just yesterday how smart it is for the DS Zelda to use the Windwaker art style, because if they’d aimed for the more realistic look of Twilight Princess, the disparity would be so much greater than it is between Hourglass and Windwaker. Yes, you can see that there are less polygons on link in Phantom Hourglass, but the impact of that is reduced by the art style.

What would you do with a game like Katamari Damacy? It doesn’t need more polygons as the art style is wonderful in all it’s boxy glory.

One thing that’s true about ‘realistic’ looking games, is that generally they date a lot faster than stuff that takes more of an artistic approach.

They have a “tree builder” program which allows artists to create a vast array of different trees with a very simple interface but amazingly life-life results. I don’t see why this kind of software can be applied to other games and we have a sort of “next gen teamwork” going with devs all learning off each other.

Actually Bethesda licensed the SpeedTree tech to do this. It really can be applied to other games; in fact, I’ve got it in 3ds max myself. The way of the future is third party plugins coming together in the framework the developer has created so they can focus on the original assets they need.

Re: the topic. This emphasis on crazy amounts of detail is just a fad. I remember seeing screenshots when bump mapping first starting coming into realtime engines, and everything looked bumpy as hell. When normal mapping stops being the new thing, developers will put it back into the kit with the rest of their tools and employ it sensibly. Then some other new tech will come out and we’ll complain about how it’s overused.

Yeah alright, your blockquote tag is fucked. I was quoting Holliday’s bit about the trees in Oblivion.

I think I edited it to how you planned, [P]. The “cite” attribute is just who you are quoting. Not the quote itself. That goes in between the blockquote start and end tags.

From my perspective I can see both good and bad in this topic.

On the good side, having access to higher performing hardware and game engines must make developing in-game models much simpler. While polygon limits still exist the constriction that they apply on creativity must surely be reduced these days such that it is simpler to add to the game what you had in mind to begin with; detail did not have to shaved off to reduce the poly count. This, I presume, should save some time in development since reducing detail has got to be problematic (incidentally, where’s Blade when we need him?) and time consuming.

On the bad side, however, there’s the potential to go nuts and just keep adding detail when it’s not necessary. As noted in the article itself, what’s the point of creating highly details character models in a game that’s designed to run at breakneck speed and consequently no one will get the opportunity to see the detail? In this respect the time required to add this detail is surely time added to the duration of the project and (slipping into software project management speak) simply represents “gold plating”. When this happens you are paying more money in labour and not receiving back any return such that you are harming the financial viability of the project.

Graphics in a game are important. That we can create better looking games today is great but I am convinced that there has to come a point where the line is drawn and someone has the sense to say “Gee, this looks great and all, but is it really necessary?” With development costs continuing to rise to stupid amounts of money and development studios closing it seems ridiculous to spend money on “gold plating” in an industry that simply cannot afford to do so. Oddly, I currently blame the likes of Microsoft and Sony. Their continued desire to dominate the market by making the most powerful consoles means that the time/money required to develop a next-gen title increases but the quality of what is delivered is not increasing as fast. I don’t believe that this is a good thing…

i have to say that there is nothing wrong with increasing the levels of detail textures etc in a game it is all eye candy and in the case of gaming on ps’s you always have the option to reduce detail hence you dont need to worry about over kill.
i think that iv reached the end of my gaming streak as when i see games today graphics are one of the last things i take into consideration (hance i ordered a ds lite). for me a game is about its storyline and how you play the game the places you visit and the characters you meet-nowadays a game is mainly about how shiny are the bolts on your gun or how high is the resolution on the shadows or how well the textures are on the foot thats about to kick your player in the arse. i find this well annoying to say the least
i mean dont get me wrong iv got a 7800gt card and new games always intrigue me into how they are going to make it look good but deep down the novelty faids quickly as looking at something doesnt interest me i want to interact with it in a realistic way. because of this i say yes games should improve on graphics as they are esentially upgrades which are never bad but i say that we should stop thinking about how well an environment looks but concentrate more on how you play the game its settings and levels of imagination-by the way i find ut does this just read the stories to the arenas you are going to play in deserter space stations etc i want to see more
if theres a games developer reading this please make a game that has some though into it rather than just how well its going to look…please