AMA: Sucker Punch Productions Co-founder Brian Fleming talks Ghost of Tsushima

[Music] good afternoon everyone welcome to gdc showcase i'm bryan francis a contributing editor at gamma suture.com and maker of things here at gdc i'm joined today by a very special guest for today's ask me anything session please give a warm welcome to sucker punch co-founder brian fleming brian would you please introduce yourself well thanks brian uh first uh just gotta say i'm super uh glad for the opportunity to talk to people today so uh as you said i'm one of the founders here at sucker punch uh so i've been here since the very very beginning all the way on the n64 days and and most recently of course we finished uh a huge project goes to tsushima and launched it last year awesome uh and just to be clear he's brian i'm bryant i've got the t at the end of my name now to get things rolling we are here obviously because you yes you in the chat want to

ask your questions for brian about sucker punches to excellent 2020 game ghost of sushima uh i really love this game i am super excited to be moderating this chat i can't stress enough that i put hours and hours and hours into it so normally i would love to ask these questions but i'm giving up that time so that you all can do the asking um you can ask questions about ghost or if you want to know more about brian's experience in the game industry that's also on the table just kind of a reminder that obviously as a game developers conference we are asking questions that are relevant to game development but hey we thought we find a way to make everything work here uh with that in mind i'm going to go over to the chat and start pulling some questions for us we've got some ones coming inside to see how many all right uh let's get started here thomas belotti would like to know what is your favorite feature activity or activity that didn't make it into ghost of tsushima if there were any and what will be the reason they were cut well you know i there's probably lots and some of those our favorites are probably you know things we keep in the hopper for future

games um you know i i think that uh it's it's a some of the really terrifically bad ideas often end up being really good ideas in the long run you know we had a whole bunch of ideas about how to do wayfinding in the world and and a lot of those bad ideas then were replaced when you add in something like the guiding wind and you really feel that sense of of progress when you're making the game you know on you know that was a few years before we shipped but um it's it's those moments where you have something that's not working and you transition to something that's really working that for me on the system side is some of the most uh entertaining and enjoyable stuff uh that we do i guess the other stuff you end up cutting most often is just for for scope you know you have one too many mini games or one too many uh interactions and maybe some of those mini games will show up um in some future work so that's a good answer i especially like pointing out that you know like the guiding wind you

know didn't just emerge from thin air it came oh no from iteration when should a merchant i don't know um let's go to brandon ledbetter's question uh when working on a game of this scale how do you manage ideas and features and when do you decide to start cutting that actually plays off the last one really well right yeah you know we talk about scope is everything sucker punch not a particularly large team we're not small you know at our peak we're probably 150 people and that includes our qa staff and uh you know all the things you need to run a business um and and so not a huge team and so scope from the very beginning was the entire you know it was a conversation throughout um and so strangely i feel like we feel like we were cutting constantly and i think that's part of the healthy process is that the things that you decide to cut kind of give room for the things that are actually really good about the game to to have space and to grow and so i would say

really from year one all the way through to three months before the game ship we were cutting it might have been smaller things you know you might be cutting a scene that isn't working or an individual lines of dialogue to sharpen things up but the scalpel is one of the most important tools uh that we wield uh and certainly uh we you know we did throughout the process of making ghosts uh we're gonna switch paths here the question that folks in chat are super seem most excited about um we've got this great thumbs up system you guys can use to just thumbs up the questions you want to hear about here answered um from raphael smears mores uh japanese players largely praised goes to sushima for being one of the games to most accurately represent japanese culture to date um their question is what were the procedures to ensure the game would execute such a good representation of that culture well um you know i think from the very beginning the original pitch it was part of our um we were aware of

of the sort of the level of responsibility that we had to have in this area so um there's a story that that people the fans may know which is that we actually pitched this to shu yoshida um he was one of the very first people we ever pitched into because we wanted not just um his guidance and his assessment but really his blessing that hey this this could be done that a team from the west could really do this um and and with his support then we we began engaging with uh you know a wider audience and that almost immediately included the team uh at sony isd the japan team and and their participation in the game throughout from the original pitches to each of the videos that we would show they would they would reflect back to us the parts of the visuals and the parts of the story and the parts of the characters um the names of the characters all the details that they found

