A Decade of Defense Grid
December 8th, 2008 was the launch date for Defense Grid: The Awakening. Ten years ago, it took the Tower Defense genre and gave it the treatment of a high-production value game. It’s seen a sequel, several expansions, and even a board game. It’s a game that I spent hundreds of hours playing, and continue to play to this day. To celebrate the game’s Anniversary, I sat down with Jeff Pobst, founder and CEO of Hidden Path Entertainment to chat about it.
Filippo Dinolfo: Let’s start by reflecting on DG1. Do you consider Defense Grid a success?
Jeff Pobst: Yes, and success is a funny thing. You can look at it as success in entertaining people. You can look at it in reaching a large number of customers. You can look at is as success as whether or not you made back the money you put in, and made a profit on it. But overall, Defense Grid has been a good success for us in that it created a game; people seem to resonate with the game. It reached a lot of people through sales and through things like Games With Gold. It won awards, people seem to really like it. We got a lot of email, people asking questions, wanting to know more, dive in. It was really fun for us to build, we had a lot of fun making it, and it seems like people had fun playing it. So, kind of on every metric I can think of, yeah, it felt like a success.
It took a long time, it didn’t instantly become a success. A lot of games will sell most of what they’re going to sell in the first six months, and Defense Grid took about three years before its sales started actually dropping. They slowly started out, and they kind of grew, and grew, and grew, for about three years, and then started dropping.
FD: So you had a bit of a slow burn.
FD: Which is pretty good,”Slow and Steady” as they say.
JP: Oh, yeah. We just didn’t know in the first two years when it was going to fall.
FD: It came out at a perfect time, really. You were basically the apex of Tower Defense at that point.
JP: And that was our goal. We had seen all of these Flash desktop tower defense games. Desktop Tower Defense was the game that we saw–we went, “Wow that’s awesome!” We were so impressed with it and it was a Flash game. Then we said, “Do we think we can bring something to this genre?” Because, while it was a cool Flash game, it didn’t feel as deep as we were looking for. And at the time the idea of the high-production-value indie game was not really a concept in 2007. Xbox Live Arcade was really the only channel for such a thing, and at that point in time there was kind of an understanding of how large the market was
For Xbox Live Arcade, so no one had a budget really more than $500,000 for a game that would go on Xbox Live Arcade. $500,000 is just an arbitrary number, but what that matters from a development point of view is how many person-months you can put on a project. How many people for how many months can you put on a game. That’s going to drive your budget and hopefully your budget is going to be under what you expect to make.
We felt like this was going to be an area of growth, we didn’t quite understand what was going to happen in the indie world. A lot of it had yet to happen, but we thought that games would grow and there would be another channel for games that were larger and more deep and you could sell at a reasonable price, $10, $15 or $20 price point. And so we worked with some investors and came up with this idea of the Million Dollar XBLA game. And what we thought was really compelling, we’d take Tower Defense and give it the wrappings and the quality and the attention-to-detail that you might if you were building a fairly large Strategy game. Obviously this isn’t a fairly large Strategy game, but we’re going to try to make it feel like it.
FD: So the plan was always to go to XBLA? That particular version came out much later.
JP: Yeah, the only plan was XBLA, and it was only when Microsoft had a management change [that] we had to start scrambling for an alternate venue. So the XBLA controls were the original controls, and it wasn’t until we got ourselves into a situation where Xbox said, “Hey! There’s new people here, your game’s great, we like it, but we don’t want to release it this year. We’re thinking that instead of the date we [said] you could release your game, you can release it a year later.” Financially we couldn’t make that work. So we started looking for alternate plans. We looked at our contract and the other consoles that were out. Getting a slot on XBLA at that time was very challenging. You had to sign a lot of agreements to get such a slot, that typically you’d only see when it was being published by someone or they were financing it. In this case, Microsoft wasn’t financing it at all, but they had the slot, and the slots were considered extremely valuable on a highly curated system.
I think that’s one of the things that led to an outcry of things like, “Let’s make it free” so that anyone could put games up anywhere. We’ve seen that growth with Steam, we’ve seen that growth with XBLA. The pendulum swings from App Stores to Curated Lists, but, each one has its benefits and problems.
FD: We’re starting to see some of those problems crop up on Steam now.
JP: Sure. It’s very hard to get discovered on Steam now with so many games.
