Folio quality in actual games.
polycounter lvl 9
Some people will think this is one of the noobiest questions.
I see so much awesome art here, and I mean just so awesome, like Dominance War stuff and the like and I know it is for people's folios for game art and they make sure it's showcased as such, with game-structure and real-time renderers and engines like Unreal and Marmoset. But I'm curious, as I'm yet to see any games with quality like that. Good quality but never as awesome as, well what's showing in the banners right now.
I know it is kind of explained by all the other processes, like AI, collision, scripts, systems, engine constraints, LOD, memory allocation, etc etc. But I still wonder...
I see so much awesome art here, and I mean just so awesome, like Dominance War stuff and the like and I know it is for people's folios for game art and they make sure it's showcased as such, with game-structure and real-time renderers and engines like Unreal and Marmoset. But I'm curious, as I'm yet to see any games with quality like that. Good quality but never as awesome as, well what's showing in the banners right now.
I know it is kind of explained by all the other processes, like AI, collision, scripts, systems, engine constraints, LOD, memory allocation, etc etc. But I still wonder...
Replies
[ame="
[ame="
and some almost next gen stuff
[ame="
and rage
[ame="
thats pretty much why. With portfolio pieces you only have to worry about the graphics. If its still shots you could technically boost up the quality as much as you want. Like add higher quality shadows and take pictures at higher resolutions than normal and compress them etc.
Graphics are always the most intensive things in games but not having to have it be shown real time and not needing to deal with programming performance etc is the reason. Current games will never catch up with current portfolio pieces.
not every studio can hire the very best for every position.
add time and personnel constraints
add hardware / distribution constraints (e.g. we need to squeeze it all on 1 disc!)
not all people deliver constant quality
not all art directors care as much about quality as certain individuals do - i.e. you don't have to be happy with the asset, as long as the AD ok's it and the schedule (time = money) requires you to move on.
There's probably many more reasons...
Then it's a lot about optimization for each platform, things might run supersmooth on the X360 but run like crap on the PS3. Then it's down to figuring out why and if it's optimizations that can be done in the actual engine or if the art just doesn't balance the performance between the two.
Then it's also about the balance between game-design/level-design and level art/environment art. There is a lot of iterations going on , lots of people changing their mind about stuff, level layouts change making it so that you have to redo some work etc etc. Which could hurt the visual quality as well.
You'll end up in cases where you have to Art up something that has a really bad approach/framing/composition from the get go. Which will make it quite tricky to look good. Good teamwork and planning within the team helps both the gameplay and the visuals.
Creating art for just art is easy, combining gameplay and balancing everything for optimization is a bit trickier though.
Thanks. Informative post.
Actually we all should be thankful for that
In a game you often give the player the freedom to look at things from less optimal angles, it's often impossible to ensure perfect angles in games.
Just look at that Futuristic City and then maps like Founders Tower from Brink.
But yeah, back to topic!
Polycount = making every prop the best it can be
Studio work = making the environment look good as a whole, spending your time on "Hero pieces", such as large statues and fountains that players will see with ease, and might even stop to admire, as opposed to a fire hydrant that regular players dont ever look at. They run right past it. Plus, memory limits are a B*tch......
That being said, not ALL studios are like that, like the examples above. And with next gen, you`ll start seeing way higher quality from the studios that can afford it.
Exhibit A
[ame="
At around 0:45, all those particle looking things arent actually particles. They are geo...
Absolutely I respect that. Even more so I understand that there would have been a whole lot of studios screaming for their art to hit that zenith and then mid-ways through production realising they had to rescale it so all the tech fit together.
This will sound mean but designers don't just have to make stuff pretty. It's a lot more guesswork in terms of what the player will do with your game, not to mention that most people say "just make it fun" forgetting that "fun" is not a scientific constant.
For the past several years Epic has been packing features into UDK that 360's just can't handle, all of the work you see is being done on PC's that are able to handle so much more and displaying so much less. One character in a UDK scene isn't going to chug much of anything without all of the other stuff, especially on a PC that was built 6mo-2yrs ago.
