- [eerie sounds]

- People don't realize how hard these things were, because today it's just like, you throw things up there and you complain if the engine's not fast enough. That's not how it worked back then. Back then, every single pixel was your fault. 

[electronic music] I'm Louis Castle, co-founder of Westwood Studios, the executive producer, art director, and technical director for Blade Runner, the game. 

[tense music] For Blade Runner the game to get to the same emotional response of the movie, we had to invent all new technologies and treat this medium differently than anybody had done before. In 1995, when Blade Runner was designed, the games that were out were think games like Monkey Island, or games like King's Quest, which were 2D adventure games, or you had things like Doom, which were kind of really pixelated looking corridors that you could shoot people in, first-person shooters. There really wasn't any kind of open-world adventure games around. And this was the very first one. So effectively, what we had to do with the game was make a living movie. And that's where we started from, which is here's these cut scenes that look very close to the film, how are we gonna turn those cut scenes into interactive experiences? 

You have to understand, in 1995 there were no game engines to use, so everything was written from scratch. So for Blade Runner, we didn't have any conception of Unreal could do this or Unity could do that. It was basically here's what we want to do creatively, how are we gonna solve that? So we got to, we got the freedom, or had the freedom, to basically invent everything we needed to do to deliver that artistic vision for the game. 

The building of Blade Runner really started with the idea of how do we create the immersive environment, that was the art technical problem, and from a game design point of view, it was how are we gonna create the feeling of agency in the player, make the player be able to decide for themselves am I a replicant, am I a human? And so the way we did that was we broke it down and said every time you play the game, you're gonna have the same set of characters in the game, but we're gonna randomize some of those characters. Sometimes they're replicants and sometimes they're humans. And because of that, they would leave different kinds of clues, and because the clues were different every time, you had to actually go get the clues to try to figure out whether or not the person was a replicant or not, because shooting a human made the game much more challenging because now you were a murderer. Versus someone who just retired or aired-out a skinjob, as they said. So that made it really hard, because from a testing point of view, from a game design point of view, we were never really on solid ground. 

We could never be sure that any given scene would play out the same way, because the player, the game kept track of the player and whether they played as replicant or they played as a human, and the world would react to them based on the things they've done. So the world would behave differently if you went around shooting everything than it would if you went around trying to be very cautious. At the same time, you never really knew if the person you were talking to in the game was a human or a replicant, and their responses would be suddenly different depending on which ones they were. So all that sense of unease and uncertainty helped to create that sense from the original film where you just never knew, that tension of any time, any moment, this slow-moving methodical pace would suddenly explode in violence, and that's exactly what we created with the game. You go through each scene, clicking around, and you start to get comfortable, and all of a sudden somebody comes and starts shooting at you or comes chasing through the scene, and that sudden explosion of violence, of action, made you tense and nervous at all times, even when the scene was quite sedate. 

One of the things you could do with Blade Runner, because it was a living simulation, you could play the entire game, saving all the replicants, get to the very last scene, get on the shuttle, get ready to take off to save them and bring them to off-world colonies, pull out your guns and shoot them all. And then you got a very special ending for that where you walked out of the shuttle and your cohort Crystal Steel walks up and says, "I didn't know you had it in you, slick." 

There was definitely a point in time during Blade Runner where we felt, okay, we're gonna need more time, we're gonna need more money, and the amount of money participation we had from our partnership, the Blade Runner partnership, was capped. It was part of our deal. So we did have to go back to our owner, publishing partner and say hey, we want to spend a few hundred thousand dollars more on this game, it's an adventure game, which nobody thought could sell, so that was probably the only time where I had a real crisis, was like, oh my God, where am I gonna get this money? I don't, I can't write the kind of check that I need to write, I gotta get the money from our owner, and are they gonna buy into this, are they gonna believe this crazy idea that we're gonna sell an adventure game that's gonna sell more copies than any other adventure game could, just to break even. 

At the time, really good selling, the best selling adventure kind of games, were 100,000 plus copies. We had to imagine that this game was going to sell 300,000 or more copies, which would make it the number one selling adventure game. So the math for the business required the number one hit. Well, to sell that many copies, you also have to sell to the lowest common denominator of machine. And so all of the problems we had to support those low-end machines were absolutely required, because otherwise we could never meet the sales goal of having so many copies sold. 

It was pretty clear we had to have some new technologies. We had to have a sense of depth in the world, so we had to create a type of buffering that would run in real time. We had to create lighting, so we had to create deferred rendering and normal vectors that would actually allow us to light the scene. We had to create real time compositing that would allow the z-buffers to work together, and we had to do all of those things inside of loops that would be repeated, but also lay on top of each other so you didn't get the sense that the world kept repeating itself over and over again. That's okay for a neon sign that blinks, but it's not so good for a car that drives by, you don't want to see the same car go back and forth. 

