Personal Updates

Lots of updates I'm sure everyone is dying to see!


Firstly, I got First Place in the Overwatch Level Design Contest! I submitted OWC_Traverse, set in City 17. The rebels have to fight their way through the city to reach an Combine forward base and destroy it. The goal of the contest was to create a new map for the mod, which pits up to seven player controlled rebels player from the first person perspective against the "Overwatch", a human player controlling all of the Combine troops facing the rebels from an RTS-lite overview. Sort of Sven Co-op meets Half-Life Wars. It provided some interesting challenges, like providing the Overwatch player with a clear overview of the map, keeping the rebels from wandering into the Combine spawns, and a lot of entity scripting work for the objectives. It will be included in a future release, so I'm looking forward to that! I'm supposed to receive some notes and feedback from the playtesters, which is also great. Congrats to the other winners!

In other level design news, I'm working on a level for the Rooftopville competition over at PlanetPhillip. It's blocked out and I'm starting to add in enemies and encounters before I start texturing and detailing. I think A Brief Detour suffered in that the combat encounters were too tight and left no room for maneuvering. The enemies were right in your face, and there were no alternate paths to take or ways to flank the enemy. The courtyard in particular, where the Combine rappelled in, had no full cover and nowhere to hide or regroup, just eight Combine rushing in, giving the player no choice but to stand and fight it out.

I knew there was something wrong with the combat, but it took someone else playing through and telling me in different words for me to realize what was it was. But now I have a much better understanding of how combat spaces work. I know I was thinking about alternate paths and varied cover the whole time, but somehow that didn't make it into my map. It was way too cramped, and almost all the encounters dropped you into a space with little to no options. The player needs room to maneuver and possible cover between them and the enemies, including actual alternate paths. Thinking back to the encounters with enemy grunts in the first Half-Life, some of the most interesting encounter spaces were large warehouses, with multiple elevations and dozens of potential paths through stacks of crates, allowing for flanking and giving the player breathers from combat. Hopefully I can incorporate those ideas into my Rooftopville map.

On the job front, I'm still sending out resumes and soliciting feedback on my portfolio, since there seemed to be a burst of level designer positions posted in the past few months. I'm certainly getting some practice writing cover letters! No responses yet, but I'll keep at it.

I also have a new portfolio up at http://samcombs.carbonmade.com. Check it out and tell me what you think! I'll be adding OWC_Traverse up there shortly.

Left 4 Dead 2 Survival Map - Office Space

Woo! I finally got this map put together. It's a Survival map set in and around an office building. Go check it out here.

Minecraft!

(image shamelessley stolen from Quinns' post on RPS)

I've been distracted by the excellent Minecraft. I'd seen a few posts on it, but what finally pushed me into buying it was Quinns' Mine the Gap series on Rock Paper Shotgun. I've only just started, but I'm having an amazing time building shacks, bridges, and beacon towers, carving stairways into cliffsides, and delving into huge pitch black caves. It's extremely satisfying to stand atop a tower you've built out of blocks you've dug out with you own block, pixelated hands and look out on an infinite randomly generated land, ripe for exploring.

There's a healthy dose of Dwarf Fortress style 'losing is fun!' - enemies blowing you up or arrowing you to death, falling off my most recent construction project, even drowning - I broke a hole in the ice and jumped in. What I didn't take into account was how quickly my hole in the ice would freeze back over, so I was left pounding futilely against the underside of the ice until my air ran out. I lost everything I had collected, but I was laughing. Each death is a lesson on the rules of the world, and an opportunity to laugh at yourself.

It helps that the penalty for death isn't too harsh. You lose everything in your pockets and get returned your original spawn, but everything you've constructed or stored stays as it was. The tiered material system makes it easy to get started or restart after you die. Almost all the blocks can be dug out with your hands, albeit at an extremely slow pace for anything besides dirt and sand. But once you're collected enough materials to build some tools, collection speed improves dramatically. The starting material progression for tools is wood-stone-iron. So you chop down a tree with your hands, then build a wood pick axe to gather stone that you can use to make a stone pick axe, which can be used to gather ore plus coal and more stone to smelt the ore into iron which can be used to make iron tools. This avoids the catch 22 of needing a tool to gather the materials to make the tool that you need to gather the materials, etc. that some other games with building fall into.

