Reimagining a Classic: The Design Challenges of Deus Ex: Human Revolution

Actually, before we start, uh they want me to remind you that they're going to send an evaluation form about 10 minutes before the end of the talk, so you should check your spam filters just to make sure you get it. And also, if there's time for questions at the end, which I'm not sure about yet, uh please use the microphones. Uh there's one there, one on the other side. And we're good. And maybe if you could, just for me, just maybe compress in the middle just a bit, just so it's more intimate. If you want. You don't have to. But as a favor, I would really like it.

Okay, let's start. Welcome. Good morning. My name is Frank. I work at Eidos Montreal. I was a designer on Deus Ex: Human Revolution. Right now, I'm a gameplay director. Uh Deus Ex was uh a big project for me. Uh I've been in the industry for 12 years now. And uh when I started on Deus Ex, I had been in the industry for only seven. So, I had a lot to learn. It was a huge endeavor, but it's it was the the best project for me. It really, really hit home. Uh It it I'm I'm a bit emotional about it, actually, because everything went so right for such a big project. And what I want to talk to you about today is what this What are the steps that we went through to make this game and bring back the franchise and also get the feeling of what Deus Ex

game is. So, let's start with the boring stuff. Me. Who am I? I started in the industry in 2000. I said that already, so let's move on. I I worked at Ubisoft Montreal for 5 I joined Eidos in 2007. I have a cat. Woah! A quick cat. Hello. And I still have trouble saying Deus Ex or Deus Ex or whatever it is. Feel free to correct me. And I have a weird fascination for David Hasselhoff. Yay! Okay, my shipped games. And my role on DXHR. I was a senior game designer. A general pain in the neck for animators, programmers, and also level

designers. This is what you heard often when I came in. I was responsible for augmentations, conversations, cover system, hacking, uh combat AI, stealth AI, balancing, and objects and pickups. Now, before you you assume that I was the only designer on the project, know that there was three of us. That's Tony in the foreground and Pierre-François in the background. And no, he's not that short, he's just leaning against the wall. Um the point I want to make before I continue with the presentation is the fact that there was a lot a lot of work to do on this game. A lot of systems that had to work together and there was only three of us to do everything. And honestly, I wouldn't have it any other way. Uh having only three designers meant that we got to touch everything. Even though I started hacking, Tony was the one to actually finish it and do all the subsystems the subnets, sorry. And it allowed us to have a more

cohesive game design. So, instead of having one designer working on something on his own side and then another designer at the end you bring together. Everybody touch everything. We knew what everybody else was doing on the project. So, what am I going to talk about today? I'm going to talk about the challenges we faced capturing the feeling of what a Deus Ex game. I want to also talk about about the conversation boss fights, how we designed them. Hacking, the upgrade system, energy system, the boss fights. Yeah, I got a few explanations for you on that. And balancing. Okay, so what is Deus Ex? Does anybody here that who doesn't know what Deus Ex is? Well, I got an explanation for you anyway. It's an action role-playing game combining elements of first-person shooters and role-playing game. It's also often praised as the best PC

game of all time. There was a sequel, Invisible War. It did pretty good as you can see from the rating. But when you go on the net and you see what some people think about it, they don't always agree. Uh there there was also another planned sequel called at the time it wasn't called Project Snowblind, but they decided to rename the the project during production. I don't really know why. If anybody here worked on Project Snowblind, I'd love to talk to you and actually know what went on. I just know it started as a sequel to Deus Ex and then they renamed it and shipped it. And there was another sequel which nobody here knows about. Uh it never went past the game uh the design document stage. Uh it's a pity though because they had the pretty good ideas in there.

So essentially we were tasked into doing what is the third attempt as this added sequel. So we started with the first two games, design documents for Invisible War and Insurrection. And that's pretty much it. We had no formal direction set by the company. Essentially they gave us the key to the franchise. They didn't even send a producer overseas to watch over us. It was only us, bunch of guys in Montreal, trying to figure out what to do, how to do a new Deus Ex game. Anybody from Ubisoft here? No? So we knew that getting the the feeling of what the Deus Ex here was was crucial. Just get

Yeah. Oh, sorry. So yeah, we we knew that getting the feeling was crucial and we knew it was our biggest challenge. And we had absolutely no idea how to do it. We knew that other games had tried. So obviously it wasn't easy. So we prepared for the big dive. This is what we we did to get our a feel for the game. So there was a creative core comprised of myself, uh the uh there was a production designer Jonathan, there was JF the creative director, and there was David our producer. And we just started to get immersed in all kinds of materials. The idea was for the creative core to bathe in all the same references so that we would have a kind of a a basis for talking about ideas. So, nobody would come out of left field with stuff. We pretty we had a self-contained idea of where we

