Iterating Design and Fighting Fires: Updating Fruit Ninja and Jetpack Joyride

[Music] alright this is um iterating design and fighting fires updating fruit ninja and jetpack joyride my name is Luke Muscat thanks everyone for coming along it's pretty exciting this room is way bigger than I thought it would be so I'm freaking out right now thanks for coming along as well because I have a talk on Thursday and it's on at the same time as like the crazy panel with Sid Meier and cliffy B and Peter Molyneux so I'm just assuming no one's going to turn up to that one so it's cool that some people came along to this line alright get started so yeah my name is Luke Muscat I'm the CCO at half Ric which stands for the chief creative officer I made these which is a jetpack joyride monster - fruit ninja and a bunch of other games that no one knows about or cares about and so this talk is going to be all about so it's kind of like a fruit ninja update postmortem so a post release post-mortem and you

know I'm gonna be going through the goals of a lot of the updates we did and the results and talking a lot about you know updating jetpack joyride and how that's been different to updating fruit ninja so just to be clear as well like I'm only going to be talking about fruit ninja iOS the SD version so just the version on iPhone and iPod I'm not going to be talking about the HD version Android bada how we go about like you know making sure the updates move across all of those correctly and you know that's a whole nother talk altogether and frankly I'm not the best person to be doing that talk but yeah so this will just be about fruit ninja SD which has had over 15 million purchases so far I haven't updated that figure in a while that might be a little bit different now and over 40 million updates downloaded - so a bit of context to be a bit useful before I start this talk so Halfbrick is 57 people now we're pretty big company we haven't put on all of those numbers since fruit ninja like when fruit ninja

came out we were already about 3035 people but internally we're broken up into teams this is my lovely very attractive team we have seven guys and we actually work on two titles at once right now we're working on half the guys are working on jetpack joyride update the other half of you know working on some prototyping for some new IP so this is just kind of useful because you know when I'm talking about how we do our updates and all that people might be thinking like god you've got 50 people how can that update takes so long we don't put all 50 people on the update all updates I'll be talking about were completed by three to four people so rocket racing hands up if you've played this game see how there's like no hands up this game was a disaster this was a game we made a couple of years ago for PSP minis and this game was a hardcore abstract racing game that we made we spent eight months on this game you know we really pulled a lot into it and it absolutely bombed it was you know it got

okayish critical reception but just you know there was no market for it so it's because of this game that we had this new goal and this new goal was to ship lean you know ship multiple games really lean really quickly and if it works we'll update it often and this way we'll be lowering our risk of development I'm just gonna move this yeah so that's where fruit ninja comes in fruit ninja is a direct result of rocket racing it was me trying to make the opposite game to rocket racing what we have here is like 80% of the design document that exists for fruit ninja this is all there has been and all there probably ever will be for fruit ninja you know pretty simple slice fruit those slice bombs and when we shipped we shipped really lean it's almost crazy like you look at fruit ninja and if you're familiar with the game at the moment you know it has you know like four modes and unlockables and

achievements and leaderboards and social features and all kinds of crazy stuff version 1.0 had nothing it was so so lean there were no modes there's only what now we know is classic mode there were no combos there were no unlocks there were very few devices we had to worry about there was no such thing as game centre or retina or any of that stuff and because of all these factors we actually shipped the in six weeks from from start to finish just with three guys plus the sound guy for a week as well I was looking for like a screenshot of old fruit ninja I couldn't even find that this screenshot has got a you know multi-touch unit 1.0 didn't even have that I just couldn't find like an older screenshot than this one so when we launched everything went really well we got a really good critical reception you know got good reviews all over the place which was very cool I couldn't find any cool screenshot so if you know us at our early stages of success so you have to make do with my artists interpretation

right here yeah I'm not an artist I'm a designer and so the game is out we have 1.0 and time for some updates so this is really you know what the talk is about update 1.0 1 so this was our very first update and our goal was pretty simple we wanted to roll an update really soon after launch and show everyone that we were committed to this game we were going to be supporting this game and although this game is simple there's more to come we started working on this before the game was even out while we were waiting for it to pass like Apple certification and and all that stuff so our current player base was zero and is pretty simple you know we were gonna add two new fruit the mango and the pear which you know we had a new fruit now no one really cares such a large roster but back then we didn't have that many so that was kind of a big deal we're going to do multi-touch support so that you could have multiple blades which seemed like it'd be a cool thing and we wanted to we knew that the performance wasn't that great especially on all the devices when you there was a

lot we could do to make that better so we were really focusing on that so I have a bit of a story here for fruit ninja came out and it was doing pretty good you know it was moving up the charts and in Australia it was doing really well so it's a Friday night and fruit ninja overtook doodle jump in the charts it was still way down in the American charts um you know position 120 or something but I was doing really good in Australia and we were so we would have an hour traditional Friday afternoon beers and obviously because things were going well we were having a few extra beers to celebrate and you know really getting into the spirit of things so to say and it got to about eleven o'clock or so and we got an email that came in don't ask us why we were checking emails while we were drinking and it was an email from Apple telling us that they wanted to feature us and that they needed the assets immediately so being in Australia you know it was our Friday afternoon they probably sent it on their Thursday afternoon or Friday morning thinking