So we approached Steam. At the time it was just PC which was not a competitor to Microsoft’s console. It was a Microsoft platform, and Microsoft didn’t have a store themselves at the time. So we went to Steam and said, “Hey, do you think you’d be interested in putting this up on Steam?” And at the time there was about 300 games on Steam when we had that conversation The closest thing to us was Darwinia. Do you remember that?
FD: Oh yeah, I do.
JP: So the folks at Steam looked at it and said, “Yeah, this looks like a game that could be on Steam. Yeah, we like it, this is pretty fun! Let’s play a bit more, oh wow this is pretty good! Yeah, let’s do that.”
So the big thing for us was, we were going to release in September of 2008 on Xbox. That didn’t happen, we were told our slot wasn’t going to happen. That was in May of 2008. So we started trying to figure out, our whole UI system was Xbox only, it was a Microsoft library for Xbox. So we had to get rid of that whole UI system and build a new UI system that could be for multiple platforms. All our code was only compiled for Xbox and we were only controller based. So we had to come up with a mouse and keyboard system, a new feel for the controls, and a new way for moving the screen around and all of that. All of those changes ended up making us ship instead of September 2008 in December 2008.
FD: So what were some of the takeaways from 10 years of Defense Grid?
JP: I think the thing that’s been really interesting, we were analyzing this the other day. In the first Defense Grid, we came out with a game that everyone could jump in, you could play it with one hand and a drink in the other hand. So at the time, we looked at… I want to say the right number, it’s been a long time, but I think it was 84 different mods and Tower Defense Flash games and we built a taxonomy of what their features were. It was this giant spreadsheet. Someone printed it out and it was like, 40 pages or whatever. We ended up going through it in a design session and saying “Well this feature and this feature and this feature seem best-of-breed.” And what we were trying to do was make what we called “The Definitive Tower Defense Game.” If you played this game, you understood tower defense.
So fixed paths, paths where you could redirect and have your own creativity. We were looking at the emotional curve of the game and we felt like it didn’t quite end right. You’re closer and closer and there’s stress, but once one alien gets through the base the emotional drop was really significant and you felt that there was nothing more you could do. It had a little bit of a rogue-like feel where it was just going to get worse and worse, and at the time that wasn’t quite as embraced by the audience as maybe it is now. So we came up with the idea of the Cores and Stealing the Cores and having to get them out of the base so that something bad could happen and you could react to it. There was this emotional spike of, “Oh no! Bad!” but I could build more towers or I can use the special weapon, and I can stop bad from happening and it sort of goes back and forth emotionally until you either succeed or don’t at the end of the level. And that was a big difference for tower defense. The idea that you could have this different emotional curve of being able to handle adversity that came in these kind of quantum chunks, whereas before tower defense’s adversity quantum chunks came only in failure.
Then the second thing was that every tower defense game we’d seen at the time was either a fixed path with build on the side, or it was create your own pathing. There were benefits and detractions to each of those. Fixed paths people understood and got into really well. Those were often more popular tower defense games initially, but didn’t have the depth so people didn’t stick with them. The build anywhere you want and cause the redirection of the aliens or the creeps or whatever you would call it, was deeper and was stickier for people, but a lot of people felt uncomfortable getting into it. So our thought was, well, if you could have fixed path to learn on, and then fixed path goes to build area that has some choices of how you redirect aliens and send them around–you know like some switches like on a railway. Anyway, we did some stuff in level design that I thought really changed tower defense and the results of that, which I think is your question, the feedback that we got back was, “Yeah, this really worked for me.” People really enjoyed the puzzle nature as well as the strategy nature and we came up with a lot of different rules of thumb from design about how large gridded areas should be, how many entrances and exits to an area before it hands off to another area there should be, how complicated things should get, or not. That really led us to trying to make good and entertaining Defense Grid levels.
FD: And you succeeded. One of the things I loved about Defense Grid in particular was the difficulty curve you had. It started off fairly simple, but by the time you got to Last Stand, you had everything being thrown at you.