were going. So, we started reading books and playing games and watching movies, checking out role-playing games, and have lively discussions. And that gave us a creative father for what would come up next. So, the lesson to me here was that nothing is created from nothing. We really needed to immerse ourselves with ideas to really fill our cranium with juice that would essentially feed the rest of the process. We also replayed the first game. Uh I remember that while playing the game, I had played it way back when in 2000. I really loved it. But, uh at the time we started working on Human Revolution and I had pretty much forgotten a lot of it. So, we replayed it. I remember that when I played it, I did not take any notes. It was something that I now call an instinctive play-through. So, I I just played the game and allowed myself to enjoy it, but

also hate it. Uh to essentially place myself in the player's position. As a game designer, I tend to take a lot of notes when I play games. And sometimes you you miss the forest for the trees. You you delve into too much detail and you forget what the player is supposed to experience while playing your game. And that's why I wanted to experience Deus Ex first and then go delve in the mechanics. And so, that's what we did. We We went We took the game guides. We analyzed everything. We also went on forums. We wanted to know what other people thought about the game. Uh we felt it was important to to see if what we thought, our opinion of the game, the way we felt about it, was in line with what other people thought. Um it it was good for our self-confidence. So, just to see that uh indeed other people thought that features that we liked, they liked too, and features that

we didn't like, they didn't like. Uh it it made us think, sorry, that maybe we were the right team to bring this game after all, to make this game after all. So, we also had roundtables with uh oh, I'm sorry. We did we looked at reviews to see uh the features that people like. And then we had roundtables. We just talked with people on the team and uh see what their opinion was. And eventually a picture emerged. Uh we saw that pretty Even though some people thought that the games were perfect, and it's fine, that was their opinion, we saw that a lot of people thought, well, I I I I don't like this, or this I love a lot, and all their opinions were pretty much the same at some at some point. So, uh we saw the beginnings of a pattern, and uh we also saw that our perception was in line with most uh people out there. Had we Had that not been the case, had our opinions been totally different from what everybody else thought, then we

would have second-guessed ourselves a lot throughout production. And uh we couldn't have produced the game that we eventually did. So, the critique of Deus Ex, the the big dive brought us to this. Having uh Now that we knew what everybody thought, uh we were ready to do a formal critique of the game. So, yes, I'm about to say some negative things about the best PC game of all times. I'm sorry. Uh it's not that I want to, it's that that we had to. Um a game well, to me at least, has to come from the gut at some point. And to to really do that, to really infuse our game ourselves in the game we were going to make, we had to be really honest about how we felt about the franchise. So, um I I love Deus Ex, don't worry about that. Oh, par stands, nice. Okay, uh the critique.

This is a critique for the action sequence in the game. I'm going to let you read it just a bit. I don't want to go through everything. If there's Oh, my mic fell off. I just want to let you know that the slides will be on the GDC website. So, if there's Can you hear me? Yeah? Okay. So, if you don't have time to read everything that's on there, just go to the website and you can download it. There's also comments in the the text section of the PowerPoint. So, this is uh the Yeah, this is actions, RPG, and finally, open-endedness. So, even though we were not ready to commit to design, we already had a

direction in mind. We knew that we wouldn't have universal ammo, for example. And we also knew that we probably would have a cover system. But we before we were ready to write anything on paper and say, "This is the game we were going to make." we wanted to spend time on the player experience before, because the mechanics would have to conform themselves to what we wanted players to feel. Yeah? So the core values. The core values are what permeate everything in the game. They're kind of a game design through line, a theme for us. We had four of them. It's about choice, obviously. It's Every choice has a consequence. Make it simpler. And make it spectacular and rewarding. These are not just words, they're intentions. But an intention is

worthless if you don't back it up with a bigger explanation. So we had uh a few phrases for each of them, just to get a sense of where we were going. This is the one about choice. And there's consequences. And finally, make it spectacular and rewarding. So these few words have a lot of implications because once we took them, we wrote them on sheets of paper, which we stuck to the wall, and we knew that everything in the game, in the gameplay, would have to uh

follow these four intentions. Next up, we decided on the pillars we were going to use. So these are the four pillars we set on. We could have used anything. We could have decided that augmentations would be a pillar, or that uh I don't know. Um what do I have on here? Yeah, we could have put pretty much anything, but we decided to go with systems that would drive the player experience. Also see that uh we put them in order of importance. So, there's a reason why uh combat is um there's more augmentation for combat and um uh we have uh weapons and stuff. It's this It's because we decided that combat would be where we would be put most of our time, energy, and resources, followed by stealth, hacking, and then social. Player fantasy. Uh before Mary came on

board, who was our narrative designer, uh we set on uh player fantasy just to have a context to our design. This greatly helped us to have a sense of uh when the game would take place and who the player will be. Um I also get at the question a lot, "Why a prequel?" Uh The reasons were that uh what I a lot of people thought that Invisible War was set too far in the future. They wanted a game that was closer to today's life. Also, the 2030s really excited us. When we looked at the timeline for the first Deus Ex, we saw that 2030s was the time when uh mechanical augmentations were introduced and that brought a lot of turmoil. It was a a theme that really, really excited us. And finally, we thought it would be easier for new players to get into the franchise. If you don't have to play the

