Why God Of War Took Five Years To Make

On today’s special bonus episode of Kotaku Splitscreen, God of War director Cory Barlog joins the show to talk about JRPGs, text size, and what the future looks like for Kratos and fam.

You can listen to the whole thing with our fancy embedded player, get an MP3 right here, or read an excerpt below:

Jason: Weren’t you worried that showing Odin had his fingers in everything but not actually showing Odin would be almost like a tease to people?

Cory: Totally. But that’s what I wanted. Some of the most compelling series to me don’t always do everything in the first — they set a lot of stuff up so the power of the characters and the interactions in the second or third really resonate strongly.

Jason: So how much of the future game plan do you have mapped out in your head?

Cory: Five games.

Jason: So that’s 25 years? Is that what you’re prepared for?

Cory: No. I want to figure out how to make these things faster, man.

Jason: How can anyone make these things faster?

Cory: I think this one – a big portion of the five years was, we had to start from scratch. We had a core engine but we really redid a lot of stuff. The previous God of War games, when we went from God of War 1 to God of War 2, the core engine was there.

And aside from the fact that the combat design team told me, ‘We’re just gonna do a little rewrite of the combat engine – don’t worry, in a couple of months we’ll be back online.’ A year later, we finally built our first character on God of War 2. It was the most stressful time ever, because they kept saying ‘Just another month’ – it’s like doing construction on your house. ‘Don’t worry, another week, another week.’

On this one, everything really needed to be redone, because we just had torn the engine apart in so many different ways that when we finally brought the team together, everyone realised, ‘OK, this is not where it needs to be.’

So even when you see E3 2016, the rendering engine wasn’t there, the lighting engine was half-there, the atmospheric engine was half-there. The core mechanics were there, but a lot of the way we were streaming and loading everything was still getting worked out, and figuring out how we were going to get it logistically to work. We knew what we wanted, we just didn’t know technologically how we were going to get it in the right order.

Jason: How do you as the director stop your team from doing the same thing when you’re working on the next God of War game? How do you say, ‘No you cannot mess around with the combat engine — we need to do this in two years’?

Cory: They’re not gonna listen to me on that one. That’s a very difficult thing to do, right? We have this group of people who are all perfectionists, and they aspire toward making whatever they’re doing better. We were at the wrap party on Friday, right?

We launched the game the day before. And the thing that was echoed over and over again when I was running into people was, ‘Yeah I’ve got some ideas on how to make this better, how we can do all this stuff.’

Jason: Make the text bigger?

Cory: Alright, so yeah, we own that one, that’s on me. You know what’s interesting about that? We did not get any feedback during all our playtests on that.

Jason: Was everyone just sitting with monitors in front of them?

Cory: Yeah, and we’re learning all the time. We’re taking in so many pieces of input as we’re studying how people play the game, the reactions to it. And even after the game is out, we’re seeing, wow, our methodology for testing had a blindspot.

And that blindspot, even when we sent kits home with people, was still not illuminated. And it’s fantastic that the community is patient with us as we’re like OK we need to react to this, figure out the best way to solve this problem.

It was definitely one of those, we were focused on so many pieces of the game that it just never came up, and immediately, the first time somebody said it, I was like, ‘Hmm, we’ve never talked about this.’ I don’t really look at the subtitles at all – even when I have them on, I know what they’re saying, so I’m not really paying attention to that.

Jason: It was the menus also, not just the subtitles.

Cory: Part of it was, our menus came online literally in the last eight weeks.

Jason: That doesn’t shock me. So you didn’t have any menus until the last eight weeks?

Cory: We did, it just wasn’t the final versions of it. We had really stand-in rough stuff to get the structure and the skeleton. But really getting it all in – it is the adage of any creative thing, it looks terrible, it is an ugly baby, until the very last second.

Then all of a sudden, there it is, everything’s there in a span of two or three weeks. And you get very little time at that point to react to it, because everything’s being done at the same time for the duration of the project: the engine, the mechanics, the level design. That was the most maddening part of the beginning, was: We’re trying to design levels but we’re also prototyping the mechanics that will drive and be the building blocks of these levels.

So the level designers are yelling at me, ‘I don’t understand how you want me to design a level when we don’t have any mechanics.’ And the mechanics guys are like, ‘We don’t have any time, we have to start doing these things, but we don’t have any tools.’

Kirk: Sounds super chill. Doesn’t sound stressful at all.

Jason: This all raises the question: Why do you make video games?

Cory: Because we are masochists. We hate ourselves.

For much more, listen to the full podcast. As always, you can find Google Play. Leave us a review if you like what you hear, and reach us at splitscreen@kotaku.com with any and all questions, requests, and suggestions.