So the idea was start with compression. So we started with the compression, we said look, we can make these beautiful scenes that are really lush and really detailed with all this lighting, and they look wonderful, and we've crunched them down to where we can fit and they look great. Then it's like, well let's animate them. And once you animate them, of course, now you end up with several seconds of animation, you have to become even better at the compression. So we had to take our compression technology and we had to actually make it more of a trilinear compression that would look through time so that we could make use of the fact that blocks were repeated over time. And this was pretty innovative when you're talking about eight or nine seconds of video. At the time, even things like mpeg would only do a couple of frames, not dozens or hundreds of frames, just a couple of frames. So we had looked ahead that was all the way out, several hundred frames later. And once we were done with all that, we had this beautiful environment that looked amazing and it's like, okay this looks like the 3D rendered stuff was better than the stuff we had done for Command and Conquer to the date, and was really great, now we're gonna put a character in it. 

And so the idea was let's go put some sprites in there. Well, they looked terrible. The idea originally was we'll just use sprites and kind of, we'll move them around and keen them and everything else like we did with our other adventure games, it clearly wasn't gonna work, as the minute you put a sprite in that environment, it just didn't look right. Then we said okay, well now we have the sprites, they don't work, what are we gonna do instead? So when we were making the characters of Blade Runner, we had these models, and I think people are familiar with models and polygons, well we had to represent them in a world that was 3D, so we used voxels. A voxel is a single graphical element that is a volume instead of a single pixel, and because it's a volume, it can be wrapped around, let's think of Minecraft, if you took Minecraft and shrunk it down, that would be voxels. Everything's made of little blocks. The voxels look good because pixel by pixel, they look very much like the scene, and the kind of blockiness you get from voxels when you go to render them back then, voxel unit by voxel unit, they look very blocky too, so they match the compression of the scene, so they felt really good. 

What was also nice about voxels is as we were going around the outside of the character, at any given time we had the normal vector of what that particular pixel was, and so we were able to actually add another layer of compression, so we didn't just have the scene, we also had the z-buffer, then on top of that we added another layer, which was a lighting buffer, sort of like a deferred renderer, that would have lighting information so that the character as we're rendering the voxels, we could actually allow light to land on the character. And that kind of sold the deal. At that point, when the character like McCoy walks into the scene and picks up all the light from the neon, and if there's a big spotlight that's hitting the water and it's splashing up with all this global illumination, the character that walked on the scene would pick up that light, and it looked very believable. 

And so we thought, okay, that's really great, we're done, this is gonna be amazing. As we started to build the game, we started to do all the sequences, we realized there was another thing that was missing. Blade Runner has all this atmosphere in it. It has all these clouds and steam. And so the problem with z-buffers is they're z-buffers, they're opaque, right. So the character would be walking into the distance and they wouldn't get affected by the steam, they'd just be stamped out over the top of the steam, so if they walked through the cloud in the distance, they would look wrong. 

So then we had to add on top of all of that another system that managed volumetrics, and so we put imposters basically, probes, inside the environment that were invisible when we rendered them, but that information in real-time was used to add effects onto the voxel characters, to make them appear to be going into the steam, to gray them out if they were into steam, or to make them go a little darker if they went into areas that were really heavy shadows. And once we did all of that, we actually even had shadows that could fall across the surface of an object, we got to the point where the characters looked so believable that they could do just about anything. And so we were finally there. 

It wasn't lost on the game development team or myself and any of them, I think, that the very things we were doing, the very criticisms we were having about replacing human actors with 3D actors, the idea that we couldn't really tell whether a person was a human or a replicant, the idea that when you're playing the game, you're not really sure if you're supposed to be a human or you're supposed to be a replicant, it was open to all this, it makes you really start to challenge and question those things, about what does it mean to be a human, and the very line from Blade Runner, more human than human, is actually accurate. If you think about the story of Blade Runner, the replicants are more humane. They're not just more human, they're more humane than the humans. And that was what Philip K. Dick was playing with in his original short story, it's certainly what Ridley Scott brought into the film, and we had to capture that as part of the game or it wouldn't have been authentic. And that was really a big part of what was exciting about making the game, and also kind of terrifying. I was like, did we get that right? 

The thing I was most afraid of is what are the fans gonna think? I would just be mortified if I had done something that they felt was a travesty for the IP and had done something that belittled it or took away from the value. I've had people actually come back to me and say, "You got it right," so I think so. [tense music]