other two games before, uh and you don't you just have this one to play, and afterwards uh you can go back and play the other games if you want to see how it is. References. We used references as a common language. It was uh easier for us to talk about abstracts such as mood if we had a references concrete references to to use. It was also an way for us to communicate our examples our ideas. Sorry. So this is our main reference for the game for the tone of it and then we had references for every single gameplay pillar. And for every reference we also wrote what we wanted in each like saying that fear is a reference is maybe too large. So we wanted fear for its AI for example. So yeah social tone of the game and

conspiracies. Yes. Thank you David. Before we ended conception we also had figured out all our gameplay loops. So we we pretty knew early on what kind of game we were building and for each gameplay loop we also had explanations for everything just so there were directions there were intentions. I got a lot of text here I'm going to skip through everything. I just want to show you the amount of details that we had before we left conception. We often hear about projects that go in pre-prod and they don't really know where where they're going and they they're trying to find their gameplay and we we knew before going in pre-prod exactly what we wanted to do and what each system was supposed to do in the game.

Combat and stuff. And hacking. and social. I'm sorry if I skipped through this quickly. And finally, towards the end of conception, we got what we call detailed system directions. So, for each system, we would say exactly what the the system needed to do. And that was a direction for game designers to hit. And it was also a way for the creative director to judge if a final design done during pre-prod would would pass or not. So, we knew exactly what we were doing and what we were supposed to to go with each system. This is the one for combat. And continued and one for stealth. So, already we're talking about feedback

and uh augmentations and cover. And hacking. And social. By the end of conception, everything I've shown you so far, we put on huge sheets of paper, which we papered the walls over with. And this was very important. The first reason is that we could not ignore it. If you have a design that you put you put on an Excel sheet or Word a Word document, and you just put it on the server and you forget about it, then you're not going to follow exactly what you wanted to do at the beginning. So, by having it on walls, every single morning when we came in, we saw exactly what we needed to do. And also, while working during the day, we just looked up and we could see exactly what we we needed to do.

They were so important, in fact, that when we moved to our our offices, it's the first thing we put up. and they were there until we did not need them anymore. So, I really think that having this kind of detail was crucial to bringing the gameplay that we did for Human Revolution. Uh we did not search during pre-prod to know what the game would be. We did not redo systems. Uh we did not rethink the game. We just went full speed ahead. We knew exactly what we wanted this game to be, what we wanted the player to feel, and the whole team was with us uh until the end. Now, before I go on to pre-production, I want to talk about the way we work during conception. Uh we work in a way we call the creative core, and that means that All right, fell off again.

That means that the creative director was not the one to come down from the mountain with the design and say Sorry. Okay. So, the creative director was not the the one to come down from the mountain and tell us, "This is the game I want to do. Follow my lead." Instead, everybody on the team had a say about about our ideas or what we wanted to do. And not to say that we did the design through consensus, because sometimes people had ideas and we they they weren't incorporated into the design, but it was done in such a way as to everybody had a say, and everybody was kind of equal, with JF, the creative director, being a referee. This meant that uh an idea would only survive if it was really good. And it also meant that we we got uh a sense of attachment to the game. It

was not something that was handed down to us. It was something everybody had a hand in. So, it was a lot easier for us to really pour ourselves into this game during pre-prod and production. I fully credit this as a big part of why I think the game kind of has a soul. It feels special when you play it and I think that's part of the reason for it. Pre-production. The blueprint process. Blueprint was a way for us to chart the whole game in advance. Uh in there you could see all the locations that would be visited in in which order, all the conversations that would be part of it, the choices given to the player and the consequences. I'm not going to go in detail about that because Mary DeMarre our narrative designer already did a talk at GDC Austin last year. And the talk is called building the story driven experience of Deus Ex: Human Revolution. So, if you want to know more about blueprint, I would suggest that you go to the GDC

Austin website and download the our PowerPoint. Just to show you this is uh half of the blueprint. It goes from the beginning of the game to about Montreal if I'm not mistaken. So, it was a pretty big, but it was invaluable to know exactly where everything went and to to get the sense of the flow for the overall game. Conversation system. The idea behind the conversation system was to have a relationship between the player's choice and the consequences to that choice not being explicit. Uh we wanted uh people to try and read the characters and to make a choice based on the character's personality and demeanor. Uh at that time we didn't know about L.A. Noire. Uh So, uh yeah, the idea is let's say you have a character with a with a weak