that we would have time but because it was gonna be our weekend and no one was going to be there we're like right we got to do these assets right away we entered like full panic mode and kind of semi drunkenly got all these assets together which was you know really cool you know we want to do that again in hindsight but we're usually a lot more prepared than that now we learn our lesson from that on so then two things happen the very first update launches at the exact same time that we get featured by Apple and we're stoked you know featured by Apple that's a huge deal it's gonna help our sales dramatically we're gonna get all this extra exposure all this extra attention awesome and it's right when the update lands that's amazing right boom this update was a disaster unfortunately in when we were working on the updates and the the performance fixes we had obviously screwed something up pretty dramatically and for some reason about 15% of players

could no longer play the game and this was the worst time you know we just sort of got this momentum going we just got featured and then this update came and you know all of a sudden the one star reviews are flying in and we're in full panic mode so we need to do a new update one one point oh one point one pretty simple goal you know and we were fortunate enough we got it fixed really quickly and it was it was actually really interesting like you know what obviously when this happened and you know with this locking people out of the game we're thinking it's this massive disaster and you know I'll the game is do forever but there was this some it was really interesting we got it fixed we ended up kind of nicknaming these builds where we fix the problem a hard firefighter builds and we got this silver lining where people really seem to respond to the fact that we took notice that we'd done something wrong and that the game had sort of become unstable and we fixed it really quickly

you know we were communicating we were on forums we were replying to emails saying we're on it and then we did fix it certainly we lost people because you know they they tried to play the game it was no good and they felt like they'd been gypped out of their dollar but you know we also won some fans because you know they were they were watching us they could see that we cared and that we were you know doing our best to keep things going so we learn a few lessons pretty quickly on this one you know updates can be crippling certainly updates can enhance a game but at the same way it can take away from it you I'm sure there's some other iOS developers that could attest for that one I think was Minnie Gore might have had a really bad one once but wiped progress and you know it's certainly something to be careful of and so you know we realized our QA effort needed to be increased dramatically for these updates we really hadn't given the update the QA time it deserved we also needed to you know assess our risk associated with each update adding new fruit easy that's cheap that's probably pretty low risk for being in there mucking around with the render code trying to make it run faster getting

into dangerous waters we also realized that players are keeping score on us which was interesting you know we really didn't think people would be paying that much attention to half Rick that turns out they are so at this point we get to the update 1.2 our player base has grown dramatically in our one hundred and forty five thousand players which at the time seemed like the entire world we were stoked and our goal was to release something big that would warrant a second trailer and another round of media and we really wanted to you know capitalize on this initial momentum we had going so we weren't really sure which way to go with this so it's tricky because a game like Angry Birds cut the rope a lot of those level based games they have such a good way of updating the game like you like our game you like playing levels boom 25 more level it's so simple it's easy for them to add the value proposition for the players is obvious like there is now more of what you like to play and it's just so easy

to communicate we didn't have anything like that for fruit ninja we really weren't sure on which way to go with it so our producer Jason Harwood had this idea and he went through and he went through every single iTunes review we had gotten so far and he collided it into this massive excel sheet and then he grouped up like common you know common suggestions that people had and common problems so we had a section for bugs section for suggestions and all that kind of stuff and this is what we ended up with so so the very top one there is our you know multi slice and then we have needs more game modes needs some kind of endless mode that there's more fruit so on and it was all organized by the number of mentions that had so number one multi slice this still gets requested all the time and every time I am against multi slice the thing is right now in fruit ninja the fruit is either sliced or it's not sliced it's

binary the call for action for the player is so obvious if the fruit isn't sliced I need to slice it as soon as you introduce this idea of multi slice things start to get trickier there's now you know it's been sliced but I could get more points if I don't there's an opportunity cost there and additionally as the fruit gets quartered and maybe eight or sixteenth it's actually getting smaller on the screen and it really stops becoming the optimal size for interacting with so multi slice is something that I've been very much against for a long time so the second thing that you know second and third thing was hey more modes would be cool and how about an infinite no bomb mode people really hated the bombs in classic mode so yeah we were like okay I get it you know bombs are kind of tricky let's try making this infinite no bomb mode and it was interesting we started working on it you know we got it working really really quickly and for the first five minutes it was really cool and then slicing fruit all of a sudden lost its meaning you it was really weird you kind

of played anymore like cool no bombs then there was no pressure and there no kind of constraints and he sort of burnt out on slicing fruit which was bad news for a game that he's only about slicing fruit we tried some other things to really spice it up as well we made this thing called a critical waves where all the fruit would come up and they would you know definitely be a critical and the blade would turn blue and it would have sparkles on the back and that's how you knew that it was gonna be critical and even that still didn't feel quite right just the points were meaningless and you know it really just wasn't clicking fortunately we were able to use that blade stuff a little bit later on so while we were working on this we started mucking around with the the idea of putting combos and that was another common thread that had been in our iTunes reviews I love this like a gif just showing what this combos look like in slow-mo I always think it looks really neat and you know we we put that in and we got to benefit from that straight away it worked in a classic

mode you know you could just slice the fruit or you could try and take bigger risks one of the big things that we did was like rather than be the school going up exponentially so a four fruit combo being worth five points and a five fruit combo being worth 10 points or whatever we really wanted it to be as simple as possible and make it so that people would understand what school they got and why immediately so we kind of made some sacrifices sacrifices to the gameplay in order to increase how clearly the game was communicating with the player and then we added a time limit to Zen mode and all of a sudden it all kind of made sense it was so weird we added the 90 second time limit and it was just it's kind of like well it's annoying because it goes on forever and you have to hit the pause button and then that's how you get out of the game so maybe we'll just add a time limit and that was it we added this time limit next thing you know you look around and all the guys are trying to be at each other's high scores and this weird thing happened where Zen mode was now like both the most relaxing mode well that's