And this tech tree of sorts can be climbed quickly, but death is quick and common, so you'll have to repeat the hunt for materials over and over again. But instead of becoming a grind, this is one of the core bits of fun in the game. Wood and stone are very common, but the game forces you to explore to find the rarer ores, taking advantage of the infinite world. And there's always another striking clifface to scale or picturesque waterfall spilling down into a cave to marvel at, with constant rewards. Another reward are the short refrains of music that seem to play at random times. They're beautiful little pieces, they drew my girlfriend into the room to see what I was playing. They reward the player just for still being alive, a sort of a variable ratio reinforcement schedule. And there's enemies and combat and construction on top of all of that. In short it's an amazing game, and all by one guy. I'm more than happy with my purchase.

In other news, I'm attending the IGDA Austin Microtalks this evening, where ten speakers speak for ten minutes on a variety of topics. It should be interesting, hopefully I can meet some industry people there.

Left 4 Dead 2 Survival Level Progress #2

Still chipping away at the Survival level. I solved a problem with the flow in the nav mesh - it kept giving me flow errors, but nothing was popping up when I ran 'nav_trouble_report" from the console. I had an area marked PLAYER_START, all areas were marked FINALE and BATTLEFIELD. I finally tracked the problem to the "goal" nav area it was using by combing through the log of the nav_analyze. In survival maps without a checkpoint, the area closest to the center of all the nav areas is used as the goal, and it turns out that the goal area it chose was on the (player inaccessible) roof. So I added on a new nav area to shift the center back onto a flat surface, and that fixed the problem. It seems like there would be an easier way to do it, but this one worked for me. Besides that, textured a few areas and moved some props around. I'll keep at it!

Left 4 Dead 2 Survival Level Progress

Still working on my Left 4 Dead 2 Survival level. It's an office building turned CEDA evacuation center, now abandoned and overrun. I've roughed out most the geometry, placed a few props and textured a bit. It's still missing something. I should have planned a little more before I jumped in to the editor. I had some thumbnail sketches of how I wanted it laid out, but it morphed as soon as I started laying out brushes. It feels a little claustrophobic, so I think I'll expand it, make the second story accessible, perhaps. It just needs some more polishing. Testing is a little cumbersome, since the nav mesh has to be rebuilt every time the geometry changes, so it's harder to jump in and test after making major changes to the map. There has to be a way to automatically fill in the gaps or update the nav mesh without deleting and adding by hand. I'll take another look through the documentation and browse some forums.

Indie Game Magazine Interstellar Marines Preview

I'm in print! A preview I wrote of Interstellar Marines appears in the Sept-Oct 2010 edition of Indie Game Magazine this month (page 5). You should check out both! I've written some demo reviews before, but this is the first full fledged article I have in there.

Just Left 4 Saboteur

It's been a while. I'm playing a lot of Left 4 Dead 2 with some college friends, which inspired me to try my hand at a Survival level. I have not worked with the navigation meshes before, my experience is limited to waypoint graphs. But I like the mesh approach, it takes the busy work out of pathfinding since they're automatically generated as opposed to the manual placement of nodes. The AI director also reduces work, randomly generating mobs and specials on the fly. I love it, it lets the level designer focus on creating combat arenas and not worry about the exact placement of enemies. At any rate, seeing how enemy spawning and pathfinding works is helping me get better at the game, or at least a better grasp of why things happen the way they do.

I'm trying to brainstorm up some settings for a Survival level that haven't been done before. Trailer park and office building atrium have been bouncing around, not terribly original, but interesting places to fight zombies. I'm analyzing the existing Survival levels, seeing how my friends and I choose spots to make our stand and that what keeps us playing again and again on certain levels.

Besides Left 4 Dead 2, I've just come down from an open world game kick, playing hours and hours of Just Cause 2 and The Saboteur. Both have cartoonish plots with unique ways of getting around the world, with a grappling hook and climbing buildings, respectively. I feel like I'm playing a cheesy comic book character in both, which isn't a bad thing. Neither storylines take themselves seriously, but that fits especially well with the ridiculous never ending explosions and car/helicopter/boat crashes mixed in with gunplay that is Just Cause 2. The grappling hook/infinite parachute combo is insane, it beats boring old walking and running around.