personality. But he he would comment you with anger, then you would try and smack him down, and you would know that he he would kind of come back and say, "Okay, okay, you win. You win." And there was a lot of tension on the team because a lot of people thought that the system was too murky. Uh they they told us, "How are we supposed to make good decisions as player if we don't know what's going on or if we we don't have good feedback?" And their opinions were valid, but it turns out that players really liked it because they thought it mimicked a real conversation. A real conversation is uncertain. You don't know what's going to happen, and that's kind of what we tried to do with this. Uh the conversations were really costly animation-wise and also because of recording. So, we didn't want to just go ahead, write a conversation, and implement it. We needed to test it first, and that's why we used an Excel sheet

with macros to to do the full gameplay before we did any code or any animations. You can see in the bottom left corner, the boxes there is essentially the gameplay. It was a simple rock-paper-scissors thing where the player had three choices of attack, and the NPC would select at random a counter. And you you needed to to guess which was the correct answer. Is something wrong? Okay. Okay. I'm sorry. I'm going to take this [ __ ] off. I can feel it in my nose. Now I'll it. Okay. Yeah, it's better. Good. So yeah, attacks and count Oh, a bit loud though.

It acts and counters. So yes, um the gameplay would check whether you had one attack that was weak against a counter, uh strong against another one, and neutral against the last one. And then the rest of the gameplay was uh the writers attempting to steer you to the correct answer without giving away the farm. Which wasn't easy, but the Excel sheet really helped us. Uh this is a conversation pane. What we did is we would uh put the conversation in Excel, and we would take the file and send it to everybody on the dev team. And the conversation has a timer, so it in evolve in a real time. You would have This is the conversation with Wayne from Detroit's police station. And you would have Wayne actually talk to you in real time with each line appearing. And you would select one of the three choices, and at the end you would get a pop-up that would tell you if you won the conversation or not. In the background, the Excel sheet would check all the players' answers. And then

the players would send us back the file uh with comments. They would tell us They would tell us if they liked the conversation, if they understood anything. Uh sometimes they didn't, so we had to rewrite stuff. And then we would do a tally of all the answers we got. We waited until we got about 40 answers um just to to make sure that we didn't base our decisions on uh not enough data. And then we would tally everything and see trends. We could see uh how many people won the conversation, how many lost, at which round, just to uh have a sense of which counter was in clear or which player attack was maybe too precise, because if everybody won the conversation, then they were then we had a problem. But if nobody understood it, then we also had a problem. Uh we did rewrite a whole conversation based on these uh results, and but usually it was only parts that we would rewrite, probably like a round or two. And once we uh we did the changes, we

sent the Excel sheet back, but this time to another group, often times the testers. The hacking system. Yes. Okay, that this is good cuz nobody has ever seen this. The hacking system was also started in Excel. Uh I had this vision of a high fidelity hacking where um you would really feel like you were a hacker. Uh so we prototyped it using Excel, and we tried to prototype the whole experience, not just the mechanics. So we the player was first tasked to select a port uh by which you wanted to connect to the computer. And then uh the ports would have attributes such as a lifespan. A port could maybe close after a minute or so. You also had traffic speed. The more traffic uh the longer it takes to download stuff off the computer, and also protection. So uh a port with high protection would be harder to break. And then you could go on the computer,

and you could see stuff that was on it before you actually attempted to hack it. This is the interface for the original hacking. It's not the one that's in the game right now, it's the our first attempt. And uh you can see on the left uh the boxes are the integrity. The idea was that the player would use programs to attack a network's integrity while the network was trying to see if there was anybody there. And if once you were detected, it would try to trace you, and then send reinforcements in the the form of guards. And at the bottom we have the the three programs that the player had. And uh the gameplay was about taking CPU cycles from your computer and assigning them to each of the three programs. You had the attack program that would bring down the integrity. You also had the stealth program that would prevent the network from detecting you. And finally, defend would make the trace longer. And you had to assign these cycles. So, if you put all your cycles in attack, the integrity

would go down really fast, but then it would be super easy for the network to detect you. That's not all. The player was also supposed to look at a map because we wanted hacking to occur in the world. So, we designed this map with You can see the red dot over there is a guard, and you're the the green square. And we had this guard just running around along in the corridor, but at some point he would come in the room. So, while you were hacking, there was a button on in the Excel sheet where you could go to the map and see where the guard is. And once he came into the room, you had buttons to change location. Uh so, this was to see if players would be able to hack while also looking around their environment. And finally, once you managed to hack the system, you had access to everything that was on it, but you had to download it first. So, you were waiting for stuff to download while there was nothing to

do. And this is the result. Nice email. And of course, David is totally right. This was way too complicated. This is one of the the pains that I used to try and configure the gameplay. This is the other one. At some point it was so complicated that I didn't even know what I was doing. The problem with the system is we ask too much of the player. Uh uh having CPU cycles was too abstract. Having to look at the map was just too much. And then asking the player to wait to download stuff, people they they just couldn't handle that. So, but I I really wanted that system. I was really fighting for it. And JF, my

boss, he had me redesign it three times, but still it wasn't enough. Uh I was sure at some point I would finally crack the code, but I guess it never came. This is uh yeah, you could even upgrade your uh everything. You could upgrade your CPU cycles, and you could upgrade your programs. And the zombie thing, uh once you hacked the computer, you could put a rootkit on it. And then when you went when you went to hack another computer, you could use all the computers you hacked before with CPU cycles to make your hacking even uh easier. So, I was really, really hopeful this would work. I really wanted to play that game. Yeah. Okay. This is a concept art done by Matthew, our interface designer at the time. He came up with this simple idea. What if we make hacking visual? I was livid. I hated that guy for a week.

