How to Iterate As An Artist According to a AAA Game Director
An interview with God of War Ragnarok: Valhalla co-director Mihir Sheth on the key to iteration, how to intake feedback, defining your wins, and more.
Season 3 of the Artwell podcast is available now! I studied God of War and interviewed its creators to find lessons that are relevant for modern creatives.
Mihir Sheth has been working on God of War for 12 years. He joined Sony Santa Monica as a Combat Designer and was promoted to Combat Lead, before being selected to co-direct Valhalla, the epilogue to the critically acclaimed God of War Ragnarök.
If you’d like to listen to the full, 90 minute interview with Mihir, you can find it on Apple Podcasts, Spotify, or wherever you get your podcasts.
*The below has been edited for pacing and clarity.
What about the culture of Santa Monica Studio allows for quality of craft?
One is just an inherent respect for craft as a value. Obviously Santa Monica Studio is part of a larger Sony initiative of first party content and bringing value to the platform. And that's absolutely the case. But I think in a more local sense within the studio, you're just surrounded by people that want to do the best work that they can do. We tend to attract that kind of talent, we look for that kind of talent, and we try to sustain that [kind of talent] which is really enticing. If you are already aligned in that framework, you want to be surrounded by people like that. When you're trying to iterate and make something better, you're not thinking like, “oh, it's because of the company strategy, we need to strive for excellence”. It's like, “no, I am doing this for me. I want this to be good”. That part of it is really, really big. The other part is the player experience and how we view the player experience. One thing at Santa Monica Studio is the level of respect we have for play testing, no matter how strongly someone feels about something, you can't argue when people are playing through it – the real honest reaction that they have, we have a deep respect for that. That coupled with what we strive for in craft makes it feel like there's purpose to what we're making, even though it's a video game, it's for entertainment. But even within that context, it feels like we're striving for something that has real purpose and value. Caring about player experience and caring about quality of craft is consistent across the board. Fundamentally it just comes down to the people. If the people working at our studio eventually change into wanting different values, the studio will change. But because we cultivate those values, we hire for those values, it perpetuates that. We're grateful that within the larger Sony context, we can preserve that identity. We're given a space to do that.
What's the key to iterating properly?
Let me first start by saying iteration is both a great thing but there are also massive drawbacks. Iteration is expensive, the value of a good first pass is often underrated in the current creative zeitgeist where iteration is praised. Don't get me wrong, I love iteration. Iteration is absolutely necessary, but to do nine passes of something that involves lots of people is an extremely expensive process. I think the first step of how you iterate properly is to have a good structure in the beginning. The value of a good first pass is incredibly high. Now that isn't the same thing as me saying get lost in theory land forever, and don't even get moving because you want to make the perfect first pass. That's not what I'm saying. You need to be disciplined. You need a time box. You need to acknowledge that iteration is going to get you there, but don't devalue the intent that you put into a first pass. If you don't have intent when you do something, how do you know what to change? When you get feedback, if you just did something “just because”, when someone gives you feedback, you might be like, “Oh, that's interesting”. But that doesn't help you move forward. If you come at it being like, “well, I know why the first pass has this. It has this because of this. It did this because of this”. Now, when you get some feedback, you can correlate the feedback to the intention that you had, and then you can develop some kind of strategy for how to move forward. You want to plan, you want to prepare it, you want to have intent. That's what I think design is, action with intention. It's kind of a corny phrase I use a lot, but if you don't have intent, then it's hard to justify one thing or another. The second part, how you validate is extremely important. Feedback is an input, but as a creator, it fundamentally comes down to experience, knowledge and judgment. When I iterate on something, I'm making sure that the way that we test something is credible. Was there intent? Is it representative? Is it useful in terms of how we'll get feedback? Can I trust the feedback? Because sometimes when something isn't representative, you get nervous around the feedback. You're like, “well, that thing was buggy to begin with so we don't know if they mean this or this”. What are you going to do with that? How is that going to make your second pass better? As far as you can, try to make it credible, then when you get your feedback, do work to make sure that you can validate what you're saying. And this gets into data analysis and there's so many subjective ways to use data. You can use it to fulfill a narrative or an agenda but do the work. As much as you can to give yourself confidence that what you're getting as feedback, you can believe it and it is helpful. And finally, don't let it decide what to do next. Take it as an input so that you can decide as the creator, “based on this, I believe we should do this next.” It's really abstract what I'm saying, and it obviously depends on context, but I think that the science of iteration is, what are you trying to iterate on? What's your process of validation through iteration? Try to find a strategy in what you're doing, where you get effective steps going forward because the more cycles you have an iteration, the more expensive it is. When you're a small team and that cost is not high, that's one thing. But when you're a AAA developer and there's lots and lots and lots of time, money, and people working on problems, you have responsibility as a lead or as a director to make sure you're being responsible about how you approach problems. Sometimes we don't talk about that enough as a creative community, on the value of a good first pass and, and really having accountability on how you manage that process to be effective in getting results in a shorter period of time.
