Dealing with Scope Change in Heat Signature and Gunpoint

[Music] I'm Tom I was the writer designer and programmer on a game from gunpoint that came out in 2013 and on heat signature' which came out just last year and when I start at gunpoint I said this publicly I'll be surprised if it takes me less than six months and sad if it takes me more than a year gunpoint took three years when I started heat signature' having learned nothing I said publicly again I'm just taking a break from learning unity to make something quick in game maker heatsink we should took three and a half years so clearly both of these games had scope creep they both deviated wildly from their original plans and scope creep is a dirty word it's a bad thing we're supposed to avoid it unfortunately it's also kind of my development process I have never really stuck to any plans and I'm mocking my past self active being so wildly wrong about time estimates but if

I'd stuck to my plan on either those games I just wouldn't be here at all I would just be someone with a slightly boring hobby project that never really went anywhere and I've already given up before the next one so all the good stuff in those games came from the stuff that I made up along the way that deviated from the original plan so there's something good here and but the warnings about scope creep and not misplaced there are definitely dangers as well so I feel like I I definitely screwed up something here probably two things one of them would be just when estimating the time that my own games are going to take me I think on that topic I have proved the world's least qualified person and I'm not just wrong about it I'm extremely wrong and consistently wrong and I'm also the only one who's wrong about this like no one else thought gunpoint was going to take six months no one told me he'd simply would be quick I'm incredibly bad at this and like a lot of unqualified people who have never been in the vicinity of being right a lot of people listen to me and so I'm just gonna have just shut myself up about that so this is my excuse for

not sticking to my plans until you make something you don't know anything I find I never have any idea everything's going to work out until I've actually made it so the whole process is about prototyping whatever idea is you have and prototyping enough of it to know whether it works and that process has gone right for me once and kind of wrong for me once because the other thing I screwed up is that heat signature taking three and a half years is a lot worse than gunpoint taking three years because gunpoint was three years of weekend's heat signature was three and a half years of full time so at some point this got really really inefficient and so I want to look into what went right on gunpoint what went wrong on heat signature and how I could improve the process to try and hopefully err on the side of the faster development process so this is gunpoint when I started making it I actually before this point I had a plan about

being a robot in space dropping fridges on people and not saying that wouldn't have worked but it isn't the direction ended up going in by the time actually started to learn game maker and mess around with the prototype fire is pretty set on you being some kind of private detective and I also wanted you to be able to click and fling towards wherever you clicked on the screen which has nothing to do with being private detective but I just did both and here we are but the only other plan I had really was that I really liked Deus Ex and I wanted to see how many things that I liked about deus ex I could capture in a 2d game one of which is being killed on site and another of which is if you do get a jump on someone you can take them out pretty easily and then I think about you know lock doors for grant waves past them security cameras all that kind of thing about infiltrating a building figuring out ways past security and how how many of those could I do in 2d and how well they work and one of those things is hacking

and I love hacking in these games I love the feeling of subverting the level to work the way you wanted to taking things that are supposed to work against you and making them work for you and so I was thinking about ways to do that and thinking back to when I used to design levels for like junik in 3d and quake in those games you would first place like a door or a switch and then after you place them and figure out how they're wired up like which thing is connected to what what switch activates what doors and you have a load of freedom when you do that and it's a it's a fun process so my idea was maybe you could give the player that freedom and let them kind of just switch to a mode where they see how everything is wired up and they can just drag those connections around to work however they want so they can drag a light switch to a door and now pressing the light switch opens the door instead of turn their light off and I tried that and it was really fun I immediately had loads loads of cool ideas to mess around with it I sent it to testers testers really

enjoyed playing with it in its most basic form it's obviously too powerful because you can just hook one light switch to every light in the building and then make the whole building go dark and sneak through it but it was obvious how to fix that problem it was obvious that you should have some things that you can't rewire at first so there's the things are here on a protective circuit and once I get to the circuit box I can then hack into those as well and rewire everything and it just immediately became obvious that there should be a puzzle game like that was just a puzzle mechanic the ability to rewire things was a really powerful tool restricting that an interesting way was felt like a sort of puzzle the obstacle and it was just very intuitive to build levels around that concepts you can't do this there's some reason you can't use the crosslink to solve this problem figure out an interesting way around that and so gunpoint just kind of told me what it wanted to be I just wanted to be a puzzle game just obviously and I just rolled with that and I just expanded the hacking mechanic to you know a crazy