JP: That’s the really fun part about these games is that they grow and escalate, each lesson you can build upon. You see that in a lot of great games, but it felt really good. Our mantra at the time was, and this is less relevant today, because so many people today, it’s easy to forget, back in 2006-2007 a top AAA game, maybe not the highest, GTA probably still broke records, but a top AAA game sold 3 or 4 million copies. Not many people got to experience a deep AAA game because the barrier to entry was pretty high for a lot of people. They didn’t feel comfortable, they didn’t feel very good, and a lot of games didn’t have an easy on-ramp. One of the things we’ve seen in the last ten years, and that we’re starting to see some backlash on, is that we’re making games too easy for people to get into. We can debate for a long time whether that’s good or bad, but at the time we felt like “Wow, there’s all these cool core game experiences that a lot of people don’t get to experience because the game is too hard for them. So can we give you that emotional feel and strategy feel and the puzzle feel of a core strategy game but make it really easy for you to get in, easy for you to access, and kind of drip it to you over time so that you feel pretty good about yourself as you go through the game. We felt like as game makers, that’s part of our job. If we were Disneyland ride designers or environment designers our job is to make you feel like you’re at a fun, cool place and participating in a thing, as a game designer we felt like our job is to make you run into this false adversity but be able to figure it out and overcome it. The more people we can do that for, the better, but if you make it too easy, you’re not doing it for anyone. So it was a tough thing. And that was one of the things that the core mechanic really gave us more flexibility than I think we even recognized.
By giving you 24 cores, some people who felt like they were very good strategy players, did not want to ever lose a core. That was how they played it. They made the game more difficult for themselves because of their preset notions. Whereas a lot of people who had never played games like this before if they had one core left, THEY WON, and they were excited. It was the same game, but people were playing it in very different ways because of their own perspective on what they needed to accomplish to move forward.
FD: I think the medal structure you had was a big part of that. You could pass a level with one core left, but you always had that incentive to do better.
JP: That’s the thing right, that you know you’re trying to do as a game developer. You know you’ve got hardcore serious gamers and you want to entertain them, and you want to make sure the game is deep enough for them, and you have all these other people that might try it and might pick it up and you want to entertain them too. So you’re trying to figure out all these systems that can encourage the person that’s going to take this game very seriously to do all the things that they want to do and not be so discouraging for the people who aren’t. So the medal system, I think I remember the first time I saw, well, the one I think about, maybe it’s not the first, but the medal system that really resonated with me, I remember when we were talking about when we did this, was the one from Burnout. If you remember the Criterion game where you crashed your car. So for us seeing a spectacular crash was really cool. If you wanted to go and try and create the most damage and get a Gold medal you could, but sometimes that didn’t create as spectacular of a crash. So for us, the Gold, Silver, Bronze medal was great in that you saw that it must be possible to do more by seeing those medals, but if you had a cool experience, you didn’t care. So that was what we were thinking when we went into Defense Grid’s medal design approach.
FD: So whose idea was the flying aliens?
JP: <<laugh>> I don’t know if it was an idea or if it came from the taxonomy. The basic premise is, I’m going through and I’m building my layout and my path and so forth, and I understand it, and then all of a sudden something comes in that doesn’t follow the pathing rules anymore and I have to deal with it, too. So that was the philosophy behind the flyers. We then said “Okay, how many towers can shoot flying things?” Overall at the end of the day when we’d shipped the game and we’d received feedback, we weren’t super happy with how flying aliens and building for flying aliens, and the reactions from players worked out exactly. It wasn’t horrible, but it wasn’t optimal. It was one of the reasons why we made the change in Defense Grid 2. I’m not sure we made the exact right change, and I can imagine playing with that mechanic a little bit more, but the fundamental idea of “I have two fronts to manage” in a strategy game. Or that I have two different pathways to worry about now that are kind of stacked, was the idea behind the flyers.
FD: I think the only complaint I had personally was that if you shot down an alien that was carrying a core, you lost that core. There was no recovery from that.
JP: That’s right, that was part of the plan. Sometimes the fiction is important to you and sometimes it isn’t. We floated the cores around the level and there’s no real explanation for that other than “Fletcher can do that,” but we never pulled a core back from a flyer because for some reason we thought that would look weird. So we said that we could only pick up a core at the end of a long flyer path so that you had a chance to shoot them down before they could pick up the core. And it made that core moment a little more frightening and scary, but it was more akin to a traditional tower defense game. In the rest of the game we’d changed that emotional curve, here we’d brought back a little of it. I think we agree, yeah, we didn’t like that. So with Defense Grid 2 the idea was that the aliens would start at your cores and with the lander you would get a chance to do some damage before they deploy, and then they just go from that point to the exit. They don’t get to go through the whole map. So that means you’ll have to do some front builds, back builds, mid builds and you’re going to have to play with that knowing that that will be an entrance for them as well. Which was the main goal of the flyers, in that respect, but we agreed with you that we just didn’t like the fact you couldn’t get the cores back.
FD: For us, what that ended up boiling down to was that when we heard the pod drop, we’d go over and Orbital Laser it and when they did deploy, they would start with their health in the red.