But he was right. And at some point JF looked at me, and he just he had to grab me from the ledge. I was ready to jump. And he said, "Look, this is not working out. I know where you're going, but it's just too complicated. This will work. This is the direction." So, it took some time uh took some time for me to get around to it. But eventually I saw that by taking the the mechanics we had in the cell sheet and just porting them to this kind of display, like the idea of uh taking over a network by uh slowly degrading integrity is how essentially you take nodes in the hacking today. Uh the idea that you could defend stuff is also part of the hacking. So, all the mechanics I had done in the Excel sheet turned out to work perfectly with this. So, once I saw that, it took 2 weeks to just port everything and we had our new gameplay. Uh it went so well, in fact, that um

hacking is the first uh game system that was feature complete before uh every anything else in the game. I think 2 years before we shipped, hacking was done and we were uh playing it. So, I'm pretty happy with it now. I wasn't at the time, though. Energy system. Yeah, this one This is a good one, too. Uh for a long time during the game, we did not have an energy system. Why? Because uh we found it frustrating in the first game. We remember that while playing it, we were always lacking energy. We couldn't use our augmentations. The bar was always empty. So, we thought uh we're genius. We're not going to put energy in the game. You use all your augmentations whenever you want. But then the takedowns came and we saw that players would do takedowns all the times. There was no restrictions for them. Uh they weren't even using weapons anymore. They would just go in stealth and do take and and everybody would fall down. So,

we we looked at each other and we said, "Well, okay. Now we understand why there was an energy system." So, we brought it back. But we decided to use cells instead, instead of a large bar, because we wanted the first cell to always regenerate. That way the player would always have energy to use. But still people were complaining, "There's not enough energy." So, Jeff would tell me, "Put more nutrient bars in the game. Put more nutrient bars in the game." So, I did. But still people were complaining. But, at some point I was resisting it. Uh and the reason is that Deus Ex is a game about scarcity of resources. If we give you everything, you're not going to think anymore. You're just going to go and use the simple route. If you have all the energy in the world, you're going to use your stealth augmentation, your cloaking all the way through the game, or you're going to do like takedowns like people were used to do. That's why we still have people complaining today there's not enough

nutrients in the game. Well, if you're one of those, I'm the guy to blame. I'm sorry. But, it I think it was really crucial to how the game turned out. Um otherwise, um it it wouldn't be a Deus Ex game. You would just plow through and would be an action game. And I find it kind of ironic that the the exact things that I didn't like in the first Deus Ex is what is something that we essentially brought back in our game. So, irony, I love it. Production. Boss fights. And I'm not sure I do. Sorry, David. Uh the problem with boss fights, how can I say that? Um remember the direction sheets I was talking about at the beginning? We didn't have one for boss fights. We kind of forgot about it. So, just it it Well, it shows how important

they were because because we didn't have these direction sheets, we didn't really know what we were doing with boss fights. Uh for us, we saw them as a way to break the pacing more than a way to test the player's skills. That's a big factor why uh you cannot go through them by with you by using social or stealth. Uh we also waited until the middle of production before we actually tackled them. There was no design during pre-prod. Uh it's during production that we we started design on them. So, we were also pressed by time. But, we thought that by putting enough ammo and enough weapons in the room, that would be enough for players to just defeat them and say, "Okay, then that was nice. Let's go on." Uh and uh play tests uh did flag the the boss fights as a problem, but they did not flag the severity of it. It's only once we ship and we saw the complaint that we fully understood, but it was a surprise to us,

actually. Uh a big surprise and not a good one, but uh So, yeah, I don't really have a solution regarding boss fights, except that next time we're going to think about it more and we're going to put uh They were a big part of the game and we should have put more efforts in them. So, truly sorry about that. The upgrade system. At the beginning, we had this idea that to upgrade your character, you needed to go to a limb clinic. Uh Adam didn't have all the augmentations. Stuff like uh Cassie, the social log, or the the cloaking stuff was all at a limb clinic and you had to physically go there and buy these augmentations by using money. We thought it would be uh more fun for players and it would be more realistic. Unfortunately, we found out that that's not at all how players want to play the game. What we saw through play tests was