extent I built the whole game around it which was definitely scope creep because I wasn't sacrificing anything I didn't I have some sophisticated combat system that I then cancelled because I was going so deep on the hacking thing I just went deep on the hacking thing in addition to all the other stuff I'd already done and I think it was good scope creep it was scope creep in the direction that that contained all the interesting things interesting possibilities that the game ended up having and so years later I finally got an artist this is John Roberts art and yeah it was just remarkably smooth on point I just got really lucky that the thing I it just happened to be within my abilities that was not in any way a cleverly judged or [Music] smart decision of my part it was just pure luck I happened to pick something that turned out to be easy for a beginner to do and the idea picked the hacking thing just evolved really well

so this is how I kind of saw the points of element at a time like this is what I thought I did and I did do this but this is only one way of phrasing it I made a little bit of everything and then I saw what shortly like the hacking thing was clearly the best thing I made and so I went with that and that's pretty good rule when it works let's look at a case where it didn't work so this is heat signature my idea of heat signature was just go inside spaceships it would be cool if you could go inside a spaceship you're a spaceship they're a spaceship you dock with them and you go inside the spaceship so it approached like that I started by just making some infinite space you can just fly around wherever you like in here and then when you come across an enemy ship you can drill for it and then you go inside the spaceship there you are inside the spaceship that I don't really know what to conclude from that there it is off you go so I'd successfully prototype to the concept but obviously there's still some

I'll answer questions here I don't think this conclusively tells you whether it's gonna be a good game or not so I just kind of had to keep making it and I made a ship generator and your hundred randomly generated chips so it generates exterior shape and then also interior layout and make sure every room is connected to every other room I made enemies with AI and a basic combat system and all of these rooms are subsystems of the ship so those guys I just shot some of those with turret Gunners and if you should take them out the turret can no longer fire that guy with a red X I just shot was my objective so I had there's a mission here to kill someone in particular I just got shot that's in there too and all of these little red dots those fuel barrels if I blow those up it not only takes out the Fuster preventing this ship from flying but it also destroys the room so all the ships have modular destructibility and if I get to the cockpit I can take over the ship and fly it myself so there's a huge amount of just systems and stuff that had to build just to explore this idea just to

find out if there was anything interesting here and I think actually that is a mistake in itself there was just too much that needed prototyping in order to test the idea of heat signature I just had to make nearly all of heat signature there wasn't really one single thing I could just mock-up and then find out is this fun a lot and to just keep making the whole thing so in future I'm gonna when I'm purchase typing stuff or when I'm thinking about concepts that I want a prototype I'm always going to ask how much of this do I have to make before I know if it works that's the key thing it's not not necessarily how long will this whole game take to finish it's just how much of it do I need to make before I'll know whether it works or not and obviously that's a bit of guesswork as well but you can make a lot much more informed guess about that so I just kept working on all the little bits and I looked for an artist and I found a guy called John Roberts is the same guy who did the Alpha gunpoint not a coincidence that you turned out to be incredibly good at this style of art too and I didn't really know what to focus on

unlike gunpoint I'd roughed out everything and it didn't tell me what kind of game it wanted to make what kind of game it wanted to be no one feature really stood out and so I was taking it to shows and having people play it at egx and rezzed and asking them like what do you think of this what should I work on next what does it need and the feedback was really positive people were enjoying playing it but when I ask them what it needed there was a very strong trend in their feedback almost everybody said something like I want to know why I'm doing these missions like the missions are fun but I want to know what the context is why I'm doing it because they were just you you would just wake up you get them in a mission you go and do it and then it just starts again and I was quite pleased with this feedback because I already had a plan for adding context to this game and my plan was to make a galaxy actually just making a galaxy would be too achievable so I decided I would make infinite galaxies by creating a galaxy generator and it would all I do

is just create empires of space stations connected by trade routes each space station mining a different resource from the nebula and each faction wanting control of a full set of all the different resources that they needed to sustain their empire and then they would just dynamically push into each other's territory and tie and take Space Station's from each other and that would be a really just as a small controlled meta game this this is stupid in many ways but I think the biggest problem with it is not so much that that it's ambitious to make that whole thing it's basically the same problem I had with the concept of heat signature itself which is just that you kind of have to make all of this before you know if it does anything at all like just making that galaxy you can zoom out and see felt really cool but it didn't answer the question of does this successfully add context to your missions does this solve the problem people are having do people now feel like they have the motivation because at that point the