My problem (not really a problem) is I have a hard time getting to a set destination in the game world, I end up distracted every time. There are so many "Ooh, shiny!" bits to distract me. And they are literally shiny - the pickups have a pulsing glow that catches your eye, besides the proximity indicator helpfully telling me goodies are near. More than six different types of collectibles, every little hamlet or outpost has a progress counter and is filled with goodies to steal and things that blow up real good. I've had a hell of a time with it, and I'm less than a quarter of the way through the game thanks the mentioned distractions. As Tom Francis so succinctly put it, it's 'the screw-around game' - a huge toy box filled with stuff to blow up and mess with. And I am more than ok with that.

The Saboteur is a little different. There are a few nice touches like being able to move the camera around during cutscenes and the colorizing of districts as you liberate them. I love darting up any building, leaping from rooftop to rooftop. There's something very satisfying about getting the drop on a Nazi soldier, then sprinting across the rooftops to duck into an attic until the coast is clear. It helps quite a bit that Sean seems to be made of Kevlar, letting me take stupid fun risks like brawling hand-to-hand with squads of Nazis armed with machine guns.

The storyline missions don't take advantage of the running across the rooftops excitment very often. Most are copied to the letter from previous sandbox games - tail person X, chauffeur NPCs around, ditch your tail, etc. They have none of the zaniness of Just Cause 2 missions, but it's hard to get crazier than grappling onto ICBMs while in flight, then blowing them up and parachuting off. When the missions deviate from previous formulas, and use the roof climbing freedom, they can be a lot of fun. Stealing a bottle of booze back from a German officer was a lot of fun, mainly because the game gave you an open ended goal and then got out of the way. They give me enough tools - sneaking, disguises, climbing, melee attacks, explosives - that when the game lets me, I can take my own approach.

The plot is pretty stale, and the characters are dry and forgettable. The crazy accents in Just Cause 2 are at least new stereotypes as opposed to the French and German ones in The Saboteur. But the real character is the setting, and I find it fascinating. It's worth buying the game just to experience occupied Paris. It makes me wish there was more interaction with the everyday people. Random events like watching an innocent civilain being arrested and loaded into a truck, then hijacking said truck and freeing the prisoner are extremely rewarding, and I wish there were more events like that to mix it up.

As it stands, the routine of hunting down guard towers to blow up became repetitive after a few hours. The only reliable way of blowing them up seems to be the same stick of dynamite to the base technique, and a little variety to the destruction would be an improvement. Maybe guy wires to destroy, or just more heavily guarded targets? There's not much challenge to blowing them up. Taking out targets in Just Cause 2 or Red Faction: Guerrilla is much more challenging, you usually have to shoot and scoot, not so much in The Saboteur. I think this might be where some of the complaints of dumb AI might come from, because it is too easy to get away with blowing everything up. There has to be some challenge to be interesting. But despite all my complains, I don't understand how little mention this game got. It made me sad that Pandemic is gone now - there's a solid core of a game there that could do with some tweaking, and it isn't the buggy mess that was Mercenaries 2. Some reviewers have mentioned that Assassin's Creed did the parts I liked - the climbing buildings and jumping about - better, so maybe that's why people seemed nonplussed.

Unfortunately, technical difficulties are getting in the way of my enjoyment of The Saboteur. SecureROM is acting up, it only reads my disk when it feels like it, telling me to insert the original disk that is already in the drive. It's been doing the same with Borderlands, and it's frustrating.

There will be more updates on my Left 4 Dead 2 level soon, so stay tuned!

Mirror's Edge

I finally got around to playing some of Mirror's Edge, after buying it during the Steam Holiday Sale this past winter.

The environments really are stunning, and it is refreshing to see so many bright primary colors. Sprinting, jumping and swinging from perch to perch is thrilling and visceral when it flows, and a terrible chore when it doesn't. The runner vision that highlights the best path helps keep you moving along, but there are a few places where I got lost or frustrated, namely jumping around the sides of a huge hole in the ground, a gigantic storm drain. Very visually impressive, but I ended up jumping to my death more than a dozen times in a row in a part where I had to run to catch a timed door.