But still that only limits developers to a certain degree, the games are still heavily influenced by art direction, design, time and talent. Most of what you see are labors of love and people are doing it because they are really crazy excited about the piece, in production that might not always be true. Often even if they wanted to they might not get the time to pour in that extra love that a personal project would get.
You are also looking at banners that where more or less, for lack of a better term, bullshots. Specifically created and tweaked for one or two angles using most of the resources in the engine to light one object.
The big hit to most games is multi-platform, often the bar gets set as low as the slowest hardware.
Consistency should also play a fair part, everything can't be a "hero piece" as said. Not to insinuate that level of quality over everything is unfeasible due to time constraints, but if everything was a unique point of detail things could get inconsistent very easily.
One of the first things I noticed about being a dev was that making a portfolio piece and making game art are completely different. Same workflow different situation.
I think the quality of indie games is definitely going to change industry standards tho, small studios are starting to find it easier to create near AAA quality, with those you know everyone on the team is doing it because they want to out of passion.
Where as not to say larger companies wouldn't have passionate people in them but out of 80-200+ people, to some that particular project may just be a paycheck.
But again its not just down to effort on peoples part, those constraints take alot out of the result.
In design there are so many more moving parts than in art. And any time you complicate the mechanics of something it gets harder and harder to work with, refine and keep maintained.
Often Designers are too close to the project and can do the most difficult tasks easily because they thought it up and have done it 400 times but the player won't have that expereince. They have to rely on other people to experience something for the first time and give them feedback, which normally gets lost in translation or focuses on things the designer can't really fix without effecting a TON of other stuff.
Trail and error and error and error and trail again and maybe success VS does this look better, yep.
What an artist tinkers with effects his/her work load its easy for them to weight the cost/benefit and work the weekend if they want with that work almost always being an improvement. For a designer one little change can effect the entire company and set the game back by months.
Games are situational and it often comes down to bunch of other factors like balance of weapons and classes, spells, the skill of the player, the design of the environment and the technical limitations.
If I had to compare the two jobs I would say, Artists are woodcarvers, Designers are engineers trying to build a car for the first time. There are just so many more things to think about when being an engineer where with woodcarving you know your tools, you know the medium and very little changes other than your skill which almost always gets better as you progress. With engineering you have to be a master at predicting the future and planning for hundreds maybe thousands of what if scenarios.
Reality is, Sci Fi and Fantasy are very 'niche', and don't have the same mass market appeal of more 'realistic' themed games, which is why only a handful of studios can even bother making those types of games in those genres to that level of expected quality.
Wow, that's a pretty rude comment that has no bearing on the topic at hand and managed to derail the thread.
-10 respect points
In my opinion artist motivation, most portfolio work is built to prove how awesome you can be.
Not every production artist comes to work each day to prove themselves, of course some do, but the industry isn't all polycount and rainbows
Production art is meant to support gameplay, often constraints in those categories make for a better game, but not as cool of a looking asset.
Additionally, artist's perception of what is cool, is often different from what the general population thinks or wants. Simply by coming to this forum you've proven your in a demographic that enjoys a certain quality of art, which a good portion of our audience does not care about. As such sometimes budgets reflect that and loops back around to time/money.
/2cents
Where as a personal piece is just art, nothing else, so like many of the banner shots, they are shown at the absolute best the art can be at without having to worry about anything else.
If you take individual assets from a game like characters and render them out alone to look there very best they rival the work that is up there on the banners. The banners are personal work that is shown off at its very best.
An example would be the characters Hanno made for Uncharted. When they are rendered alone to just show the highest quality they can be at without having to sacrifice anything you can see how they would be "banner" worthy if they were just personal art.
http://www.hannohagedorn.com/professional-work/uncharted-2-among-thieves/
So its not that the art for games look worse that personal art. Its just personal art is displayed at the highest quality that can be achieved without having to worry about any other aspect of game development. Personal art is just personal art, its not a game. To be a great game artist you have to understand that pretty art is not the only thing that makes a game great or even great game art.