that um both worked well and the things that were not working for them and we were you know super uh appreciative of that feedback and and responsive as a result and so uh that process was was really a healthy part of the ongoing development throughout um and certainly uh without their guidance we would never have achieved uh what we achieved and there's even little stories and you know i one of the details that i really love is that team was so committed to the project that they made sure that every single kanji in the game that they understood that that kanji actually existed in the 13th century um and so if they had you know there are often multiple different ways to describe something and they went and found the exact uh kanji that would have been used back in in that time to make sure that we were being authentic even at that level so uh tremendous effort in partnership with that team oh that's such an excellent anecdote uh it sounds like international cooperation you know it's

one of those things that yeah and you know i mean it's just one of the joys of the project was being committed to being a learner not uh remotely pretending that we understood it to recognize that we needed to go learn and that you had to go find your selection of teachers and mentors to really help us uh on that journey gotcha we're going back to thomas belotti here who seems to be asking some some really questions that are resonating with chat um what were the biggest hurdles transitioning from uh the infamous series to ghost futsushima it's a great question um the i think the single biggest change for the studio was um you know infamous was an open world game but a kind of a small open world and we were going to a large open world and effectively that change you know kind of across the board destabilized every aspect of our game making tool set um we had you know all the ground in infamous would be hand modeled by by

a 3d artist um there's no way you're going to 3d model the individual vertices of something as big as this so you have to change that you have to change the you know we went from you know like a hundred tiles to 1200 tiles uh the tiles were four times bigger um everything broke and so the biggest challenge really was the journey to getting to the point where we had solid earth under us again um because for really many many months you know 18 24 months things were still really quite difficult to synthesize into a game you could sit down and play um because there was just so much that that used to work that had had was not able to deal with the scope of what we were doing so first couple years were actually quite difficult uh at getting the game to stand up and then as we got farther into it you sort of at least had you know you weren't sinking into the into the

swamp uh constantly and you could you could begin to iterate and make progress talking more about the um kind of the the nuts and bolts making of ghost sushi um uh muhammad uh al salman i hope i pronounced that right uh feel free to correct me in the chat um what was the hardest feature to implement in the game and why was it so hard yeah it's easy answer and the answer was combat the combat system uh was something that you know the team that worked on that handful of programmers and designers and animators you know they worked on non-stop for six years um we worked you know and and built multiple versions of it with multiple approaches and you know you're really dealing i i the term i use for it is kind of ring zero i always think of a bullseye and and and the combat system along with a few other systems like the way that the hero moves through the world they're kind of the center of everything um and they have to work in in every situation they have to work in

every lighting condition they have to work in every terrain they have to work in every uh every strange mode that the game might be in combat could potentially break out in and so as a result they they are just exceedingly uh constrained right you have to it's not an isolated part of the problem where you can just say oh well in this case we'll just ignore button inputs nope can't do that oh you know you in this case we'll ignore damage nope can't do that oh we'll ignore ground matching nope can't do that so combat and and i think you know you have on top of that this goal of of making it so that it isn't so simple that you can just push x but it isn't so complicated that the player's head explodes trying to deal with the control scheme and the different modalities and so it was an elusive uh non-stop effort for that team over you know literally a six-year journey

um continuing to work on that and and i think the results were really good but it was a long difficult road gotcha we've got we're going to pivot the discussion subject quickly to talk about the game industry getting into it living in it uh surviving in it um both connor shaw and christopher zawatowski would like uh they're asking a question of the same vibe so i'm gonna read conor's copy but they're both asking it um uh copy um uh can you elaborate on your path into the game industry uh would you recommend uh your pathway to others hoping to break in what are specific lessons you would impart on students and other hopefuls well you know i'm old enough that the industry is is changed a ton so i don't know my path is exactly right um i worked on games when i was a college student um i worked at epochs making an old apple ii game called california games so i had some experience with it and then i walked

