God of War postmortem interview — How the ending and its surprise came together

GamesBeat: You build a lot of that thinking into some of the dialogue, and then you may have actors coming in and changing it.

Barlog: But that’s good. There are so many times where they just bring something that surprises me. I had to get more flexible with the idea that I’m doing 7-minute takes with six actors, and one of them is 9 years old. It’s hard to say, “Well, one actor said this one part of that take wrong.” So I automatically went into all the shoots being open to the idea of, “That was really good. A bit more of that.” But acknowledging that I can’t stitch all of this stuff together. I can do a bit here and there, but I’m looking for that perfect seven-minute take between everyone. Empowering the actors even more to own it a little bit, and giving them direction that’s a bit more focused on finding what they already know to be true, but just getting it out in the lines.

I figured out a different way to work with everybody and direct, because of that. Because literally, one camera shot in a cinematic means you can’t just say, “Oh, that’s fine, we’ll get it in the coverage. We’ll do close-ups and you can nail it there.” Everyone has to give a perfect performance, every single time. The more invested and owned they are of it, the better a performance you get. It’s amazing.

GamesBeat: Because you named the kid at the end, I do wonder whether you then felt — well, you’re committing to a certain sort of story path here. Red Dead Redemption 2 is a prequel. You know who’s going to be alive at the end of that story, because they’re alive in the first game. They had to commit to a certain storyline, where these guys survive. It almost seems constraining. Are you constraining yourself by telling people who this kid is? Is there a certain path that has to unfold in a future game?

Barlog: That’s interesting. For me, no, because I already know what I want to do with it. To me, this is part of the story. Everything that happens in this one is leading toward this vision of how we’re going to resolve all of this.

GamesBeat: Did you cut off the game at a certain point and decide that the rest is part two?

Barlog: No. It’s definitely one of these things where — everything beyond the game has the tent poles of “I want this to happen. I want that to happen.” The important thing about finding out who he is in this game is it creates that sense of, who do you think he’s going to become? Because part of it is, like you’re saying — when you know what’s going to happen in Red Dead Redemption, Red Dead Redemption 2 isn’t necessarily about getting surprised by the actual ending. It’s being taken along for the journey of the characters.

I think that’s such an amazing thing, because it’s appreciating the moments as they pass, as opposed to saying, “Oh, I called it, this was gonna happen!” So many people, when they watch a movie or play a game, “I knew 10 minutes in this was gonna happen!” That’s great. But you don’t win anything. It’s not the point of watching something, to guess what’s going to happen. Sometimes it’s so predictable that you’re saying it to point that out, but the reality is, it’s about enjoying those moments and understanding what’s happening to these characters. How are they growing?

To me, playing around with the identity is such an interesting concept. If you’re presented with two possibilities of who you are — it’s what happens to us when you go to college, right? You leave high school and you decide that you aren’t really happy with who you were in high school. You’re going to reinvent yourself in college, because nobody knows you there. You can be somebody completely different. That doesn’t last very long. Usually you end up folding right back into who you were. Perhaps slightly modified here and there.

But that’s such an interesting concept to play with. You take somebody that everybody thinks they know, because of Marvel’s interpretation of it, but that’s not necessarily the whole character.

GamesBeat: Sometimes I wonder if the thing that goes through game designers’ heads — “We have a nice story here, but how can we stretch it out 10 more hours? People don’t want a 6-hour game.” Is there ever actual pressure to make a longer experience?

Barlog: Not at Sony, not at Santa Monica. We’ve never had that. In fact, it’s usually the opposite. “Oh my gosh, too big, too much, get rid of it, we can’t do this!” I kept trying to convince people that the game was 20 hours: 10 hours for story, 10 hours of exploration. I was convincing them for a couple of years before the reality set in, that there was no way this was 20 hours. Then I was able to switch tactics and say, “I know, but we’re already invested. We understand what we’re going to do. We can do this. Let’s just be economical and find a way.”

Some of it’s a trick we play on each other. But for me, I didn’t want to make it long for the sake of being long. I’ve learned from long format storytelling, like great seasons or multiple seasons of television, that the best payoffs are these longer payoffs. I don’t feel like just learning something and then paying it off an hour later. It’s cool, but the weight isn’t there. The feeling of experiencing this entire thing and then having a realization, the weight is so much stronger. It allows you to take your time.

We, in the game industry, are far more analogous to the storytelling of TV in how we’re dragging things out and having moments. Great TV still can have an “off” episode, but that episode gives you the breather, the palate cleanser, to appreciate the “on: episode, the one after that. If you just cram it all together — it’s difficult for movies sometimes, because they want to make multiples. Here’s movie one of a five-part series! The movie doesn’t feel like it pays off, because you wait a couple of years for the next one. Whereas with Spider-Verse people are saying it’s a world in and of itself. The whole thing is an inclusive experience where I got what I wanted from beginning to end. I walk out wishing I could see more of that, but I don’t feel like I’m waiting for a payoff.