what it was intended to be it was this relaxing mode where you know you didn't have as much stress and pressure to it's extremely hard core competitive everyone trying to beat each other's high scores which was absolutely fantastic and all of a sudden that felt fruit ninja again it felt like it had a place and a lot of the times it's just this kind of intuitive sense that it belongs in the game we went through and we tuned the waves in a Zen mode to make him play off that as well and they're all combo able in different ways the other thing we added was the coming soon fruit we wanted people to know yes this is a big update but we're not done yet we're not even close to done yet we wanted everyone to know that we are still working on more stuff so when we released everything seemed to go pretty good we got quite a bit of coverage around the place which was neat you know everything kind of worked how we expected we we said we wanted to warrant doing a second trailer and a big media blitz and that definitely happened got really widespread coverage which was

great and then this is the graph for what happened so the orange line is updates the blue line is sales they're not necessarily on the same scale which is kind of compressed so you can see what's going on otherwise the blue line is really flattened at the bottom because the update volume is always so high and you can see you know we were on this downward decline and we did this update and it helped but not a great deal you know we spent a long time on Zen well relatively long time we spent three or four weeks on it which is almost as long as it took to make the game and you know we got this bit of a bump and it stabilized us but it certainly wasn't this big boom like we were expecting so again lessons learnt we need to listen to the fans but read between the lines and I mean this isn't even something that's specific to you know iTunes reviews and community feedback this is just play testing in general right you know one of the ones I always remember is back when I was working on an old licensed game we had

some kids come in and they would design this puzzle game and these kids were coming in and they were playing the game over and over and over and you know it's kind of playing through the first part and keep got about 20 minutes in and he's going you know what this game needs I say to him what does this game need he goes this game needs a tank I want to run over things I want to blow up the walls and that's what I really want to do and so the interesting thing about that was obviously I can it had a tank to this puzzle game but you know what it really said to us was okay this game needs variety it needs a greater sense of empowerment and the same thing can be said for you know iTunes reviews you know community feedback and community engagement we really need to read between the lines and be smart about what we do they know what the problem is they don't specifically know what the solution is always although there are often some really great ideas and the other thing was just that updates can help keep the momentum but there are a lot of work I mean I'm sure a lot of people have done iOS updates know that they take a while to get right so update one point two

point one current player base has grown to 350,000 so that's pretty cool and we had this idea for this update we kind of had taken inspiration from the very first update where we botched things up we said okay we're introducing a new mode that's a pretty big change that's probably a pretty big risk so let's do this update and I'm gonna call it the way that we kind of internally phrase this update is a chaser update so the idea is this there's going to be an update that drops again really quickly after the major release update and it's gonna have a couple of new interesting things in it nothing major not like another mode or anything just some nice little extra bits and pieces and we're also gonna roll all of the bug fixes and all the problems that come up into this update the other thing we really wanted to do was you know spark some community in Gaga man hmm okay there's no spellcheck in Photoshop I apologize so what we did is we got some

minor stuff ready beforehand we added some coconuts and you know fixed a couple of bugs that we already knew of and we had this stuff ready to go before Zen mode even even dropped the other things we added were right us so really simple thing come ups it comes up every now and then you know hey how about you rate us we were a bit smarter about it than just putting it in randomly it waits till you've played a whole bunch of games because then we can assume that the player is engaged and then you know if they get a pretty good score or even their best score then we think that might be a good time to let them know that now's a good time to rate us if they had it like right after crummy rom and that's not so good also if they no thanks we never bug them again I don't see any point pestering people into it I don't see like when they see it for the ninth time that they're gonna have a change of heart and give us a five-star rating the other thing we wanted to do so this was to meet our user engagement and

community engagement was the dragon fruit so the dragon fruit is really really rare fruit it has a 1 in 40,000 chance of coming up with each fruit spawn only in classic mode there's some other kind of probability clamping and we fudge if the numbers quite a bit to you know make sure it doesn't come up too often or too rarely and yet we put that in as well just as nice little thing to try and get people talking and the really cool thing is once it came out people really were talking about it these are some comments they're just copy pasted straight off the touch arcade forearms and it kind of started to bring on this mythos like people weren't even sure it was real or not and people would claim that they'd seen it is amazing the dialogue that we had going and this is my favorite one people speculating about what the dragon fruit does speculating that maybe it avoids your slips it most certainly does not but that was just so cool people were coming up with that so really felt like we met our goal where the other thing

that happened is you know our ratings went way up with just adding that little rate at now thing no longer were people going to rate us to let us know that they hated us or that it was broken you know we were getting a lot of 1/u we were sitting at about four stars previously and a lot of that was you know a lot of five stars and a lot of one stars it wouldn't get many in between and all the time and the one stars it was people telling us about its brokered which is great we need to know when it's broken and under what conditions and what needs to be fixed about it but there was no one that not that many people going there to tell us that they liked it and it was great we did this and all of a sudden people were on outside they they were there and letting us know that they liked for the game sometimes they do things like awesome game and give you one star often accompanied by this game is terrible five stars sometimes I wonder if like maybe they're confused and it's like golf so lower score is better hole-in-one good job guys once so yeah and then this is what happened so the orange spike on your left there