that players wanted to come up to a challenge and be able to upgrade right there on the spot. So, they saw a big crate or uh an impassable stealth challenge and they they they said, "Well, I want the augmentation to move that crate or I want the augmentation to cloak myself. And by asking them to go to a limb clinic, we were essentially forcing them to anticipate their needs before they actually needed anything. And that wasn't fun. It was counterintuitive. So, we we said eventually decided to uh nix that design, and we put all the augmentations on Adam at the beginning of the game, and we allowed players to unlock stuff at their own rhythm. But, of course, that broke other stuff. Like, the first conversation in uh the Detroit manufacturing plant with Zeke Sanders was never designed to use Cassie, the social log, because we knew that the first uh limb clinic was in Detroit, which was afterwards. So, there was no

way for the player to get the augmentation. Now, it was possible. So, we needed to go through the first parts of the game, the whole manufacturing plant, and balance it in a way where you wouldn't get enough experience points to unlock Cassie before you uh arrived at the Zeke conversation, because Zeke was already done. Everything was recorded. Everything was animated. There was no way for us to just go back and add the Cassie support. And I'm going to tell you trying to balance the first level just so you wouldn't get Cassie first was a big big big challenge and a big problem. And Mary, our narrative designer, wasn't happy because she had to rewrite the conversation with Vera Markovik in the uh the first limb clinic for the third time. Balancing and debug. How did we balance this game?

The first thing we decided upon was to balance the critical path independently from the exploration path. So, the first thing is we made sure that on the critical path, if you did not explore, you would get get enough stuff to get through the game. And everything else was gravy. Uh we use these balance boards to get a sense of everything that was in the game. This is about half the boards we had. We had six in total. Uh and the reason we used boards is because the metric system wasn't working. We had one, but it broke down and we didn't have time to fix it before uh we had to finish the game and finish the balancing. Uh I did the first pass at balancing uses using just an Excel sheet that was that rested on my computer. And if I remember correctly, it was a total failure. There was about 10 times too much ammo in the game. Uh it was hard to balance the game without having something visual, something that would give me a big picture of what went on.

So, after doing a first pass that was a failure, play test showed us that people were just gunning everything. You know, they they weren't thinking. It was just a huge frag fest. Uh so, we use these boards. And then again, because we didn't have metrics, the only way to actually know what was in the game was to go in the editor, open every single level, every single inventory, every single shop, and then pin stuff to the wall. And uh but doing that, we knew exactly then how much ammo was in the game, how much experience points, uh how much well, pretty much everything, grenades, and uh uh and whatever practice kits. And we could also see trends. We could see where we were giving out too much experience. We could visually see where do we give the laser rifle for the first time. Or um what's the featured weapon in this map or this other map. So, it was a really really useful tool. And it's only after

we we did the balance boards that uh the game really started to take shape in terms of balancing. The other thing I learned about balancing this game was that uh as I said, we put in too much ammo at the beginning and then we drastically reduced everything. We we removed a a lot of stuff. And uh during play test, people would complain. They would say, "Well, I I don't have enough ammo to kill everybody." Uh but then again, that's exactly what we wanted. The first time people went into the manufacturing plant and they they emptied their combat rifle during the first combat, they understood at that time it's actually what the game was about and that it wasn't a game for you to just run and gun. It was a game for you to hide and think and check out your inventory and your augmentations to find out um to actually explore and find other ways to get through the challenges and that's where the fun lies. At the beginning, when there was a lot of ammo, people thought, "Well, the game is fun, but

it's okay, nothing much." But as soon as we removed and we toned it down a lot, the fun factor went through the roof instantly. People were liking it. Yeah, they were bitching at the beginning, but that's because they did not understand the game. They they were new to the franchise. But afterwards, once they got through the first part of the level, they really liked it. They they really thought, "This is new. This This is something. There's strategy behind it. I I got to think." And thank you uh for for that. They were really happy about it. Conclusion. So, in conclusion uh what I learned during Human Revolution is that there is no magic recipe. It takes a lot of work. It also takes people that you like working with. We were a tight-knit group and we still we still are. We we have confidence in each other and we like working together and I think that shows in the final in the final game. You also got to follow your gut.

You you got to pour yourself in in what you're doing. Uh we lived and breathed the their sex for a long time, a lot of years. Also got to face the truth, you're not perfect. We made a lot of mistakes and people would tell us this is not working and we we had to listen to them and we had to to correct these mistakes. It's we could have decided no, no, we don't want to hear anything, but that wouldn't have been a problem. The game wouldn't be as good as it is right now. It's because we listened to people. It's because everybody had a say. Everybody was kind of equal in the team. And you also got to question yourself. You you really got to look at what you're doing. Like hacking, I wasn't questioning myself. I had a vision. There's I was I really wanted to create something. Had an agenda. And that was a problem. I wasn't thinking about the overall game. I was thinking about me being finally a

designer that creates this totally awesome hacking gameplay. And it turned out that it was my it was a good lesson for me actually. And finally, the last thing I learned was work for company that has trust in you. Deus Ex was a huge endeavor. It was very complex to do, but still the company was always behind us. And I think it was a very important because if we had somebody constantly second-guessing everything we did or just coming in and saying, "I'm not sure about this and I'm not sure about that." We we wouldn't have been able to do the game that we did. So, I thank you very much. Thank you for coming. If there are any questions, please uh use the mics.