missions don't connect to that meta game it doesn't do anything so you've got to make the whole galaxy then you've got to make hell the galaxy relates to your missions and then you find out whether the last six months were spent on anything worthwhile at all so this is the exact same slide I used last time too much leader pressure diving here I should have how much of a Vista I have to make before I know if it works and so I had made quite a lot of that galaxy stuff and it wasn't really working out and I around that time I went to this developers retreat to called Steuben in Sweden where you go to cabins in the woods and focus on your game with about 20 other developers and it's a really good opportunity to like focus but also to reevaluate like I got there and I was still working on this galaxy stuff which

just wasn't really panning out it was it was there and it was adding context your missions technically but it felt kind of arduous because you'd do a mission and then you'd see how affected the galaxy map and if it captured the station or if did something that you wanted it to do then you would just kind of see the whole game stretching before you in kind of a bad way like oh I just do that again and again and again until this whole galaxy is cleared it wasn't an appealing thought she made the game kind of more tedious and I've been wrestling with that and reinventing it and trying to solve that problem in all kinds of ways when I went to stew can I the idea of just resuming that work and just tinkering with that same thing that wasn't working was really unappealing to me and so instead I switch gears and I decided to just I'll just work on one of the many other things I know I'm gonna have to work on at some point and I worked on the interior game like when you're inside the spaceships what kind of vitamins can you find I added teleporters grenades

added armor types turrets things like that all the stuff I knew was going to add anyway but I decided to add it then and there and the game started getting better really fast it was incredible how much how efficient that was it was just you had a few items and there recall and then you had a few more items and not only are they cool but they also interact with the items you just added in really interesting ways so back then I'm actually rewinding see if we can show it I am about to put down a proximity grenade and then use a swapper teleporter on someone in the next room so that when he switched places with me he triggers the proximity gray areas put down and I can shoot his friends so stuff like that just emerged from the rules I was adding all the time and that's exactly the kind of game I wanted to make you know this game was also aiming to be a bit like Deus Ex and anyway I could possibly make it and that's exactly the kind of thing I loved in the ASX is when a solution emerges from the rules and the more

items I added the more that happened the more fun I had with it and I was just getting to a point where every time every day I worked on it the game would make me laugh out loud at something that I didn't expect that's was surprising and fun and made sense with the rules so this is clearly what I should have been working on instead of the galaxy thing this is what I should be working on all along but that lesson is kind of useless for all of us because I'm not going to make heat signature' again you're probably not gonna eat signature I don't recommend it and so what I want to know is like not what I should have done but how I could have arrived at that conclusion sooner because I did get there in the end but this was you know two and a half years into development that I finally realized what I should have been working on so I think the question I asked what does the game need most was the wrong question to answer to ask if you could answer that perfectly I'm sure that would get you a useful answer but it's just too difficult to answer that question I'm

really you want a question that focuses you on what the answer to that is likely to be and so now I think the question I should have asked myself is what is the core of the game that is a question that if I had been asked that one year into development like two and a half years before I finished the game I think even then I could have told you it's the interior game the stuff that's on board that's what you're going to spend your time doing it's got to be interesting and the way I think of like what is the core of the game is to ask this would it ruin the game with this bit was kind of bland so the Space Flight in heat signature could that'd be kind of bland yeah and actually it kind of is we didn't make that very sophisticated or fun because the more put into that the more people it just put off the game because people who couldn't handle flying stuff but otherwise might have liked the game then weren't able to play so we just basically left that extremely rudimentary and that's fine and they could have told you that would be fine

even two years probably finished so if you asked that about you know the Galaxy Maps the strategic layer would it ruin the game if that was kind of bland no it actually is kind of bland and that's true for loads of elements releasing chef but if you ask me that about the on-board stuff like when you go inside the spaceship would it ruin the game if that was kind of bland yeah it would absolutely ruin the game it would be terrible there's just not nothing you can add surrounding that that's ever going to make it a great game if what happens when you're onboard the ship is just kind of boring and so that the answer to that question is gonna be different for all kinds of games but you know if you're making a Monkey Island game can the dialogue be bland no the dialog can't be bland like you and a bunch of other things maybe can that there's gonna be one part of the game that just has to be you know unusually good for this to work at all and obviously if you're answering yes to that on many elements of your game then

cool boots that might take a while right I'm now looking for there are certain kinds of games where like multiple things all need to be good and those are the kinds of games that you probably don't want to make if you're alone I'm looking for I always start working alone and then bring in other people if and when I end up with something that's worth making and so I'm always looking now for things that just have a single core like something that I think okay this is probably gonna be the core of the game and so this is my process now I I'm going to choose a game idea that's quick to prototype and quick to prototype means the you can test whether the idea will work by making some small part of it not the whole thing so heat signature is bad for that because going inside spaceships you can poach like that you don't learn much unless you also make a whole game around it and then approach like the important bits of it you get to decide what the important bits of it are you probably

