Hey Peeps, I was discussing this with my friend as he's studying game art and design right now and we had a very interesting discussion on something we noticed with past gen games and the latest entries.
This is something that most AAA titles tend to do and both of us agreed it was better before.
It seems like like nowadays, levels and game environments are built with too much focus on the aesthetic and low attention on the ''fun factor''.
''They're nice to look at but not fun to play in''.
Both of us felt like level design kind of went downhill these past years.
In my (ignorant) head, whenever developers wanna build a level, they take the ''Game Design'' and Gameplay logic first, they take the concept art second and then hand it to the level designer and say
''make something fun out of this''.
The level designer then take the artwork and modifies it, makes a layout out of it and start figuring out how they can make this ''concept art'' fun. For me, this looks like a nice way to start designing a level and when I played previous games, I felt like they were built that way. I quickly figured out why a lot of environments in past games never looked like the artwork. Of course, there were technical constraints but I thought to myself that because the drawing was nice to look at didn't mean it would be fun playing in it.
Then logically, if you're developing an FPS like Counter Strike for example, you wanna have multiples elements and factors in your game that fits the need of the game design, not the art! You may be the best concept artist out there, even if you would paint a ''desert level'' with beautiful sand dunes and an amazing lighting, nobody would play in it..
For example, I'm going to take this random artwork (I haven't played the game, I'm just using this as an example).
Let's analyze this picture. More and more, I'm kinda amazed at how artists manage to recreate and keep the mood of artworks into game, which is good. The problem is; in terms of fun factor, I have hard time seeing what the level designer did and/or would pull out of this.
Ok yeh, you put some super beautiful pile of rocks here and a super pretty lamp post there, but why are they here for? What's their purpose? Can I hide behind them? Can I pick them up? Or it's all just eye candy?
Imagine there's a gunfight happening here, how would you modify this picture to make it fun to play in?
Honestly, there's a bunch of stuff I would change in this image if a gunfight situation would appear in this sequence. I would add some assets, remove others, add more elevation variation and etc.
See that's the thing, the more AAA games I play, the more I have the feeling that they directly take the concept art and mold it straight the way they are into the game without modifying them...
This is something that is getting prevalent especially in Open world games. Open world now are not a always a good thing, especially when you see that half of the environment is made of recycled assets and empty places void of any ''level design '' value...
You're basically adding water to you beer to fill the bottle...
Sorry if I look like an ignorant on the matter, I just wanted to discuss this with you guys...
Replies
Put that sort of thing in a game these days and people would complain about getting lost and how arbitrarily the keyhunts pad the game play. Would they be right? Kind of. I think the best place for single-player FPS level design is somewhere inbetween the Dark Forces levels and the Medal of Honor: Warfighter levels. Levels that are going to be played rather than looked at should always, always go through a greybox or CS1.6 stage to make sure that they're interesting to play through, well-paced, and easily navigable even without all the bells and whistles.
The designers don't use traditional concept art. Usually they graybox things directly in the game, or via something like Sketchup. Some designers like the oldschool grid paper, or using a flow chart.
If you're an environment artist, you're usually working closely with designers on the flow of the level. You either get a graybox with some ideas about where things need to be obvious for the player (red light shining on a door, etc.). Or you get a mostly fleshed out level using modular pieces, but it's unlit and un-set-dressed, so you're there to prettify it without destroying the flow.
At least that's how it's gone for me.
We have some stuff on level design here that might help
http://wiki.polycount.com/wiki/Level_Design
I think the true problem is the constant push for higher resolutions and greater fidelity without understanding that it lessens what you can do with things at hand. You can make things pretty, but you then have to simplify enemy encounters and lessen the numbers at hand, but it also affects the level design because of that.
I often find that games like Duke Nukem 3d, Quake 1 and 2, Blood, and even Unreal to an extent put a larger emphasis on action and interaction- cause and effect, which the player is more directly affected by and has to confront or be an active participant in. Games these days put alot more emphasis on showing you a thing happened rather than having you directly affect. Call of Duty Ghosts is a pretty big offender in this, where they constantly intersperse things that go against the general gameplay to show you something pretty. Or worse, pull you into a minigame to fly a drone to shoot things that you never interact with nor would have actually been intended to go near. That sort of thing is designed to fill time for a section where the level design itself is boring. (In one instance, they shove that in to hide you going up an elevator) It deviates the game away because they couldn't think of anything better to do, or to keep the action constant. (Which sometimes you really need breathing room.)