That just reframed something for me. I often say, “don't be afraid of a bad first draft”, but what you're saying is don't strive to make it a bad first draft either.
Eventually, over time, you get better at your first passes. I'm not saying to build a masterpiece on the first pass. Iteration is part of the process and it's a great thing. But there's a reason why you'll see people in their lane that they're just so good at what they do that you're like, “man, their first pass was amazing”. It's not the final product, but what people don't say is, “that saved a ton of time, effort and money”. At the end of the day, they have saved a lot in areas that you don't even see. Because a lot of times it took less iteration to get to the final result. That's a more practical framework at looking at this but I agree with what you said, don't be afraid if the first thing sucks and it isn't succeeding… that's part of the creative process. But then your process of validation and analysis and the next steps and strategy need to be really high. Otherwise, it could just be bad for a long time and you can do 10 iterations and spend a lot of time and money and that's not great. That's not what we want as developers.
You iterate a lot based on feedback that you get from playtests and from other people, but when do you trust your gut?
That's the million dollar question. It's always a mix of both. Data is an input, but for the types of games that we make a lot that comes down to judgment. We strive for [our games] to feel human. There's a human connection at the heart of it. We do a lot of things that are counterintuitive. There's a lot of things in our games that would have never shipped if we just purely listened to feedback. We are a studio that is quite director driven in the way that we are structured. In my opinion, the role of a director, fundamentally, is to bring a level of cohesion to a title. That is very hard to achieve without some kind of singular or small group of vision holders at the top. The value that cohesion brings is extremely resonant and it's hard to see until the very end. Why? Because as someone that's in that position, you see things that no one else on the team is seeing. I think the role of quality comes down to our playtest and validation method. This is my framework, this is not official Santa Monica Studio framework, but I see directors as really responsible for ensuring a level of cohesion that makes the game feel resonant to players and our process around play testing and improvement and analysis and all that kind of stuff as funneling quality over time. That mix between cohesion and quality is kind of our stamp as a developer, I feel like that's something that we try to make sure feels unique to what we make. So judgment and trusting your gut is something that you learn to get better with and you be humble about it too. You make some bad decisions and you be humble to it and you be open to it. You look to improve that but you need to have a vision and understand what you're trying to do and why you're trying to do it to know when to push back against the data. That's the hard part. That is the struggle. I like that struggle. I like that we engage with that struggle as a studio. When you're making art that's what you're doing, you're looking and you're validating, “what am I feeling? Why am I feeling this when this is seeming like this?” You're doing the best you can and you make decisions and you move forward and you trust that your team is going to be there right with you and execute at the highest level. It's worked out for us, but we're also humble that we'll make mistakes in the process and try to account for that in the way that we build things.
What's something you still have to get better at today?
So many things. This is going to be a little more on the personal side, but I struggle being effective when running at different gears. What I mean by that is my framework for a very long time was like, I'm always in high gear. I enjoy intensity. I enjoy passionate work. I like getting into problems. I enjoy that pace. I think that there's value in that. I've seen some impact of that on my own growth professionally. Where I struggled is when things are a little slow, I wouldn't be as effective, both in terms of productive output, but also just myself, like how I'm feeling. Intensity is kind of tied to me feeling good about what I'm doing and myself because I'm just running at this high gear. I'm sure many people experience this with maturity in whatever they're doing, but you see the game differently later on. You see the playing field in a different way. It's kind of all about the value that you're bringing holistically. If you can do that at a high pace, if you can do that at a lower pace in between, that's real resiliency and flexibility. For me, this is very relevant because the nature of my day to day has changed throughout the years. When you're the core implementer for a character like Kratos, you're in the middle of everything. You’re deep in the weeds on everything every single day. Then over time you get abstracted more and more to see things at a different level. Sometimes your wins are not, “I got this mechanic in, in a month”, but, “hey, three years from now, I hope that we can achieve this goal”. Things move at a different speed and learning how to be effective and how to bring value to others and how to be happy for yourself with that shift is something I need to get better at and something that I'm trying to improve
I'm asking as somebody who very much relates to operating at a high gear and not being comfortable operating anywhere slower. Do you have any tips for operating at a slower speed sometimes?