is the Zen mode update and then this is the chaser update so you can see it dropped pretty quickly afterwards and our sales went way up you can see there it's made of really obvious a really dramatic difference on sales which was absolutely fantastic and I really attribute that mostly towards the ratings not so much the dragon fruit so that was really excellent and we also managed to address all of the bugs that really helped stabilize our star rating as well and so the people who downloaded it who got the chaser update who had been having problems had all their problems addressed really quickly so they were happy and then the people who weren't having problems everything was working fine they got some extra new stuff they got the dragon fruit and the coconuts so they were happy as well so we were really happy with this we really felt like the chasers was a good model it was important to make sure that we had the content side of the chaser ready to go before the previous update even landed so we could respond to the bugs in a timely fashion this is a model that

we've been using a fair bit I won't cover any other kind of chaser updates because they'll follow the same sort of model but we even use it for jetpack Joyride and Monster Dash and all of our iOS games now and also we learnt that ratings matter it's probably no great revelation so now we move on to update 1.3 at this point our player base has grown to about half a million people and our goal is really to make the game more sticky we really wanted people to spend more time playing fruit ninja it seemed like you know people loved the game they told us how much they liked it but they would kind of play it for about an hour or two and then they would burn and ID on it you know maybe some friends would get some really competitive stuff going on and they'd be really socially driven but we certainly weren't getting people spending a huge amount of time in there so to make it sticky we figured hey we need some long-term goals with rewards which pretty much sounds like every Shopp system ever made in every game so we started working on that we already

knew what we wanted people to be able to buy and unlock you know we already had these funky blades from when we were on the critical waves for Zen mode where we could easily color the blades and particle effects do kinds of funky stuff with that so that was a no-brainer and again adding backgrounds that's a nice easy one as well it was easy pre retina days at least it gets a bit trickier now with how much space it takes up on a on a Retina display so we had that sorted quite quickly what we weren't sure about was whether people should be allowed to have custom backgrounds just yesterday in my hotel I realized that I didn't have a screen shot for this background so I got on scoff and I talk to woody one of our programmers back in Brisbane I said hey can you just hack up a fruit ninja build and like hack in some random stupid background and send me a screenshot of it this is the first screenshot he sent me so is this one all right I like that is perfect and really what we decided it

was you know we didn't want to allow people to do their own custom backgrounds cuz we really felt like it kind of took away from the vibe and the feel of the game and sort of people were obviously gonna do all kinds of crazy stuff we did had you know you can kind of imagine the stuff that's going to end up in Google Image Search and not only that but we spent a lot of time mucking around with the backgrounds we had to make sure that they you couldn't the fruit was really visible on there like there's a lot of contrast with the fruit and we often use really dark playing backgrounds because it provides a lot of contrast with the juice which is a really important part so you know we wanted to maintain some authorial control over that so then we started working on the shop and the shop was damn tricky we'd come up with this really clever novel design for our menus that everything is sliceable and there are no shops that involve slicing a pineapple to buy things the thing is is people had come in and they bought fruit ninja' and they've done it in itunes store and that uses such a different

interface convention to what we were using this was just so crazy kind of off track and it was really a massive massive challenge for us to do something that would intuitively rely mostly on swiping and slicing we really didn't want it to go into you know some kind of finicky picky touch based thing we wanted to keep it very fruit ninja' so this is what became my wife I couldn't find any work in progress from those unfortunately although rest assured there were plenty of terrible horrible designs maybe scoot I'm not showing you it might be a little bit embarrassing but this is what we came up with in the end you know we made a couple of little concessions as well we're like okay there might be people that are still kind of expecting it to work like the App Store so we actually made our icons for the things that you could unlock glossy and rounded corners and if you tap on them then it would automatically scroll up to there as well so then the one of the really big problems we had then was um you know we had it in there it was selectable that

was fine originally we had at the instead of the current system where it is now where you unlock it by completing tasks it originally worked like a shop you know by playing you would kind of earn currency and then he would spend that currency you know unlocking things and again it was this weird being where you know we're playing it the whole team is playing it and they're like yeah it works but for some reason it just doesn't feel fruit ninja' we kind of had this same problem again where something that we expected to work and that seemed fun really wasn't working for our game it just didn't feel quite right in there it didn't feel like it belonged so after a lot of mucking around we ended up coming up with this this system where you know it basically like achievements we spent a long time tuning it so that you would unlock them kind of in a fairly progressive manner early on you would do it and if you were a really good player you would unlock like five blades in the first game and then you'd have this big drought where nothing happened so we spent quite a lot of time tuning that and in one release you know

we had a really solid reception again we were really happy with the iTunes reviews instantly changed from asking us to add multislice and new modes and things like that to just I want displayed I want this background I want custom backgrounds it really dramatically changed the way people were thinking about the game and what they wanted out of it and we didn't have any analytics at this point but we just know anecdotally is that people were sticking with the game for longer so we were really happy you know we felt like we achieved our goal with that we also got our our ratings crept up a little bit higher again which was grow and this is what the release looked like so again it's kind of similar to this Zen mode one where you know it has a bit of a bump and you know you can see the sails kind of get pushed up there but really it hasn't done much at all you know and again this is a lot of work who kind of bummed we were hoping again for it to explode up there but the good thing is is we weren't in that downward decline like we were we keep the updates were coming pretty thick and fast and we