Of course I forgot to factor in time as so many have mentioned, of course that would impact on quality.
Of course I know that, it's why I got into a mini-argument in this very thread about the capabilities of designers. But simply I was confused about why stuff game artists were showing to impress game studios and not necessarily movie studios was something that most game studios couldn't even show.
Sometimes it's the other way around though; the pressure from your coworkers' expectations from you makes you want to push the quality so that it matches the rest of the art. Many times you get some great concepts to work from and/or have art directors that help you with ideas to achieve a better quality. When you work on your personal art however, you might want to create something from your own imagination and what not and the lack of discipline as well as assistance makes your art look worse.
There are many factors that can affect your piece and personal art isn't always better looking than in-game stuff(not saying that any of you said this, just throwing my 2 cents in).
http://www.gamasutra.com/view/feature/173545/fun_is_boring.php?page=3
Fun is really difficult to define. And why should everything in a game be fun. I know quite a few games which have purposeful mechanics in them which arent fun but the games are better for them.
Games really are difficult to design. With art you at least have an end a clear aim. You know when you are there. The art is fixed.
With design you cant be sure of the end or the aim. The design has to be fluid.
I would say, think of it like a restaurant. The picture of the burger in the menu looks beautiful and carefully detailed. However, when you get yours on the plate, it could be close to that, or more realistically, it's probably half as pretty. It's just not realistic to take the same amount of time on your burger as they did on the burger for the menu.
People also like to 'blame' this issue on old tech, the 8 year old 360, etc. However, with enough time and care, that is not an issue. Diablo 3 is an example of this. There are hardly any normal maps or "next-gen" features. It is just good ole fashioned art direction and polish.
Basically, having 16 cores and 32 gigs of RAM is not going to fix your sloppy UVs and bad topology.
While I love Diablo 3, the game was in development for 10 years, thats a lot longer then the average 2 to 3 year dev cycle on most games.
Also with Diablo 3 you have a fixed Iso camera, you cant run 360 around objects so its easier to hide seams. You just put them in them on the opposite side that the camera is facing.
No, I meant it as an example of a game where the quality of the art in the game was equal to or better than the "portfolio quality" the OP was referring to.
Sure, there will always be seams, UV issues, etc., and their respective ways that they are dealt with. The issue is, why are most games not up to par in those areas like a respected portfolio piece would be? And it basically comes down to time = money, as others have mentioned here.
Typically a game in development is 100s of gigs of data, thousands and thousands of assets, and that has to be boiled down to a 4GB disk or even smaller download or streamed packets or whatever. Most companies can't take years to really finesse every asset so a lot is done quickly and procedurally. Once a game's content is complete and everything is finally view-able together (skinned and animated characters, environment art, interactive physics, vfx, destructable objects, lighting, etc) a combination of tech artist(s) and engineers will look at what is making the biggest performance and frame-rate hits. You name it they will cut it or scale it down. This can include but is not limited to:
-removing draw calls from shaders (removing shader functionality, maps, features and so on)
-decimate geometry/LODs
-Compress and/or resize textures
-simplify collision
-remove lighting/shadow influences
-reduce/simplify physics influences
-dynamic lighting
The list really goes on and on. A lot of this can wind up being done procedurally by scripts so you wind up with verts and UVs getting collapsed or deleted that probably should not have been collapsed or deleted. Ideally these cases are caught and fixed but sometimes slip through.
There is also the likelihood that a game you are looking at is a port from another platform, often ported by another studio altogether, all of which will likely be procedural optimization by a handful of people.
In short, what you see in people's portfolio is what they created and hoped to see on-screen in an ideal world. Even in something like marmoset toolbag you are free from constraints of frame-rate, lighting, draw calls and so-on. You can turn on all the bells and whistles and see your work at it's best. It's a blow to the gut to most people to see what happens to their work further down the pipe.