For me, that was the big thing in our game. I wanted people to be able to have a complete experience like I had when I was a kid. I’d take a game home and it wasn’t like, maybe the developer will give me a bit more later. You got it all. You got the entire experience there, and then you wondered what might happen next.

GamesBeat: Did all the Easter eggs make it out by now? Have people found them? Especially anything that would give a clue to what’s coming or not.

Barlog: I don’t think there’s anything people haven’t talked about in some fashion. Some things have only been briefly mentioned. A lot of our team, they’re avid monitors of Reddit and ResetEra and stuff. When people find a small secret, it may only be a small group of people that even acknowledges they’ve seen it.

GamesBeat: Such as the name in the house at the beginning.

Barlog: Right. We never intended on that being the last secret. That one, honestly, I thought people would find fairly early. The one I really wanted to be the final secret, the really complicated one that I spent months on, where you had the collector’s edition map and two languages that we made up, and then a bunch of coordinates — if you realized that with the collector’s edition you got two toys, in the code it would tell you where to put the toys, and then you’d follow the code and find, at the end of all this journey of moving these toys around the map — in between them is the location where you do a secret code input.

You use the two coordinates, get the two toys to that point on the map, and then you translate the two languages we made up – and you can only find out about that if you played the game – so you know what to do when you’re standing in that position. I had to keep that a secret. I used the visual development guys and a designer to implement it, and we had secrecy among everybody, so nobody knew about it. I had a special tester work on it so we didn’t let anybody know the secret.

People found it in three days. Three days after the game came out. It was maddening. That was the one I hoped would take a while to figure out. Nope.

GamesBeat: I feel like at least half of a great story is a good beginning and a good ending. If you have that, then it all pulls together as a work of art. The Last of Us has a beginning that resonates back through the ending. You guys have that, too. The funny thing, when I was talking to Neil Druckmann, he was saying that in the latest Uncharted, they didn’t know what the beginning would be. They had this sword-fighting thing and they weren’t sure if the mechanic was going to work. Eventually it did work, but they took it out of the beginning and had it in the ending. They wanted to have the symmetry in the sword-fighting at the beginning and the ending, but for some reason it didn’t work.

He said that they never do the beginning until they know which mechanics are going to work in the game, and they’re sure that they’re not going to throw something out. You can’t have a sword fight in the beginning of the story if it doesn’t work mechanically yet. I wondered if you have a similar approach, if your beginning wasn’t there yet.

Barlog: It’s interesting, because we’ve always done the beginning last. What we ended up using for the beginning was something we were doing, but thinking we were throwing out. It was based on a short story that I wrote in the first year, to get everybody comfortable with the idea. I wrote it in the style of that Cormac McCarthy story about the father and his son. I never said “Kratos” or “Atreus” in the story. It was just the father and the son going out hunting.

That was a tone piece to help people think, “OK, I can see that. I get it. I see where you’re going.” And then we had to keep doing these things where — okay, we have to come up with a vertical slice. We don’t have anything? Let’s just use that short story. When we were brainstorming the beginning of the game, well, let’s use that short story. It ended up becoming this thing we kept leaning on. Surprisingly, we were doing the beginning of the game very early.

The elements, the real heart of the elements, came after we had cemented everything else. Things we were adding to it that really made it much better. Uncharted 4’s beginning, getting out of the orphanage, I thought that was so good. Such a brilliant beginning. The beginning of Last of Us was done in the last bit of the game. They’d done it, and apparently they had to reshoot everything, because they had a better idea.

That, right there, is why they’re the top of the game. They realize they have a better idea. They know it’s a better idea. Or they believe in it, so they’re willing to take the flying leap. I try to live that as much as possible. When there’s something better, it doesn’t matter what the consequences are going to be. We have to follow that. We have to say, “This is going to make it better. Let’s do it.”

GamesBeat: It sounds like you can’t get wed to your story, either. You can’t say, “This is the beginning, the middle, the end, I want it to be this way.” You find, at the end of the process, that you could rip it up.

Barlog: Sometimes what you’re doing is figuring out how pliable you are. There are certain anchors throughout the story, anchors that I still argue about with the writers today in some cases. But they’re non-negotiable. Those things won’t move. They’re important. They need to be there, because everything fans out from there. But by nature, we want to mess with everything. Knowing which things to say, “Nope, this is too far, can’t mess with that, but here, that’s fine.”

When you get to see everything together — we had a unique experience on this one in that we were constantly revising and writing the whole game. We didn’t have any “a-ha” realization at the end. Everything worked out, because we’d gone “a-ha” all the way through. I think our big realization was that we had to cut a huge chunk and we needed to figure out how to fill it. What we filled it with made the game better. It surprised me. “Why didn’t we do this before? This is amazing!”

So in some ways it was great that we cut that, but — not only being flexible with the story, but knowing what not to be flexible with. You can easily get lost. I see a lot of games where they lose the thread because they cut too much. They rewrite a few things, and all of a sudden they don’t realize that they’ve just completely, fundamentally changed the story.