kept on adding a lot of value to the game and as a result it was holding quite steady so yeah we were happy about that but you know we had been hoping for more so the lessons we learned was you know while we were receiving our goals of you know making the game sticky it was really important for us to not compromise what have made the game successful in the first place you know we could go through and hack things up and change it around but you know if fruit ninja doesn't remain what fruit ninja is about and that's going to be a huge disaster a huge mistake and also again you know updates not a magical silver bullet so update one point for our current player base is now at about 1 million players it's time to make another mode that worked out pretty good last time you know the Zen mode have got a lot of media coverage so that was cool and you know we figured this time we could do it another mode it would be bigger and better and more badass than Zen mode so what everyone was saying and I mean it was pretty obvious like a lot of people are saying at any of it in the

office as well was power-ups right like hey let's have a mode with power-ups in it will make them bananas because it's a mystery banana mode that's even cool so we started working on that at the time you know we were kind of influenced a bit by module blitz which is ironic cuz I think they were on right before me in this room you know we took a bit of inspiration from there we had a 60-second time limit and we started adding power-ups go war so you can see there a screenshot of an early kind of build that had the frenzied banana in it and then we had the freeze banana and the score multiplier banana which is all stuff that is still in the game but we had all kinds of crazy in there we had this gravity well banana where you would slice it it would make this black hole and then the fruit would come up and kind of get sucked into the black hole and all just sit there so you can swipe it all at once and get like a hundred for combo we had crazy waterfalls of fruit and time-traveling bananas that would rewind the game but not your score and all kinds of wacky crazy stuff but

somehow it felt like it needed more so again kind of you know at this point we were drawing things from you know bejeweled blitz and taking a lot of inspiration from that because I was just like hopelessly addicted any time I wasn't working on fruit ninja I was playing that we came up with this idea of Stafford when you sliced a fruit coins would come out and then at the end of the game if you had enough coins you could buy these power-ups so you know like turning off bombs and things like that so we were working away on that for a while and at this point we're at about six weeks worth of work so again about how long it took to to make the first version of fruit ninja which is kinda nuts and things are looking good but it's just not quite popping yet it still needs some love so you know we're still working along on it and then this happened we get to the office one day and surprise game Center is a theme and it has all these crazy cool features like inviting friends and matchmaking and you know real-time multiplayer and

leaderboards and achievements so we changed our mind on what we were doing and even though we had like this kind of half complete mode that was getting there it's starting to feel good we're like right screw that let's put that on the back burner let's be one of the first real-time multiplayer games out using Game Center let's do that let's strike while the iron is hot let's be those guys so here's a screenshot I don't have any work-in-progress shots of this because somehow someway this mode was designed in four days flat this is right off the back of spending six weeks on a mode and only getting halfway there multiplayer was knocked out in about four days and implemented in just over that as well because we're kind of implementing and designing at the same time to to really get it to market extremely quickly it was kind of cool because because the multi-player was so novel it was kind of almost easier to design for kept all of

everyone's fruit on the same screen just to kind of create that sense of connectedness and like you're really interacting with someone else and you kind of sharing that spatial location we did a lot of tricks to kind of hide how much lag there was in the system because it was pretty early and you know and yeah the coverage we got for that was just mental we got a huge amount of press for being one of the first we had a really positive reviews on the mode which was encouraging because they hadn't spent that long on it but you know definitely that kind of novelty factor of playing against people and being I'm better than you at fruit ninja' really really worked and most importantly we got a lot of love from Apple for doing this because you know as soon as they've released something new they really want to promote and tell everyone about how good it was and we were there right at the front right at the beginning when it first launched so you know we got refeed should they did a game center category and we were right near the top of that and a multiplayer category and we were in that and you know we got it so much attention for it which was fantastic so this one here the

very first spike on the left that is multiplayer and then the second one is when we brought in achievements and leaderboards we actually left that off the first implementation of Game Center just so we could hit market fast and you know almost first and then you know we also put in a couple of tweaks to multiplayer in there again not unlike the kind of chaser model and you can see this had a profound long-term impact on the game it really did lift it up and the amount of exposure just really helped fruit ninja immensely so we learned a few lessons at this point you can't plan everything this is obviously not part of the game plan game center coming out the mobile market is crazy and sometimes we feel like we've got to be crazy to keep up with it even though we kind of dropped but we were doing and we did this mad dash and it was always kind of insane scenario we definitely feel like we made the right decision in doing that and you know that top level of visibility matters when when you're there and this you know getting massive amounts of coverage across the site's that made a huge

difference there's absolutely no doubt about it so where were we oh yeah okay so okay mood is now 1.5 we now know we spent quite a bit of time on the you know supporting a multiplayer update we're at about two point 1 million player base we've really want to finish off arcade mode and we want to completely blow everyone away we're starting to feel the pressure right like this mode we've been teasing for a long time and people are starting to get a bit sketchy about you know whether when we promise there's gonna be another mode this banana mode whether it was a real thing or not so at this point the mode kind of looks like this there's some interesting things you'll notice about this that are different from the actual version of arcade mode that's in there first of all the combo blitz is there it's down the bottom there was actually a bar that ran along the bottom as well told you how much blitz you had left we have like frenzy text up the sides and then at the top we have a bar that shows you how much frenzy you have left as well and it's just not quite popping you

know again with that weird like it's yeah I mean it's good but we really feel like it could be better that's got of we can do better than this and we had we have our sanity testers so this is um not Emma and Chloe who are all partners of you know people working on the game and they've always been kind of our foundation for playing at our casual testers none of them are gamers in any way shape or form and you know they were really big support as a fruit ninja before it had even come out and you know what through all of the updates and they were just not quite liking it like they were saying to us in that way where you're trying to tell you go up to them and say ah what do you think do you like it and they're like trying to be polite being like yeah it's pretty good but you could tell you know their heart wasn't quite in it they weren't really feeling it and we had this kind of division right where somehow this casual bloody game we'd made had this hardcore fanbase