away and worked in commercial software at microsoft for a number of years and then came back to games when i started sucker punch and so the start of sucker punch was was pretty bruising because there weren't any really good entry points pc gaming at that point um was kind of going uh really really quite difficult um console gaming was great but you really had a tough time getting nintendo or or uh sony to give you even a development kit because you were a small you know coming out of nowhere uh studio and and so i think right now it's actually better um it's a better situation because there's a ton of different platforms and entry points for small teams and and my advice is um go make games uh figure out how to make games and and keep making them because we we you know when we ship ghost i you know i've said and i i truly believe that anyone who ever worked at sucker punch contributed to this project

because over the years that we've been making games we've been learning how to in fact make better and bigger games and we don't make uh ghosts without infamous and we don't make infamous without sly and you make fly without rocket like that series of games is is a journey that that we needed to go on and so my biggest advice is just make games find a way uh to make games and and and you know continue that journey um there's no substitute for building and shipping games because you will learn more by doing that than i think anything else um while i grab the next question here uh in chat um uh um excuse me i just wanted to say that a couple folks did give a they said they gave a shout out to california games so if anyone else played california games i haven't fortunately well i wrote i wrote the assembly language uh hacky sack and surfing games so if you want to go play

uh you know get your emulator fired up and go play them uh you know those are the games i worked on um uh oh amelia sailor auguita uh would would like to know um what would be one thing you would have done differently during the development of ghosts of sushima with obviously hindsight being 20 20. that's a good question um and one i really haven't thought a lot about um i think that the regrets that that when we talk about things as a team um you know it's going back almost to the question about cutting i feel like um we maybe waited too long on some of those cuts um and and it is it is really difficult oh i i know the answer to this question i'll give you a different answer so cutting i think being brave about cutting earlier would have served us better i think we postponed some of those cuts and as a result we spent more time on the stuff that was going to get cut anyway um you know i think if you're in doubt about things

on a game of this scope you probably ought to cut it um because you have so much other content the other big i think aha lesson for us was that complexity will overwhelm the game and the player and the reason for this is that the game is so big it has so many systems and interactions that um you you you tend to think about hey we're working on combat or and then another team is working on navigation another team is working on mini games and each of those teams can think oh i need to make my part individually rich and detailed and the problem is when you have so many different components and they're all rich and detailed the player can get overwhelmed so i think if i had any big takeaway that i wish i knew that going from going in was make sure that each of the individual systems are almost held to the standard of being as simple as you can make them um rather than thinking oh each one needs to be complicated you ought to

really focus on each one being simpler because the combination of all of them is actually quite complicated so oh that was a really thorough answer um joaquin de losada would like to know how do you guys know how long a game production will take to make and how do you all make sure that the company is able to meet the deadlines obviously game production is fluid but it's still a you know scheduling yeah it's how games get made well the real answer is we didn't um we are in a situation and we're very fortunate to be in a situation where our relationship with sony um you know and we've made uh you know games together for for you know almost two decades at this point and i think we have a very healthy relationship with them where we both understand that the real goal was to make a game like ghost it wasn't to make a game at a particular date it was to create something that was

a great experience for players and that that's the most it was a new ip and it needed to reach a certain level of of success and so if that took us a little longer that was the right path but that only comes because sony is confident that we're not being wasteful and that we're you know really diligently uh protecting their investment or we are continuing to make progress um but that you know it would be quite difficult i think um you know we did not originally forecast a six-year project and it was on the strength of our the progress we'd made and on the strength of our relationship with sony that everybody agreed we should keep investing to to try and and find the true diamond that we thought was there even if it meant it was going to take a little longer so um was it difficult it was a difficult you know miss we missed we were not close um we did not forecast you know oh five