Mainly I missed wall runs when jumping from platforms to platform along the edge of this hole and plummeted to my death. So after a half dozen or so tries, I switched to normal jumps and made it just fine, but I still wanted to wall run across the gaps. Wall running seemed so much cooler, it just seemed like the natural move. Why would I make a normal jump when I can run on the freaking walls? But I have a hard time getting the key combination right for some reason. Maybe it was the curved walls of the storm drain that was making it harder to register the wall jumps? Or I'm not hitting the buttons in the correct combination?

The huge circular storm drain was also a pain because I wasn't sure which way to head. The checkpoint pointed me in the wrong direction, facing a platform that I had just jumped on but forgot about. So I ended up dying another dozen times before I figured out my mistake. Solution: Maybe make the platforms more unique with different colors or lighting patterns? Maybe have some of the platforms collapse or retract or get blown off after I've used them?

Another spot where I had trouble was in the first level after the tutorial, where I had to make a long jump from one building to another and catch a pipe on the other side. Again, this took me more than a dozen tries. I kept missing the pipe, or hitting the wrong button, I'm still not sure what I did to finally catch it. I'm not sure if I need to activate pipes when I hit them, or if I wasn't crouching at the right time, or if I just wasn't jumping at the right time? There were too many factors to pin down what I was doing wrong. This could perhaps be fixed by having context clues, like having the pipe flash red when it's the right time to press a button to grab it? Or have me grab a pipe earlier in the level in a safe spot where I don't have to worry about making a huge jump, then have me make a long jump without a pipe, then combine the two once I have become confident doing both together.

That's my main complaint, that the controls can be frustrating. I can visualize what I want to do, but it's hard to get Faith to do it. Maybe just a more clear tutorial on wall running? Force me to practice it a couple times in a row where I can try it again and again without penalty? Maybe it's easier to pull off wall running with a controller, I am playing with a keyboard and mouse on my PC. Or maybe I'm just not that great at the game, one of the producers of Mirror's Edge in this postmortem on Crispy Gamer mentions that if he could go back and add something to the game, he would add a variable difficulty system that made it easier to make jumps, gave the player a little more room for error. Maybe that's what I need, but I still feel there were a few spots where I feel like I should not have gotten lost.

I don't want it to sound like I'm bashing the game, it's just that the fun bits fly by, and when I get lost or have to make a jump over and over again, it gets burned into my mind. That does not make the good bits any worse: any game where I can slide down an office building, jump off and drop kick goons as I land is a winner in my book. Having bullets flying by and glass shattering all around me as I outrun an entire squad of cops and a helicopter is an awesome experience, all without firing a shot.

One idea I was thinking of as I played was having free reign of the world, Mirror's Edge as a sandbox game, finding and delivering hidden packages scattered throughout the world, crossing from one end of the city to the other on the rooftops. Probably a nightmare to design such a huge space without the player getting lost, since I already had trouble with that in the linear game, but I think it could be done, and it's a fun daydream, finding my own fastest way through the world, the time trials writ large.

SXSW Interactive Day 3

Day Three! I presented Robot Revolution for the 2010 SXSW ScreenBurn AAA Game Design Competition today! I competed with some awesome guys, Fredric King & Ryan Doyle with Tall Bike Joust, and Patrick Cunningham, who had Grimore and Shadow Wars.

King and Doyle presented first, followed by Patrick, then me, then Patrick again.

King and Doyle had implemented the first of the planned stages for their game, an iPhone game about bike jousting, which looks pretty intense. I think it's awesome that they were able to actually implement their game, that gave them some serious points with the judges. The iPhone game wasn't the AAA part of their plan, they're also going to make a browser game and a console version, with more than just the bike jousting. There were some technical difficulties, the sound wasn't working for the laptop set up for us, but they improvised and gave their own voiceover.

Patrick Cunningham had two game ideas and presented twice, which definitely worked out for him. Grimore was an open world fantasy RPG, with a complex sounding magic system, and Shadows Wars was a cyberpunk MMO. A unique idea from Shadow Wars was doing mini-games or puzzles before a mission to give you an advantage, like more info about the level. I thought that was pretty cool. His wife did all the drawings for his slides, which were also awesome. His slides were pretty densely packed with text, but the illustrations helped, and he seems like a great guy.