I'm sorry that's like it's not a really fair representation of our hardcore player base because the hardcore players are great right like they're what they're the ones on the forums telling everyone how great this game is and how everyone should play it they're our evangelists at the same time we have this casual market which is obviously makes up a huge amount of people and this mode right now is satisfying the hardcore players but the cash your players just aren't quite getting it you know it's become too complicated so it's time for us to start cutting start removing some stuff so we removed a fair bit of stuff the whole kind of shot back end and the currency system gone got rid of all of that it was just kind of making our results screen too clunky too complicated whenever a casual player would get to that screen you just see their eyes kind of glaze over and you know it there's a lot of button there's a lot of interactions going on there the other thing we did is we started kind of stripping back the amount of information that we were presenting the player to make it a bit more vague it's easy so like one of the

big things we do do is you know the frenzy bar at the top we actually removed that bar and we just let it play out for about seven seconds and we weren't telling people how far they were through it we started relying on people to just kind of intuitively get a sense for how long these things we're gonna last we did the same with the combo blitz as well and once we cut things there was all the sudden that it felt like this kind of new life and this these new opportunities before things have been complicated and we kind of been missing these these interactions we could do as soon as we cut these bars we could stick like fries and friends even honest together and that was awesome all the fruit pouring in from the sire and it's in slow-mo and it was totally about us and just by stripping things back and simplifying we'd allow out all these new opportunities the other big thing we did was you know we made the blitzing visibility really different before we'd had bars counting up on which again we're trying to do so much to explain to the player exactly

how it worked instead we came with this new philosophy of let's kind of hide it from the players of it casual players might not ever get a combo blitz and that's fine we don't need to have like this kind of empty bar on the side that's just sort of taunting them telling them they're not doing very good if they don't ever know about it then that's fine so we kind of rather than having bars we have this you know these lines in the background that are kind of quite faint and they fade away and slow down and that's how you know how much you have left we bring the score right up in your face because we wanted people to know that the reward they were getting but not necessary really how it worked or why and the great thing is it's casual players now that is playing along that's fine as they get better as they start stringing combos together they start to see this stuff pop up they're like okay I really want to get a better score I'm gonna learn this system and you know they spend a little bit of time figuring it out and you know for an advanced player it's fine they've got it understood and so even by changing the visibility and how much implement information we were presenting the player with that really really improved the design and then we

couldn't imagine the game without it at some point we were just spending all of our time playing arcade mode we've taken screenshots and like sending it to the QA guys and having these massive battles all of a sudden the productivity on the mode started to plummet because we spent six hours a day competing and it's kind of that's when it struck us that that's how we knew that it was ready and now whenever we add another major update to the game that's kind of our measure for whether it's ready or not once it's at the point where we can't imagine it without it where we go home and I use my phone that has the store-bought version on it and I go straight for that mode I'm like damn it's missing feature one two and three in the new version that's our metric for okay this thing's ready to go let's get this out there and get this out to people yeah if we finally released it yeah it took so long it took like almost three times longer than to make fruit ninja which is crazy so we poked a bit of fun ourselves in the trailer I only noticed when I grabbed

the screen shot that the Arcade Mode trailer somehow has five times more hits than the release trailer which is kind of nuts and this time we got a really really good reception again you can see like a significant sales bump there and then some really good sustained like held kind of velocity there I know now from using analytics that so there's four modes in the game arcade mode makes up 50% of the total time played on fruit ninja iOS so we learned a lot of lessons here as well we took a large risk by investing so heavily into this update and it ultimately paid off though we really feel like it was a better result than kind of eating it out and putting a version out there and then iterating our way to success which is ironic because that's exactly what we did with the very first version of fruit ninja but player expectations were changing and you know the App Store is just so crazy and so fast-moving that that was changing as well the other lesson we were learning

at this point was the bigger and the more complete the game is the harder it is to create a substantial update that adds to the game and doesn't detract from it so I have a picture of a pomegranate here I'm just gonna color this quickly rather than going through every single update we did we got to the pomegranate and that was oh geez maybe our 10th or 11th update title and our only spec really was add something to the game which will you know reinvigorated and refresh it and this talkest age is everything that we tried adding detracted from the game or didn't quite feel right even though you know the pomegranate is such a simple addition a fruit that you can slice a couple of times and then it explodes you know there's not much to it the game felt so full and complete and you know had such a sense of identity it was really difficult to add anything without disrupting the balance that was already there because the game just you know feeling like such a complete package all of the low-hanging fruit sorry for the

pun was already taken at this stage so this is sales and updates breakdown for the life of fruit ninja so again Blue Beam sales are insuring updates for scale on the side there I'm not sure if you can read it maybe if you're up the front is you know the number of updates you can see the blue there's like these two massive spikes that's Christmas Day 2010 2011 Christmas Day is a good days and you can sort of see you know the the spikes in updates right down the very bottom left corner there's like this tiny little orange dot and that was our very first update that like completely freaked us out and now you can see like how much the stakes have been raised on every update so that last update right before the second Christmas 2.4 million downloads in one day and that is a lot of stress and a lot of pressure for us when we're creating these up because you know especially when we're QA and it's got to work on retina and