For me, I always ask this question to others: “Where do you get your wins from?” If I ever meet someone that I'm interested in, I like asking this question because everyone sees this quite differently. I mean, this is the most open way possible. Being more conscious of where you get your wins from and learning how to define your wins can change your perspective of how time is used. What I used to say is like, “four years from now, when this game comes out, if people love this weapon, that's my win”. That's not a great framework for a mature approach to a long-term goal. Now it's like, “over time, if I see improvement in my relationship with someone where I think we're on track to form a really good partnership, that's a huge win”. Even if we have one conversation, I'm like, “did I see signs that showed that there's more willingness to be more vulnerable, to be more direct, to talk about things because who knows what possibility that could create”. I'm not saying that in a transactional way, rather that it's creating a possibility space for just positive things to occur. Who knows what the value is going to be defined by that. That is a real win. Learning to be more effective with how wins are defined and then doing some analysis on yourself of, “do I feel that as a win? Is that honest? How could I feel that as a win? Is there a different framing I need to use?” Then you are able to adapt it a little bit better because to me, my anecdotal psychology is that, frequency of wins is a very, very big factor of how we operate as people, how we manage burnout, how we navigate life. There's a baseline frequency of wins that we're almost wired to want and need. One way of doing that is turn things up to 11 so that you can force those frequency of wins. Another way is to carefully look at that and question, can you redefine where your wins come from? That's another way of potentially solving that problem of frequency of wins. You could try that. I'm still struggling with it, but I think it has helped for sure.
Do you think there's things that are wins that don't feel like wins?
All the time. There's some things that you can't possibly know are wins until far later, which is why I like this retrospective approach. It's a very humbling exercise to be like, okay, let's say something good happens. Then you think about why this good thing happened? And maybe you'll realize that the largest contributor to it happening was something that you had almost nothing to do with, or was something that you did that was so insignificant that turned into something. That's valuable just because it helps you be grateful. It helps you view the world a little differently. It's also a huge ego check for how we see the role that we play in controlling outcomes. That is very, very useful to do. It is extremely humbling to remind yourself in the moment you have a very specific narrative of how you contribute to outcomes. And just like how we validate and play test when we design things, we need to validate as we kind of go through things in life and be like, is my perspective correct? Is this the way that things happened? When you think of the question that you asked of like, are there wins that you don't think are wins? Absolutely. It's impossible for there not to be. You are limited in what you see in the present, which is a very human thing. That's the value of looking back and asking questions and being open and being a little humble about it.
What does art done well mean to you?
Resonance is a key part of it. I use that term a lot and it's kind of vague, but it's something moving that at a human level connects you. I don't know if there's a science or there's a thing behind it, but it's a quality, it exists. When art is done well there's a relatability to it. There's an appreciation of it. Obviously quality of craft and things like that are a factor, but I don't know if I would say that that's necessary for art to be done well. I think resonance is really the core factor there. Another part I would say is honesty or integrity. Does it come from a place of vulnerability in some way? These giant AAA games are huge machines, but fundamentally it’s tons and tons and tons of people working on them. I view them as art because in that moment, when people are working on it, they are putting different parts of themselves into the game. You might not even understand how or why or whatever, but it exists. There's something kind of poetic about that. It's like, there's a part of you that's in this. As a developer you have a sense of this. You're like, I made something that a part of me is in. I think that's meaningful for art.
Thanks for reading! If you haven’t already, make sure you subscribe to the newsletter for more essays that help you become a more thoughtful artist. If you want to learn more about God of War, you’ll like Season 3 of the Artwell podcast where I studied the making of the games and interviewed its creators to find lessons that are relevant for modern creatives.