JP: Right, and it became a sink for Orbital Laser use.
FD: That’s one thing he and I sort of disagree on. I liked that mechanic, it’s just one of those things I need to be sure that half my build is enough to deal with them.
JP: That was another piece of feedback we got from the first Defense Grid. “The Orbital Laser is cool and all, but I don’t use it.” “Why don’t you use it?” “Because it doesn’t give you resources when you use it. You’re actually losing resources when you use it.” And it was true, because we didn’t want to OP it, right? In the first game, your score and your resources were super tied together. That was good for a core player, but it actually turned out to be kind of bad for a poor player. A poor player would get in, would not build well would not get any resources, and get into an infinite loop of not being able to move forward. So that problem showed up and we were thinking about it for DG2. The Orbital Laser not being used because of not wanting to waste resources was another problem. So for DG2 we said we were going to give you resources for the Orbital Laser but we’re going to put it on a cooldown. We’re going to let you have different kinds of Orbital Laser in case you don’t want to destroy with it, you could give yourself bonuses or other things like that. And, we’re going to separate score and resources so that the new player doesn’t get into an infinite loop that just fails them out.
FD: That did technically cap what the max score on any map could be, though, didn’t it?
JP: Well there always was a theoretical cap because of the amount of time, I mean, even on the first game I’m pretty sure we stopped scoring if you started bouncing. We didn’t ever want it to be an infinite score problem.
FD: You could get away with some juggling.
JP: Absolutely. People liked doing it, so we didn’t want to take it away, what we didn’t want was for the people who cared about getting high scores to go “Oh, I have to play this way in order to get a high score.” Which isn’t very fun. So we wanted to limit, there are several, what you might call exploit mechanics, they’re not exploits because they’re part of the game rules, but we wanted to limit their success so that you couldn’t just lean on any one as the only way to win.
FD: One of the fun things you guys did was a cross-promotion with Portal 2. I remember being without internet for a few days, then when I got my connection fixed I check the DG Steam forum and find out there’s something weird going on. How did that whole thing come about?
JP: Valve had tried out doing an ARG themselves for the announcement of Portal II a year or so before launch and it had taken off across web sites and content sprinkled around the internet and gone much farther and was more rabidly followed than they had expected. So about 3 or 4 months before Portal II was going to launch, they invited around 15 indie developers to come over to Valve for a visit to talk about a possible group project, but they didn’t specify more than that. We were one of those indie developers who attended.
Gabe Newell entered the room and basically said that they’d like to do an ARG again around the launch of Portal 2, but this time, they wanted the ARG to take place on Steam rather than randomly across the internet. They thought it might be an interesting idea if many of the popular indie games on the service worked together, had access to some Portal 2 content, and put together an ARG as a group. Valve would provide access to the content, and would be willing to adjust the PC launch date of Portal 2 as an incentive for people to participate in the ARG, but the rest would be up to us. Would we be interested?
A couple indie developers opted out for their own reasons, but the rest of the group got excited at the concept, broke into subgroups, came up with pieces of a proposal and after that meeting at Valve the Potato Sack group was born (named after seeing the Potato battery that the player builds during one of the game’s levels). The group came up with a 3 week or so ARG progression where GLaDOS invades all our games to prevent Portal 2 from releasing, and by players overcoming her disruption, the release of the game could actually happen. Portal’s release date was in late April, so we had the idea of starting the ARG on April 1st (April Fool’s Day) which seemed like a good idea at the time, but in retrospect was actually a mistake as players weren’t sure they really believed that things we said were happening were actually happening.
But on April 1st that year, 13 different indie games all got an update with new content. The content was strange and the new content didn’t make a lot of sense. Then a bundle to let you buy all the games (and only get charged for whichever games you didn’t already own) on Steam went on sale so people could participate in all the content across all the games and it sold amazingly well. Some say it was the first ARG that actually made money, but the indie studio’s goals really were around entertaining players and hoping to recoup the extra money of development each studio was doing to add in new content to the game for the ARG. There was constant communication between all the studios, and we arranged for different types of clues and puzzles coming from the solution of one game which would then help start the puzzles in the next game and so on. Each game was updated several times over the following weeks all together, all coordinated, and when I think back, it’s pretty amazing that we pulled this all off.