know why it excites you and what's interesting about it and at that point decide what the core should be so in the case of gunpoint that question I didn't even need to ask it because the gamers told me this is what the core of is it's going to be about hacking that's by far the most interesting thing with heat signature the game didn't tell me that and so I kind of got lost in the reeds and start adding things that weren't really relevant to the to the central core of it if I just ask that question I think I could have answered it and exploring that that area is exactly what ended up making the game good in the end so once you know what the core is or once you've decided what the core is because you can make it up if you like you can scope creep as far as you like in that direction like scope creep is good when it's related to the core of the game which is once you if you're right about the core of the game is then you can add to it as much as you like you can make it more complicated you can make it more difficult you can make it more sophisticated you can expand it you can make more work for yourself as long as it's in the area that is actually

what's special about your game in the first place so that's what I'm doing next time and I want to stress that this is this is my new process I put that specifically at the top to stress that this is just what I think is going to work for me and for particular kind of games I like to make I'm looking for things that are mechanically interesting and don't to make things that generate exciting experiences and I want to make those games as quickly as possible I don't have a good track record to make them quickly but that's what I'm optimizing for and so I just want to make sure that it doesn't sound like I'm saying this is how you make a game this is just my idea for how I can improve my own process and the other thing to say is that heat signature it would be easy to talk about it like a success story because it sold very well I should say better than gunpoint it reviewed well I'm really happy with it I have a great time playing it when I play it I'm very proud of it but all of that is only

possible because gunpoint paid for it it was it took way too long what really happened is I failed to make good game in a reasonable amount of time you know two years in I did not have a good game I devoted a huge amount of time and money to it and I had failed to make it work and if I was under the same kind of financial constraints that most people are I would have just had to release it in a bad state or just give up my whole career may well have been over so it only worked out because I have the enormous privilege of being able to afford to make those kinds of mistakes and learn from them so I feel like the least I can do is to share what I did learn so I hope it's been useful thank you so we have time for a few questions if you'd like to ask him a question please come to one of the mics in the aisles and I'll start things off then Tom you have a background in the games press right you know do you consider the press when you're coming up with you know the

concepts for the games you're gonna be working on and I guess I'm always I can fall back on the the fact that maybe because I was pressed over for related reasons if there's an idea that really excites me I just assume it's going to excite some other people as well and I think I should get the game the seen more and more supports lots of little niches and lots of well actually big niches and it feels like now if you're really really psyched about a game idea and you're right to be excited about the game idea ie it pans out then there's probably going to be a bunch of people who are excited about it too and I think I've been especially like either the kinds of games that excite me are like I say Deus Ex and stuff which are not like the biggest selling games in the world but there's you know several million people who love those games to death and so as long as those million people continue to exist they're probably hopefully you're

going to like things I think of thanks on the subject of that how did you find kind of wrestling with your audience in terms of them kind of like feeding into your scope creep so like you know then you kind of talking about something that was sinking your time and then the audience being gone gone and more features game what's the thing how did you kind of counter that yeah so the question is how do I counter other people feeding into my script by suggesting things and I listen to people suggestions and say oh that's a cool idea and then I don't do them you have to get very used to doing that you just have to do that a lot because there are some good ideas and they definitely did benefit from people's suggestions on heatsink sharing gunpoint and especially friends who I you know specifically asked for ideas from but sort of 95% of the ideas you're going to

get are not necessarily bad they're just outside the scope as you say it's if you're going to be able to resist your own scope Creek you at least need to be able to resist other people's ideas for your game and so heat signatures was the worst for this because it could be a million different games you know it could be Eve it could be there's something about space games in particular that makes people sort of want them to be everything I think because there are space games that are trying to be everything so people assume that every space game wants to be everything so people keep asking why you can't mine asteroids or build your own ships or start your own Empire and have a crew and all this other stuff and it's the answer is because that's nine other games and it was hard enough to make one so yeah just I don't shoot down people's suggestions or just quietly ignore them thank you hey I think when you spoke about the development of gunpoint you had mentioned that because you only had weekends you were specifically picking