iPad and all the different os's and all the different game center cases and all that it's like okay if there's a use case in here that has like 1% chance to crash in the game 1% crossed across that many users that's gonna be a lot of one-star reviews there's gonna be a lot of complaints and you know we really do our best to keep everyone happy which drives us slightly mad at at times but you know that's that's just kind of our thing so this is a really interesting scenario and be really interested to hear like after the talk if anyone wants to come and have a chat to me I'd be really curious to hear about this but for us we hit this problem where we're working on it and it's been the same team working on fruit ninja updates and by same team I mean like same three guys working on the fruit ninja updates for I don't know 14 15 months now and we're just completely shattered like we're fully d fruiting injured we don't have very good ideas anymore on you know easy ways to make the game really good and you know new

exciting kind of inspiration and because the market again has changed so much there's so many platforms so many devices so many you know trying to update it so it works at the multitasking and then iCloud and we're really you know my team we're really good at making gameplay and you IP but engineering side of things not so good you know that's really not where our strength is and we're putting all of our time into managing the tech so we eventually kind of made the decision that we were going to pass fruit ninja along to another team to work on so another team to do the tech stuff and another team to come up with the content updates and that was really hard decision for us like you know we definitely felt like fruit ninja was our baby you know we'd be grown very attached to it at the same time it was clear that we would no longer the best people for the job we've our hearts weren't quite in it anymore it really needed fresh blood so we passed that along to you know some other teams inside the company and again yeah I'd be really interested to hear whether other people have had that similar problem and how how people have dealt with it so

that kind of leads me into jetpack joyride originally when I got asked to do this talk they wanted me to talk about the jetpack Joyride updates I don't know that you're gonna get an hour out of me for jetpack joyride updates because we were making jetpack joyride and we knew how important updates were but there were a lot of new problems we couldn't ship lean anymore but app landscape was so different in the time that our fruit ninja had been out not only that but you know all the press and everyone's expecting something amazing from us because you know this is the the next biggest game we've really since fruit ninja so the pressure was on because of that you know we spent nine months in development on jetpack joyride and it came in it really close to twenty mag like 19.7 meg fruit ninja she originally shipped that 9 mag we had heaps are so you know we move right up against that ceiling of how much stuff we could add we tried to make it easy to update and really extensible you know we're like oh cool we will do this

vehicle system because then we can add new vehicles and this level system so we can add new levels and missions and we've really kept that in mind while we were working on it but the pure size and scope of the game made it so difficult to to update is exactly like we're fruit ninja or it you know had eventually gotten to the game was so full so complete is really difficult to add things we was out of space to do it to keep it in under 20 neg and we'd spent all that times fighting fires we had this currency system and it connected to the internet and all that and that was a huge drama and we spent all of our time you know chasing down problems so we actually announced update we just announced it yesterday here at GDC called the gadgets update for jetpack joyride we have spent four months on this update and I think that really you know really shows how different it is to the early day early days it sounds like it was so long ago even though it was less than two years ago for for managing updates and the way

that we're doing these updates now you know this is a major major release for us we're almost treating it like another game release we've got a you know we got trailers coming we got previews press releases release date all this kind of crazy stuff so that's kind of where we're at now with our updates and that's what we feel like we're getting the most out of as well as you know doing a lot of minor kind of chaser updates in between because we don't want to have too long a gap between them and yeah I mean we've got continued support we're still working away on other updates um there's so many more things now that we're trying to keep up with you know adding languages adding new platforms we have this crazy one we just worked on recently with Mattel this apptivity thing where you've actually got a sensei that interacts with the screen and ipad knows that it's this particular figurine touching the screen it's all pretty crazy and you know really good fun to develop for because it's just so oddball and different but we're still learning

you know that the up way of updating games is so complicated it's so specific to the game specific to the company we you know we've got a lot of answers especially early days you know we kind of figured things out but again it's changing so much with time we're constantly learning constantly coming up against new problems and new challenges we certainly don't have any hard and fast rules that we feel like everyone should follow but hopefully by going through some of these key updates and sharing with you guys how we went about doing it and the successes and problems we had with it that you all got something out of it so yeah thank you very much and we actually have 10 minutes for questions so if anyone's got any questions by all means please use the microphones so it comes up on the recording for the GDC volt yeah I was wondering how much worry do you have

about the announcement tomorrow and how much that's gonna factor into your plans moving forward I would be really worried about it if I hadn't been in like a permanent stressed State about doing this talk and I talk on Thursday but yeah I mean you know it's tricky and there's already people on touch arcade and all that showing their games at iPad Retina resolution without knowing for sure that's what it is it's hard maybe that'll work and maybe that'll pay off for those guys but you know we we have to wait and see what happens and take it as it comes right now you know we're so deep in development for a bunch of different things that we have on the boil that you know it's gonna be difficult for us to be super reactive to that sort of thing that yeah hopefully yeah it would be interesting to see what comes it's hard to know until what happens yes hi thanks for the great talk I was wondering about the bottie platform and divorce from Herrick are you using any of the motif air from engines or okay so so the