Watching the players on chat and forums as they worked to solve the puzzles, or try to crack the games open and look at the updates in code, or do whatever they wanted to do was amazing. I remember that the Tiki Tori team in Amsterdam did a puzzle that required players to come and climb up the side of their studio’s building a short ways and find some clues ten feet above the sidewalk. One player actually came and climbed up and they captured footage of him from across the street with their cameras sitting in a coffee shop. It was amazing footage! We also saw his video he posted online and captured it too.
We needed to get an encoded clue out to players for a Defense Grid puzzle next, and so I remember getting the footage from them, adding some special characters onto the side of a Garbage truck that drove past using Photoshop and Premier – looking like those characters were drawn on the dirt on the side of the truck. We then uploaded the footage anonymously and pointed people towards it hoping they’d find the clues.
I remember wondering if the encoded text was way too subtle for people, and then having the ARG players find it, decode it, and wonder about what lengths we must have gone to in order to get those characters on the side of the truck in Amsterdam and have the truck drive by the window at the exact right time. Of course we didn’t actually do that, but I remember how the story grew so much larger than what we really did. It was so crazy.
Eventually GLaDOS appeared and invaded all our games, and each game had a special section or level or something that required the player to overcome the way that GLaDOS was messing with the rules of our games. It was a lot of fun, and I believe Portal 2 ended up releasing 24 hours earlier than planned or something like that because of all the participants.
FD: The ARG that was part of that ended up adding a bunch of fun content to the game. Not just the maps themselves, but the console stuff as well. The Unix shell replica was a nice touch. Was this an obvious choice based on what the DG UI already looked like, or did that evolve from something else?
JP: When we first talked about all that we wanted to do inside our game, it became clear that we could only do so much with new levels and that would only open up so much puzzle content (for example a top down view of the first level we released was actually a legitimate QR code that took you to a special website that had the next pointer to the next clue in the ARG). Because we wanted to do dozens of puzzles and mini-games, we came up with the idea of adding in the console as consistent with the IP of the game, adding in new voice content and new puzzles as part of it, and then making some fun ASCII versions of games that were fun to play but wacky for being in Defense Grid such as a Castlevania-style platformer or a single song version of Guitar Hero for your keyboard. We had a lot of fun in making quick, fast, and entertaining content for the ARG.
After shipping the two levels for the ARG, we approached Valve and asked if they’d be open to us making a full 8 mission expansion story featuring GLaDOS because we had so many new ideas for ‘modified-rules’ Defense Grid that would be perfect under the auspices of GLaDOS. They gave us the green light and later that December we released the “You Monster” expansion pack for Defense Grid on PC & Xbox 360. Valve was very kind to us and very supportive.
FD: So let’s move onto Defense Grid 2 for a bit. The first question I have is, do you think DG2 was just too big?
JP: I don’t know that it was dramatically bigger than DG1. The levels individually were larger, but from a number of levels and number of modes point-of-view it was pretty similar to the first. It added multiplayer which we thought our players wanted, and I would say they probably really didn’t want. Especially PvP, which at first we were told PvP is the way to go—everyone wants PvP. And I would say that statistically not many people play PvP. There are some, and they really like it, so I’m not trying to say that at all, but statistically from a large group of people, it’s not a common thing. The one request we did get a lot of in DG1 that I’m glad we put in DG2 is couch CoOp. A lot of people wanted to play with their kids or with their spouses or with another person cooperatively and just have a fun time building stuff together. Talk about what you’re building and do it at the same time. That mode we have seen a fair amount of use of. When we went to do the VR game we pulled all of those modes out, in part because they’re very demanding on performance, and in part because the VR audience wasn’t large enough for multiplayer.
That would be one area where we probably went too far. Multiplayer, unfortunately from a budget point-of-view, is super expensive. Because you have all these different console services, the Steam service, you have all these things to do to get two people together, matchmaking, reconnecting, making sure you have a good connection, supporting chat, handling it if someone leaves, just all of these cases where a ton of programming goes into multiplayer and I wish we hadn’t done as much effort in that. So that’s a size thing that we probably did too much of.
We also embraced—we were talking to the Valve guys and DOTA2 was becoming a popular place for people to upload their content and then Valve would sell it to other people and give the content creators a piece. So we were one of the first games to do that as a third-party. But our audience didn’t really—again, the idea of selling the content didn’t really resonate with them very much. They just wanted to make levels for each other. So we allowed that too, but we didn’t put that inside the game. If that makes sense.
FD: There were a few sites that showed up, I think DG2freemaps was one of the biggest.