I think the most offensive game of recent memory had to be The Order 1886, where you really never got level design of any meritable interest when played. It was very dead-set on the idea of being a movie, to the point where the game part was walking and shooting a few things. As much shit as people give Gears of War, it was atleast more dynamic and interesting than The Order, even if it was very much shoving you into corridors when you weren't locked in a battle hub. The Order never once hides that fact, and even story sections can end up with you walking down right angle corridors of brick and metal.
On the opposite side, Lost Planet 2 has the best level of any game: the train cannon, and to a lesser degree the hover carrier. it's a level that is also your greatest asset- you and your friends have to fight a very large enemy, maintain the cannon's structure so it won't blow up, find the best shells in the structure to fight the giant worm boss, load them in manually while one of your friends (or you) aim and fire the cannon. the better you do, the bigger the reward for killing the thing. Even then, the rest of the game puts huge emphasis on choke points, elevation, destructible buildings, secrets, and large area to play in. even the battle hubs are massive, and they have buildings and structures to climb on and take better vantage of the fighting. It's a very well designed game, and everything about it works together to create a fun, albeit very goofy experience. It's not the prettiest game by fidelity standards, but it does what it can with what it has.
To sum it all up, things are more condensed because of technological limits that we're constantly hitting against in the drive for 'more better prettiness'. I think at this point there should REALLY be a change in focus away from this, and put more emphasis back into cause and effect and looping structures to allow the player to see their changes in the level. backtracking can be fun if you make sure the player never actually has to stop moving forward to get back around. Just... don't make them go back the way they came, but this time have even more things spawn in. That's the worst.
Both Mirror's Edge and The Order are linear however M's E proposes levels that are not only gorgeous visually, but they're fun to play in (and allows the player to choose and create his own way depending on his skills).
After playing it for the first hour you realize how empty it really is and just rush from one mission to another. I don't see any reason to go and collect stupid banners, coins (whatever).. unless they offer you some additional info on the story/setting (and it better be voice acted, atleast a big part of it). I do realize that putting some generic stuff out there without any reward is cheaper though.
I only have a very basic understanding of level design.. but I could imagine that there's a fight between gameplay and realism. Realistic locations just don't always have the perfect layout. Older games could be a lot more abstract and some settings and genres still offer that.
http://bobbyross.com/tutorials/
I will say though that I'm glad we moved passed the Lion King game for super nintendo. Fuck. Dat. Impossible. Game.
Donkey Kong Country was 100000x better but good lord is it me or any game that has water levels usually made the water levels impossible. I'm looking at you Zelda...
I wouldn't say level design is dead though. Valve games are usually great at holding a candle to level design (though some have argued with me that they're boring linear maps so I guess to each their own.)
I would argue that there still is a lot of level design done in open-world games, it's just not as noticeable as a traditional linear corridor. Someone had to place all of those patrol points, weapon emplacements, cover, choke points, vehicles, helicopters, etc in MGSV as an example; and I'm sure they playtested many different setups before they found what felt right. The AI would probably seem stupid as hell without a level designer to adjust everything to the Ai's strengths and to give enough variation in the nav paths to keep the player on their toes.
I'm still in a somewhat-junior position so everything I say has to be taken with a grain of salt...
In a very simple way, usually the modern workflow goes as follow:
- A small team of seniors gather to imagine the core locations/encounters
- Designers take over concept arts and design guidelines to flesh out every location
- Artists start working on the first art pass
- Designers add relevent gameplay elements
- Artists polish the level, add non-gameplay relevent props
What changed compared to 10 years ago, is now there is barely any blockout made in the second phase that will stick until the game is released. There is so many iterations, so many people responsible for so many things that everything that can change, will change.
I don't think designers got necessarily worse with time, it's just that a lot of their work was taken off their hands or spread around so many people it's hard to keep a single direction.
That's it..but then, I feel like people give up so quickly over difficulty to a game they don't even bother trying harder anymore...