question is how do we handle multi-platform development and we actually have our own in-house engine that we use which is called mortar bricks and mortar come on I think it's kind of neat and yeah it's uh it's tricky you know I don't have that much to do with the multi-platform side of things I'm really focused on the design side but yeah it's a constant challenge because the platforms will have their differences and their kind of quirks and it definitely is a big challenge and you know we're doing our best to keep things multi-platform but yeah it's always challenge you mentioned in your talk that you initially didn't use analytics for developing fruit ninja hmm obviously you've changed your cycles lod now you know then you know 50% all that sort of stuff and you're obviously using analytics now how has it changed your process to begin using analytics as part of your design and was it heavily influential in terms of how you approach tunes which actually

ride yeah we use analytics and I mostly use them because they're interesting rather than using them to inform my design processes you know there's like the old old saying there's lives damned lies and statistics I'm definitely very much in that camp I'm sure there's a lot of good uses for analytics I tend to prefer to design like quite intuitively sometimes we'll get some interesting insights out of analytics a really excellent one that we did get was when we were setting the score cap on the light version of fruit ninja' having analytics there was a good way to know like you know what is the average score people are getting where should we set that should we set that 200 no that's too high not enough people get there but yeah really it hasn't influenced our design that much at all it's you know sometimes they'll be a very specific problem that it's good at addressing but a lot of times it's it's interesting and you know it can be used to inform our decisions but we don't use the analytics to yeah to really drive anything cool

just that the graphs when it had the updates versus the sales yeah they seem to be spikes about once a week for sales that would like to be a big increase and then it would drop off wondering how did that occur oh yeah I mean we see that across all of our apps I'm sure a lot of other develop developers see it as well straight up there's a big spike on Fridays and Saturdays and to a lesser extent Sundays as well I guess you know weekends people have got more free time more likely to to get the game so yeah it's a it's interesting and it's so uniform like you see it across all those graphs that's so pronounced so you mentioned how an earlier version of fruit ninja had the store system with currency and whatnot and what's funny is that jetpack joyride incorporates that so I'm pretty sure like I don't know me personally I find that system successful

and jetpack joyride you guys ever considered bringing it back in fruit ninja and actually having some sort of cross-pollination like support to where currency and fruit ninja kind of feck jetpack joyride and vice versa basically something that could run as a scan - you know how many other Halfbrick games that that individual has on their device at that moment you guys ever you know came across that thought in terms of using currency in fruit ninja at certainly you know something that we're willing to play around with and figure out when we did it last time we were unsuccessful and you know we didn't manage to get it right that's not to say that isn't a right way to do that implementation I just don't know what it is yet yeah definitely felt like it worked really well in jet pack and jet pack such a different kind of beast that it just sort of felt right for it in terms of cross pollinating of currency between apps I don't know maybe it would be cool I mean oh there's probably some pretty wild stuff you can do that's honestly something and never even

considered before sweet it's all yours run with it I think you said you are a a game designer not an artist yes that and so I was wondering if you could talk about the makeup of your teams and just like who you know what contributes to code and who you know has sort of the art direction and like whether that influences things sure thing below yes so basically the makeup of the teams there's a Steve what are you doing up at the microphone though so the program or a fruit ninja is that guy who's standing up at the microphone just now [Laughter] so we it's pretty simple we have a designer an artist and a programmer who will fill those roles and but the thing is we're so small like you know three or four guys working on it we get a lot of cross-pollination you know like we all have a say on the art style I had a lot of things that I was really kind of really wanted to do and you know these

guys who aren't designers often have really good design ideas and implementations and you know it's a very collaborative process and you know that's across all of our teams at Halfbrick as well we really encourage everyone to get stuck in and help out and you know sometimes some programmer art might make it into the game if the programmers got a bit of a gift we have some guys that are extremely kind of multi-talented you know everything so it really depends on the team and the project yeah Steve just wondering if you talked about how you decide what date to release each update like because I know there are certain things what we think about when we do that yeah I mean we're in terms of setting the day for when we want to release the update it really depends if it's a firefighter build and you know we're we're just trying to knock out bugs we try and make sure that happens as quickly as possible if it's a major update then we want it to kind of you know go with the we might want it to coincide with some other major releases

or you may be looking for a quiet part-time quiet patch to release it in it really depends on the exact time and the nature of the update but it's certainly something we don't just pull the trigger willy-nilly a lot of the time we have updates especially major updates ready to go far in advance give it a lot of time to go through our internal QA and then I'll even be ready and you know ready to press the big green Go button on for a little while as we set the dates and you know really do it properly doing like a proper game release I'm a huge fan of Jeb Toyota and food ninja as I'm sure many others are and so whenever I played Jeff I tried it I just was thinking about the design process that went into doing it because it feels like three or four games and melded together you have a slot machine all the different vehicles have different gameplay and just all that can you talk about the iterative process or whatever it was that sure yeah yeah so we announced jetpack Joyride last year at GDC and I went to at all cuz it was my first GDC last year I went to this talk and someone was I don't remember

what you talk it was what they're talking about this process called JSI RSO which stands for Jam in and rip out and when I was hearing this talk I'm like oh my god that's jetpack joyride um and that was really it like it was really a lot of ripping things out putting things in that's actually like a really good time for the last question cuz I can use it to pimp my talk which is all about the design process jetpack Joyride which is in this room Thursday 11:30 but yeah yeah I mean it was a lot of ripping things in putting you know taking things out and you know hacking and slashing design-wise to get up there so yeah thanks everyone for coming along and yeah