JP: Right. Supporting the level editor was a lot of work and a lot of cost that I’m glad the community got to do and got to build, but from a return on investment point-of-view of number of people that used it, people who benefited from it and the time and money we had to put in, probably was not the best use of our time.
FD: Which is a bummer. A game like this could really benefit from a vibrant community of map makers. I wonder if part of it was that with DGArchitect, you fire it up and it’s like, “Whoa, what am I looking at?” It was not exactly the most approachable piece of software.
JP: No, and editors often aren’t, and we didn’t want it to be so approachable because we didn’t want a lot of bad content. Typically the people who are willing to make good levels are also willing to learn or familiar with tools that are similar. So, we kind of understood that, but I think the biggest mistake we ran into a few things. One, we partnered with 505 Games to not just release on Steam but also on console. A year or two earlier that probably would have been a good idea, it wasn’t 505’s fault or our fault, but the console market for mid-priced games had gone away with Games With Gold and PS+. So we spent a lot of time on console versions that maybe, maybe we shouldn’t have had. The first one ended up on PC and Xbox 360 and ended up reaching a million and a half people on 360 when we did Games With Gold. So that seemed like a really good way of getting people to know about the game, but when we came out, a lot of metrics were telling Indie games to raise their prices, when in fact the indie game market was going through a price crash. So the metrics were not helpful. So we priced the game too high at the time. We came out at $25 for the regular and $30 for the special edition. The market was just not right for that. Today the game is $15 and seems to do about what you would expect a $15 game to do. It definitely, both because it came out at too high a price and we didn’t get the marketing support that we were hoping to get. That’s a typical developer complaint, you know, there are reasons I say that, but, it’s not a fundamental problem. Then two other things happened. A lot more games came out on Steam and lot more came out on console, and then the day we came out, Steam changed its front page approach and went with the individual recommendations. That did not end up working out well for us at the time.
FD: It also just goes to show that $5 either way can make a huge difference. If DG2 had launched at $20 and $25, it might’ve done a lot better.
JP: I’d like to believe that as well, I wish I could go back and change it, it’s not one of the things I have the power to do. But yeah, I agree with you. It’s a really weird business world, we go and we spend money to make a game, and we employ a bunch of people and we spend our money, we spend our investors’ money, we spend some money from 505, and we build the thing. Then we take it and we sell it at a price to try and get that money back. If you pick the right price and you get the right number of people the equation works, and if you go too high on price, or go too low on price, or don’t get enough people, the equation can fail. So it’s a really weird business in that respect. You know, you’re always trying to find how much value does this have to a person, if they buy it for this will they be happy? You care about that as a game creator, very much, but you also want to have the numbers work out so that you can keep making games.
FD: Did you feel that you were on a bit of a time crunch with DG2?
JP: I think the reason that it felt like we were on a time crunch was that we were more on a money crunch. We had a limited budget, and we actually went over the budget and Hidden Path ended up covering the overages, our investor and 505 didn’t cover those.
FD: So 5 years into DG2, how has it done?
JP: It’s done okay. It’s got a good audience from a console giveaway point-of-view, Games With Gold and things like that, it’s done similarly to Defense Grid. But from a sales point-of-view it’s probably sold to about a third to a half of what Defense Grid had sold.
FD: That’s unfortunate, because there was so much this game did right.
JP: Well, thank you!
FD: There were a few elements that did feel a little unfinished. The tower items in particular. I didn’t feel like there was any sense of accomplishment in those. It was basically, if you played enough of the game, you got all the items. They do alter the gameplay in substantial ways…
JP: And what we created was some difficulty levels that really you needed the items to do. Without the items you wouldn’t be able to, so what we were hoping was that you’d try out the Very Hard, see what the challenges were, get the right items, find the way that worked for you so that you could take on that difficulty level. It was also pretty early, before item drops became as big a deal as they have become. We probably didn’t put the energy into that system as people do now, because they use that as a main monetization system. For us, we were doing something, we saw the tower items as a thing for the player who was going to spend more than a hundred hours.
FD: I certainly fall into that catagory.
JP: It’s interesting, we have these peeks where we look at our audience… We have the people that play 0-10 hours, they’re a group, we’ve got an equal group that plays probably 20-40 hours, we’ve got a group that plays 100-300 hours, then we’ve got a group that plays over 500 hours, it’s kind of crazy the different audiences.
FD: Any fun stories you’d like to share?