After the presentations, we were lined up at the front of the room and asked to give our elevator pitch, then the audience applauded for each contestant. The judges (Adam Martin, Chris Charla, Jesse Redniss, and Lori Durham, with Kain Shin moderating) then publicly decided among themselves which game was going to win it. Robot Revolution wasn't a favorite of any of the judges, so I think it was ruled out pretty quickly. Shadows Wars won it, with the judges saying it seemed to have the most commercial appeal, being the most likely to be picked up by a studio and made. Tall Bike Joust got a mention for also being marketable since there was already an audience for tall bike jousting. The judges seemed to have a little difficulty choosing, and there wasn't a clear cut criteria or rubric. Having an even number of judges also seems like an odd choice, since there was almost a tie.

I wish I had gotten more feedback on what wasn't as great about my idea. It lost, so there has to be something wrong with it, and more feedback would have been nice. But I got a sweet trophy for being a finalist and got to meet some cool people, so it's all good.

If anyone would like to take a look, I've uploaded my presentation here. Make sure you check out the speaker notes (click the little face with a plus sign), they explain a lot. Any feedback is welcome, either on the blog or at sam.combs@gmail.com.

The other finalists were great, and congratulations to Patrick for his win!

P.S. - I also saw the Casual game design contest, with Escape From Planet Zero by Lance Meyers taking home the prize. I might write up a bit more about it later!

SXSW Interactive, Day 1

So I had my first day at SXSW. I came by Thursday and got my pass early, and I'm glad I did - the line today was huge. I got a chance to see three panels and have a quick look at the ScreenBurn Arcade.

The first was "Take Over the World With XNA Indie Games", with Chris Williams presenting. Pretty small crowd, but I think people were still registering and getting their bearings at 2 PM. He did a very general overview of XNA, nothing new if you've checked it out before. But I hadn't heard of the Windows 7 phone though, that apparently has multi-touch and will support XNA, which might be an interesting combination. Learning about the peer approval process for Community Games was also new to me. Still, I was expecting more of a look at up-and-coming games, sales figures, that sort of thing. Still, it would probably be a good intro for someone that had never heard of it.

Next was "Drawing Board: Innovation Lessons from Cartooning", with Tom Fishburne. This wasn't one of the ScreenBurn panels, but I'm glad I sat in on it. The basic lessons I drew from it:

1. You can't expect creativity to just come to you, you have to exercise your creative muscles and constantly be on the lookout for new ideas, you can't really force creativity.
2. If your work is polarizing, that's not necessarily a bad thing. If you don't get any sort of response, or only neutral comments, you're probably doing something people have seen before, nothing innovative.
3. Even niche subjects can have a broad appeal if you're doing a good enough job. Blogs can help to give context to inside jokes.

He also mentioned Russell Davies' blog post "How to be Interesting", which seems to boil down to the idea that people will find you interesting if you're interested in them, so have diverse interests and pay closer attention to the world around you, and be able to get people to talk about themselves. He suggesting regularly taking pictures, making blog posts, reading about subjects you haven't heard of, and just generally paying more attention. Makes sense to me.

The last panel was pretty interesting - "With Great Power Comes Great Responsibility - The Future of Video Games". I wish this one could have been longer, it lost some time due to a false alarm evacuation of the room fifteen minutes in. Jesse Schell's "Design Outside the Box" presentation at DICE was brought up, and the general opinion of the panel members was that the use of achievement systems to foster good behavior was a good thing, if a little creepy if taken to extremes.

I wish this had been discussed more, it seems like such an interesting concept. They suggested that as long as individuals can control and mix and match the achievement systems they want to use, modifying them to fit their own ends, that achievement driven structures could be used for good, and I think I agree. I wonder if artificial achievements are all that appealing to people that aren't gamers or people with addictive personalities.

I'm looking forward to tomorrow's round of panels, hopefully I'll have a chance to get a better look at the ScreenBurn Arcade. In the meantime, I'm really looking forward to presenting my entry for the ScreenBurn Game Design Contest at 3:30PM on Sunday. A little nervous, but still excited.