and a half and it turned into six you know we were not close on how long it would take um and again i think it's it's really just a reflection on our relationship with sony that we were able to finish and complete such a good game yeah um alexandra soria would like to know um how and this may be a question more for the programming side but it was i it resonated with me because i loved the fast loading so much that i wanted to ask it uh how did the team manage to achieve such an optimization that it made it possible uh to load the game so fast um it's a it's a great question and i'm gonna say we were all pretty shocked at the reaction um again we lived in this world where the game worked the way it worked for years and so i think when we shipped and and the news was like oh my god the game loads so fast and we kind of had taken it for granted how big of a deal it was um the the engineering team did a spectacular job at and the art team once the engineers

had done their work at really understanding what was sort of core to the game and we didn't have to reload all the time and then what was localized uh in and around uh the player and that we could bring in um and they did some clever things where we have really low resolution versions of all the textures in the game so that we can begin uh to do the fades uh while we're still finishing the last little bits so there's there's cleverness uh here but the truth is that i think if you had adrian one of our lead engineers on he'd tell you it was just the basics it was just being careful about uh organizing the data so that it could be loaded without seeking the head all over and um it was just good solid engineering work uh and then of course the artist did a great job working within those constraints so uh kind of ordinary success speaking of artists uh that is what you call a segway uh valerie bourdain has a great question

um one of the most memorable elements from ghost of tsushima were the visuals and the art direction yeah valerie um how did you go how did sucker punch go about creating assets that were historically accurate and yet had a dynamic and vibrant feel and i guess we can throw in there uh that they also were you know optimized and worked in the constraints of the engine i guess yeah you know jason cannell who leads the the art team and was the art director on the project i think one of the things that we had from the beginning was a fairly clear idea of what we were trying to do with the game and that certainly includes the visuals and as we started pushing forward on things that we wanted to achieve visually you know a lot of the inspiration came from the source material you look at curacao movies and all the movement and wind that he used and so that was a big push for the team to capture that sense of movement and so

that was an early investment point and then early in the project the team made a couple trips to japan and they go to sushima and they they went to historically uh preserved locations to kind of draw that inspiration from the reality and the combination of all that leads you into the directions uh that we went so um you know you have kind of the source material and visits to the sushima you have the inspiration from the curacao films and then our exploration really was just how do we how do we capture that and i think the last really big you know uh center point for jason was just continuing to try and um remove noise from everything to look at the individual textures the individual shapes and and look for simpler uh less sort of noisy shapes because it added a sort of sense of calmness to the world i think the combination of all of that and you know it's just a long journey um but but uh hopefully a successful one

yeah um uh auntie lenon here has a great question i think is about leadership um usually in a game project at the start everything is shiny and exciting after working on something day to day for many years the shine and excitement naturally fade away and it becomes hard to stay motivated how is your team how to stay motivated for such a long production cycle and i'll tack on to the end of that you know as a studio co-founder like what are the ways you think you can do to help keep folks motivated as a leader well um i don't think we have any secrets i think that the first thing that really helped us on the project was that i thought the vision for the game was really compelling you know that we weren't talking ourselves into it we believed from the beginning that this was the right project and that it was a it was a going to be a successful idea now the challenge is that even after you do the initial pitch you know years before you ship the game i always think that um after i've pitched to a hundred people at the studio you know we show the slideshow

there are a hundred different pictures of what this game is going to be at that moment right each person has heard that presentation and they have an idea and those ideas are similar but they're definitely not the same because one person is thinking oh the combat is going to be you know um souls like you know it's going to be bloodborne because they just played that and another person's thinking oh the game is breath of the wild and another person is thinking uh you know oh no it's going to be infamous because that's the game we just made and it's going to really be different and the journey is really helping everybody's vision of the game kind of converge over time right and it doesn't converge uniformly but you try and and so every six weeks we get together as a team and we do a review of our progress and we would try and select demonstrations of progress we'd made for two things you wanted progress that was individually exciting and you know maybe it was a tool demonstration or a