JP: One of my favorite ones was, after DG1 I went on a business trip and I turned to my business partner Michael Austin, “I haven’t read any good books in a long time…” this was probably in 2010 or 11, maybe. Maybe it’s earlier, I can’t remember exactly, maybe it’s 2012… Somewhere in that 2010 to 2012 range. I said, “I need some authors.” he said “Okay, you need to read Brandon Sanderson and Patrick Rothfuss.” Oh, okay, I’ve heard his name but I don’t know anything about him. The Name of the Wind, great, I’m reading and I’m going, “Man, the way he writes about characters and stories I could read about someone brushing their teeth for ten pages.” To me he’s an amazing writer. And so I go to a Child’s Play charity dinner and I kind of know some of the Penny Arcade guys, Mike and Jerry a little bit. And I come over and say hi to Jerry, Jerry says, “Hi! Oh here, meet Patrick Rothfuss.” And I go, “Oh, I just read your book, I think it’s amazing! Really nice to meet you!” And we talked for a little while, he goes, “What do you do?” I go, “I make games.” he goes, “Which games?” And I said, “Oh, Defense Grid.” and he says “Defense Grid! I love Defense Grid and I can’t play any more of it or I won’t be able to write anymore.” I’m like, this is my first “I love your stuff, you love my stuff” experience. I’ve never had that before, ever, really.
And so, what ends up happening is, that was fun, we talk a little bit, we emailed about some different things. Then a year later I see him and I say, “Hey, we got the funding, we’re going to move forward with Defense Grid 2, would you be interested in writing for it?” And he said, “You don’t want me. I don’t make deadlines, I write and rewrite and rewrite, and that’s why people like my writing, but it probably won’t work in a collaboration. You want Mary.” I’m like, “Mary? What do you mean?” he goes, “Well, when Twitter first started, I decided to get six different twitter handles. I told my followers, one of them is me, but the others are other writers. It’s up to you to figure out who’s the real Patrick Rothfuss.” For three months or something they all tweeted about stuff that Patrick would tweet about, and at the end they all voted on who was Patrick, and they all chose Mary.” So she impersonated him quite well on Twitter, as it turns out. She also writes her own books and her own short stories, Mary-Robinette Kowal, she’s a Hugo award winning writer, and he introduced us to her and we hit it off and she ended up writing for Defense Grid 2, she’s written for Brass Tactics, she’s done a lot of writing with us. That came about completely because Jerry introduced me to Patrick who introduced us to Mary.
FD: That’s pretty cool. I felt that was a collaboration that really added a lot to DG2. Those text blurbs on the loading screens really added a lot of depth to the world you were trying to build. That worked out really well.
JP: Thank you, yeah! We basically had these two storylines. The text blurbs were all completely written by Mary, and Sam and Jim who had worked with us before on Defense Grid before they got Haven, they ended up being the show-runners for Haven, had done the dialog up to that point. Then when we wanted to do the special edition radio play and when we did the expansion for VR, Mary ended up writing all of that.
FD: The DG2 Radio Play, A Matter of Endurance that was part of the special edition was a really fun addition. It was a separate install, though. Did this turn out to be a problem, or did most people find it?
JP: Most people seemed to find it ok. There were a few support questions asking about Matter of Endurance and how to run it, but typically the Steam installer would add both the full Defense Grid 2 game and the A Matter of Endurance radio play app into your Steam library if you bought the Special Edition and at that time at least, Steam libraries weren’t quite so full that people couldn’t find it. That may be different today. We also had links on the Defense Grid 2 Launcher to run the Matter of Endurance radio play, to access either of the art books, or to access Russ Pitt’s Making Defense Grid book that all came with the Special Edition as well.
Matter of Endurance was originally planned to be a comic we were going to create and give to players along with the game. We approached a few popular comic artists, talked with them about drawing it, and had talked with Mary Robinette Kowal about writing the comic. We had an idea of filling in the gap between the end of the last Defense Grid 1 expansion pack story and the beginning of where we pick up Defense Grid 2. As we started locking down the price for getting the comic created, we realized it was going to be too expensive for the budget we had set aside for additional Special Edition content, so unfortunately we had to leave the comic idea behind. While we were talking about alternatives, Mary, I think, suggested that maybe we could do the same story with the voice actors, add sound effects like a radio play of old, and play that for people. We loved the idea and were able to make that cost work. We also got to feature a lot of existing concept art while the audio play was playing, and our internal concept artist John Thacker also made some new paintings specifically for the story. That led to the final result we have today for Matter of Endurance.
FD: Tell me a bit about the VR Edition. How did that end up coming about?