tasks that fit within a day or the weekend hmm as a it was a natural constraint was that something that you thought over tried to apply to this or what did that discard that window because you didn't feel the need to arguably I should have I didn't sorry no that's wearing out yeah that was a Gunpla it was mostly weekends and that did change the pace of development in lots of ways and yeah I said earlier that heat signature is probably three times as long because in a number of days it was certainly at least that but there is a sort of hidden benefit to having five days of not working on your game in between the two days that you do work in your game because you come back a lot smarter I think and I think there are some tasks that do not fit into a weekend and that's that was a problem with gunpoint and I'd lost a lot of time to sitting down to it on a Saturday morning and thinking what was he doing like how was I going to solve this I don't remember but equally it had a week to

sort of sanity check everything so they come back and thinking you know sometimes sit down and think wait why was I doing this and I discover I should not have been doing this this is a complete waste of time so it had ups and downs and ya know I didn't try and apply that to heatsink show I just assumed that having more time to work on a game would just be outright better and yeah that was probably contributing factor into me sort of failing to control the direction of the production as well as I did with gunpoint thanks you go ahead over here I think you were first so this one during a few years ago you did an article of how to explain your game to an as you I'm curious do you explain your game to you explain it how do you explain gunpoint how do you explain heat signature I practice explain it to other people so explain to myself would be too easy because I know I can read between the

lines and I know what I mean by everything but yeah gunpoint I changed the how I described it to people continuously for all three years I was working on it and actually still now like there was a we did a steam sale like a few years ago it was like two years after gunpoint came out and I decided to change the pitch for gunpoint I think I have a better way to just explain this game now so something I constantly work on and yeah her heat signature yeah just keep writing like a little blurb and actually conferences are the best place to kind of hone this because when you explain to someone at an event you get to see their reaction you get to see whether they grasp it or not and that's the the key thing is just have that you can understood what I'm saying and I definitely changed my mind about how to explain that a lot of times Thanks okay Tom I just wanted to say I think a couple years ago you did that game maker series about learning program games I just want to say thank you cuz

it came to me at a time when I kind of stopped with flash dying and stuff so thank you for that and some of that when you were making heat signature' did you kind of feel like how far into it did you feel that you had to sort of shift gears before working the interior or was it kind of like frog and water starting to boil where you just didn't notice for the longest time and yeah I noticed that it wasn't coming together and I actually got kind of burnt out at GDC two years ago I remember talking to friends here and saying like I think I'm just going to take two months off and try and do something else and take a break what I end up doing in that break was making another game called Wharf blade because it turns out that I if I take a break from game government I just think about game to them and the entire time so the only thing that can actually distract me it's just a different game so yeah I wasn't unaware that it was failing to come together and it was like I said it's stupid when I started to to figure

it out and that was sort of June of 2016 I want to say so like a year and a bit before it before finished thank you and in general I sure I think that's that seems to be a pattern that like after you figure out what your game is and get to the heart of it and have a good working prototype it's about a year before you're done okay we only have time for one more so when figuring out your core how did you start looking at your game in terms of mechanics that add to it and then things that you were bringing into it like what kind of things did you start as like a rule set to add and subtract from your game yeah for the gadgets and stuff and heat signature I started with teleporters because it was the easiest thing I could imagine to code you just click what you want to go and then you change the players position to where the cursor is that's a teleporter and then I knew that would be too powerful because the whole game is about trying to get to a certain locations I came out with

downsides for them and actually yeah that was that was why I was planning on teleporters and I'd actually been planning them for a long time I had the sort of document of the ideas for devices and stuff and I just kept having new ideas for teleporters had about seven ideas for teleporters for some reason and adding those in with their constraints there's one that can only swap with people there's one that can take you wherever you want but it takes you back to seconds later and then there's one that can take you to anywhere that you could get to by walking anyway but instantly and those all just felt really interesting to mess around with so mostly it was just experimentation it was just I focus on the easiest things to make first and then see if they're any good and it usually they are or they need just like a small tweak to make them good and so I didn't have like a big like unifying rulesets of what items could or couldn't be but I did need to do is figure out a system for like how their program sort of how they're implemented and and so in heat signature

its traits every item has a list of traits and it's just a list of strings the data that we say you've doesn't store any stats about the weapon at all the game just knows oh if it's long range that means it goes this far and if it's silenced that means it makes this much noise and that applies to like melee weapons grenades teleporters all that stuff and so that was Ryu SFIL and that also starts to inspire you to like come up with new ideas of like what if there's a grenade that had the hacking trait all right I think that's it so please make sure you fill in the feedback forms to let Tom know what you thought and let's give him one last round of applause [Applause]