an exciting visual effect that we were going to have and that's inspiring but then also things that would help converge the everybody's idea of what the game was going to be and so if we knew that you know one of our ideas was that we knew it was going to be an m rated game right but we wanted to talk about that because some people think of m rated game as you run around and you chop heads off all the time um and and so we had to actually have conversations about how does violence get used in the game and where would if somebody's you know gonna lose uh you know if we're actually gonna chop someone's head off we're only gonna do that in narrative we're not gonna do that systemically and why do we feel that way and so all of that was about helping each of the hundred people who heard that pitch start seeing more and more of the same game and uh that process continues throughout the entirety of development but it's a really essential part of of of building new ip it's what makes new ip so challenging god i want to dig into

the decision around the the mature violence that you just said but yeah i'd love to don't have time for that gotta keep moving before i grab the last few questions for today we only got about five minutes left um so we're gonna get one or two more before we get out before we go i just wanted to tell everyone they should totally stick around uh we've got more great free sessions today the rest of the week uh more great folks from the team you know the demon souls team the spider-man morales morales team are going to be here later in the week the ggc live podcast i am so excited for all the stuff you guys get to watch um just by sticking around here on swap card um uh also before they go brian if there are any questions we didn't have time for today is there a way they can ask them is there a way they can get follow-up what do you recommend uh you know that's a good question i would have turned it back to you is there a process for them to get you a list of questions i you know i'm certain we can do an email with you but uh all the questions are on your they're on your screen right now excellent good point uh we will we'll

you know what we will we will figure something out and we're gonna go back to the questions now for the rest of our time here all questions okay thomas belotti would like to know what technology are you the most excited about for the future of game making across any discipline what am i excited about i you know the thing that i'm the most excited to get to to try is uh the new ps5 uh storage system the the loading systems there will change the way that we think about how we make games um it's so fast that um even the idea of unloading the things that are just off screen on the camera and as the player turns bringing them in just in time is possible and that really fundamentally could change how we think about making games so uh lots of exploration to do there but i'm i'm intrigued by what that could mean for us oh um uh uh alexis harrington has an interesting question uh in researching for the game we talked up front about you know researching um you know and you've mentioned before visiting sushima itself um in

researching the game were there any specific resources that the team found extremely helpful beyond we mentioned the sony team before but i guess if there's anything you want to throw in well i mean i the the critical thing is is is people you know is having the right uh individuals uh for me you know there was a there's a book uh the mongol invasions of 1274 and the subsequent invasion that that was really helpful for me at just getting oriented into the content um and and certainly lots of people looked at that and there's there's numerous reference books but i think in the end the critical critical thing was finding people who knew the culture knew the history um and and who you know were able to guide us with their expertise it was really more about that than than a particular resource yeah um i think this might be our last question raul mann has kind of a leadership company leadership question since you've been around for a while what do you think are some of the main

directions that sucker punch took as a company and as developers to be stick around for so long um you know i think we've been relatively uh conservative about hardware so we never used our games never required some kind of add-on it was always sort of aiming at the big part of the market we've chosen great partners uh you know we worked with ubisoft first and sony second um and we we you know that's really been helpful um i think we've also always viewed our publishers as partners rather than as adversaries and i think that's really helped us i think last we probably uh over indexed on realizing that our team isn't just the sucker punch employees it's our uh banker and our lawyer and our uh you know the person who does the insurance who sold us insurance they're really kind of expansively thinking about your team not as those people are vendors but those are teammates i think the more

you can really invest in those relationships and people uh the more when the moment comes that you need them uh you will have built the relationship that lets you get through the next you know whatever is the obstacle that you have to overcome roger that well brian thank you so much for joining us today we've got to restore my questions up yeah uh we can't wait to see what's next for sucker punch productions and who knows maybe i'll just hop off when i'm done today and go play some more ghost of sushima sounds good uh i'm gonna go ahead and wrap us up a little a little early thank you so much i hope everyone has a good day and like i said before stick around there's more great things happening at gdc showcase this week bye everyone thanks everyone [Music]