JP: A year before, 2014 or so, we had reached out to Oculus, we knew a bunch of people who had joined Oculus, so we said, “Hey, we think VR is pretty cool, we signed up for your DK1, we’d love to make some stuff with you, do demos or something.” We sat down with them, had a meeting and we talked about the different things we could do, and basically it never went anywhere. They either weren’t ready or didn’t like what we’d suggested, anything. So, it just kind of died. This happens a lot in the games industry, lots of meetings, lots of happy people, never goes anywhere. Probably the same in the writing world, I’d guess. But then about nine months later I got an email a week before GDC saying “Hey, can you meet with us at the Games Developers Conference?” Sure, happy to meet. So we go in there and we sit down and we, hi, new people, I had not met these people before, and they said, “We think your game would be really good for VR.” And this is 2015, to me when people said “your game” in 2015 it probably meant Counter-Strike. So I replied, “No, actually I don’t think Counter-Strike would be very good in VR. I think it’d be really nausea-inducing, and in fact it’s Valve’s game, not our game, I’m not sure this would work because they’re doing Vive and you’re doing Rift….”
And they go, “Not Counter-Strike, Defense Grid.” And I just sat there and looked at them like, “Defense Grid? In VR? I… I… I don’t get it.” To be completely honest with you, and I love Defense Grid, I love it tremendously. And they’re like, “Well, here’s why your game might work well in VR. There are people who love rollercoaster sims and motion and stuff like that, some people love that, but a lot of people hate that. Some people like being immersed in first-person environments, but other people don’t like that as much. But we found that almost everybody likes if we put them in front of it, is a little tabletop, kind of dollhouse simulation where you’re basically a god and you’re messing with stuff. Everyone seems to feel really comfortable with that and they like that. When we started thinking of what would be fun from that perspective, we thought that your game, Defense Grid laid out on a tabletop and you can mess with the aliens, that would be really cool.” I went, “Oh, I can kind of see that.” So they said “Why don’t we do a little demo, put the VR camera in there, let you control it a little bit and see how it works.”
So we did that with them, for a month or so, a quick demo and we were shocked. We were like, “Oh my goodness, this is so cool!” We’re seeing our own game in ways we’ve never seen it before. My situational awareness of what’s going on in the space is dramatically different then when I jut scroll the screen. And my desire for what I want out of this environment is completely different. The ability to look around plus manage the game and everything is just really different. So Oculus said, “Yeah, this is really cool.” And they financed us to basically make a VR version and try to make a premium VR version. And you know, we were 30 frames-per-second on console which is basically a single eye, and we needed to optimize the game to become 90 frames-per-second for two eyes. Two different cameras, and we wanted to make the graphics better, and we wanted to fill out the levels, so it became a really big project, that ironically didn’t change the game dramatically but changed the experience very dramatically.
Then we said to Oculus, “You know what would be really cool, is if you let us make an expansion for this.” Because up to this point, the sales for Steam and console just weren’t quite enough to really justify an expansion. At least not for the quality level that we wanted to make the expansion. So Oculus said, “Sure, we’ll pay for an expansion, but it’s got to be VR exclusive.”
So we go, “Okay.” so we added an expansion to it as well. We have the new experience, we adjusted the levels to work in VR. We also created, believe it or not, this was during development, it’s be really fun to do a hidden-object game. So we created these Gold Cores, I think there were five per level that you have to find, and that turned out to be, especially when we did press interviews, no one wanted to play the game they just wanted to find the Gold Cores. It was bizarre. So we updated the levels, updated the graphics, made it all work with VR. Initially it was controller because that’s what Rift shipped with, later we updated it to support Oculus Touch, so you could point and pull and do stuff, we did some nice things with the Touch controls. We added the expansion and really improved the performance of the game dramatically. It shipped on Rift and it’s done very well in the VR market.
FD: How much re-tooling was needed to get that game going in VR?
JP: It was a team of 5 to 10 people for a year. It was not a small project in any way.
FD: Have we seen the last of Defense Grid?
JP: I actually don’t think we’ve seen the last of it. I think there’s a possibility of more. But I’m not sure what form that’s going to take quite yet.
I’d like to thank Jeff for sitting down with me to talk about the game leading up to its anniversary. It was a lot of fun to sit down and take a look back at what worked and what didn’t. Hopefully, we’ll be seeing more of the series, soon. I’d also like to thank James Carter (email@example.com) for the banner image, and for helping to push my scores ever higher. Both in the game, and in life.