Thanks. Hi. So, I was interested in hearing what you guys how you guys approached the cameras in the game for Deus Ex: Human Revolution because in the original Deus Ex, I used them a lot as tools during my gameplay uh to point and look at where I was going to see some of my objectives Yeah. and to uh to get a layout of how the guards were moving around. Um and in Human Revolution, I found them to mainly be obstacles more than tools for the gameplay. I would turn them off so I wouldn't have to worry about them catching me. But, I didn't really have any opportunities to actually use them to kind of scout ahead to much and uh see what I was getting myself into after I hacked them. Okay. Uh Yeah, we did see the cameras more as an obstacle than a a way for you to to see in front. Uh we wanted originally the cameras to be controllable by the player. And it's not something we had time to do. And um again, we're not perfect. We had a lot of stuff to do and um

Yeah, I don't think I have a a good answer for you. It's something we should probably look at though if there are any other sequels. Absolutely, yeah. Next question? Um hi. So, um Human Revolution is sort of unique in that it's um was sort of accepted by mostly accepted by a rather vociferous former fan base and that's not something that happens a lot. Um and some people have also pointed out that it it follows the original Deus Ex uh almost act-for-act story beat for story beat. I'm wondering how much you feel that contributed to the acceptance of it, whether you think it's a prerequisite for it being accepted by the fan base and if not, you know, do you think it's now that you established yourself as demonstrating that you understand it, do you think it's safe to move away from that? Uh Well, first off first off I'd like to react to your comment that the game

follows the first game beat by beat. It might be true. It wasn't our intention. It's not how we designed the game. It's not how it was written. So if that's the case, it's a coincidence. As for the fan base and the reason why we got accepted by them, it's because we did listen to them. When we went on forums, they were the guys that we that we we were listening to. We also got people from the community come to see us before the game was finished just to see what they thought. So to answer your question, I think the reason why we were accepted is because we first off we love the game. We tried to do the best job we could. We really infused ourselves in that game. Uh it it came from the gut. It came from the soul. And secondly, we we also knew we had a big responsibility. Uh in one of the slides at the beginning in the directions stuff, we said that we it wasn't our intention to dumb it down and we knew it was a big fear of people.

So we were very aware of that and it's just keeping keeping reminding ourselves of what people expect from the game and I think that was the secret to it. Thank you. Yes. Hi. Um Well, first off I just want to say that I really appreciated the game. I thought it was a great addition to the to the canon. Thank you. You know, it's a subject matter that's very near and dear to my heart. So I have two two questions. One at what point in the development of the game did you guys start thinking about multiple endings? Or like how did you go through the process of thinking about the multiple endings in the game? And then I also was just curious about how behind the scenes the relationship with Mary like how the like writing informed game play and gaming game play design and foreign writing like your thoughts on all of that.

Okay. Can you repeat the first question, please? At what process like what was at what point did you guys think about the endings? Yeah. The process for the endings at the beginning we knew we wanted multiple endings. That was a given for the franchise but actually thinking about the endings we did during production. There was two parts to the ending. The first was well as you all know the buttons you had to press. The second part was Adam's voice over that is driven by decisions that you made before in the game. We were looking at stats like how many how many people you killed, how many conversations you won and that would drive some of the dialogue. We did not think the whole thing through at the beginning. It came pretty late which explains a bit why it was implemented the way that it was. Does that answer your question? Thanks. Yes. I had a question about balancing. You spoke about balancing the critical path first and then balancing all the

exploration stuff. Yes. I've had problems in the past with systems that I've built where I balance for the critical path and then and then the exploratory stuff and players who do all of it excel way too fast and and sort of break the game. What did you guys do to prevent that and if you could talk about your experiences it'd be really helpful. I would say that the secret is play tests and playing the game yourself afterwards. We did have problem at the beginning where the exploration path would give you way too much so we had to tone it down. I'm not sure there's a secret answer to to to what your to your concerns. Um, when we just going to remove myself being naked. Thank you. Thank you. when we balance the game, uh, I think what we did actually was that we said ammo would more be mostly on the critical path because ammo was

our biggest problem. We didn't want the game to turn into a shooter. So, we very carefully decided where we would put it. And then the exploration part portion would be more about finding a few practice kits or finding medical stuff or nutrients. So, it would be stuff that would be helpful but not too much. It wouldn't break the game. Had we put a lot more ammo for instance in the exploration path, yes, we would have had a a big problem. Thank you. No problem. Hey, um, in one of the earlier slides you mentioned that your core or your core creative team immerse themselves in literature and games and stuff to sort of get a feel for it. Um, as someone who's read like six or eight cyberpunk novels, I'm quickly running out of them. Do you happen to remember any of them? Uh, so, you want cyberpunk novels? Yes. Well, I would tell you that actually what we looked at wasn't necessarily cyberpunk.

Um, if you take a look at the references when you download the slides, you're going to see we went really broad. We we looked at we didn't want the game to be informed solely by cyberpunk. We wanted something we want we wanted a feeling and we wanted themes from other stuff, too. So, I I did read Gibson for example way back when and I I was a shadow master at Shadowrun for many years. But that's not enough. We I don't remember exactly what kinds of movies there was there was Children of Men I think as an example, but anything really anything you that feels remotely like it could help you go and read it and watch it and just expand your horizons. What I would tell you. Thanks. Hello. As you mentioned that you don't doesn't have plan to the boss fights, right?

Have you figured about your take off then? And well, how how is the process to balancing the the boss fights because people who is who use the stealth mode has a lot of problems to to deal with it. Yes. Well, when we balance them, we did have play test so we would watch people actually play them and we never saw that big of a problem. That's why it was such a surprise for us when we read on forums and we read in reviews that people were getting stuck and still we get mails by from people who would say I'm I'm not able to go through this boss fight or this other boss fight. And unfortunately, it's not that easy for us to just patch the game and remove the boss fights or change them. I'm not sure I have a good answer for you about that. It just something that happened and we as a design team we we actually were surprised. Unfortunately, yeah.

Hi, 5 minutes? Okay. Yep. So I just wanted to know um how much of the feedback did you incorporate from the forums when you when you read them such as like the highlighting decision to remove make highlighting optional? About the conversation system? Well, like you know how you had the object highlighting and there was a thread on the forums that really wanted that to be optional so you could have that and there was a lot of other things that the like just the uh hardcore like Deus Ex 1 fan base that were that were quite opposed to. Um Yeah. Like what So, what did you like how much did you um just incorporate there from the start? the highlighting, at the beginning it was mandatory and we decided to make it optional uh from comments we got from forums from uh from the fans. Uh the problem is uh when you start communicating about a game, when you start having demos and stuff where people can actually see what the game

is, it's pretty late in production. It's only when you do have stuff to show, but by that time it's also harder for us to change the game. If people had wanted a third-person shooter, it would have been impossible for us to do it, for example. But smaller stuff like putting in options, uh if we could, we would do it. It wouldn't even be a question. It was mostly a question of production time. Yeah. Thank you. But we were really listening to people. If uh people on forums uh thought we weren't there, we weren't listening, we just weren't replying, but we were there reading all the time. Thanks. Hi. I'm sorry, I just wanted to uh thank you for coming out today and congratulate you and your studio team for coming out of the gate with such an awesome game. It was personally one of my favorites of last year and I'll probably the only one in the room, but I didn't really notice the boss fights weren't that great until people started complaining. I had a great time with that game. Um you had uh a ton of writers working on this project. What was that like? I mean, between the Icarus Effect novel, the whole comic series, I got completely

sucked into that. And the amount of emails in that game, I'll be honest with you, I like tried for the first like 2 hours to like read them and then after a while I was just like, "Okay, okay." like collecting them and not even bother reading them. Was that something that you uh intentionally want to overdo for people that really wanted to get into the sort of the whole side story and stuff or is that something that you're going to maybe tame back in I assume the next one, maybe in in I cannot talk about the next one or whatever is going to happen in the future, but I can tell you the reason why we had so many emails, so many books is because there's a lot of story to tell and we wanted to the world to to feel alive, to feel that there was something else going on aside from your own quest. And NPCs and conversation is one way to do it, but having all the stuff for you to find and explore was also to to kind of immerse yourself, to to show you that the world is bigger than you are, but you are still a big part of it. Okay. And uh the writers were great. They they outputted a lot a lot of writing and they also

uh had the help of the LDs. So, it wasn't just writers, it was also LDs helping out from time to time writing uh writing stuff and uh so it was a big team effort. I was just going to ask that. Did I answer your question? Yeah, that's great. Thanks. Thanks. I'll have to be quick because uh there's some people behind me, but uh you're talking about how you balanced each level by listing the amount of ammunition and on the slide you showed how the premium pack ammunition was listed there as well. So, I was wondering if you could talk about really quickly about uh how you balanced for the optional uh pre-purchased DLC that you knew that not every player would have. How we uh balanced for the DLC? Well, we we just treated it as a Got to wrap up. Okay. Well, uh we we treated it as a a part of the game. Uh so, uh we knew we were introducing weapons, so we made sure that there was enough ammo for them, but it still was a matter of balancing for the critical path and for the exploration path separately. So, we did not do anything

different for them. Thank you. Thank you. So, that's it. Uh thank you all for coming. It's been my pleasure and uh hope to see you around the GDC.