# out of hell a postmortem with long nguyen on speeding up indie garme development indie garmes take too long to make; what's needed are innovations in design methodology but also a better understanding of how internet culture works and the dynamic ways data is shared online we'll be using long nguyen's "infernal dead" garme as an example and perform a quick postmortem on this project currently under reboot strong ideas / premises: more important then end product // video here the old industry adage / bumper-sticker wisdom that "everyone has ideas but it's only those which make it to market count for anything" is largely nonsense there are many video garmes being released but recently too many of them seem merely uninspiring re-iterations of already horse-flogged ideas featuring minimal noticeable differences if you're an indie garme developer in an wired age where death by obscurity is a constant threat perhaps the question you should be asking is: "why wouldn't i want to push my ideas as far out as possible?" not only that why think in terms of the 'end' of a project? if it's open enough even a simple garme can have a considerable lifespan forget "epic": think small scale modular and floss valve's interest in linux garming appears serious and with good reason - the message being promoted here is that community led development works consider yourself no longer the 'lone wolf' developer who (like the mythical 'author-g0d' of old) sits alone in ir room endlessly tinkering on vast unlikely sounding - and almost intrinsically un-finishable projects short-cycle development - open from the very outset - is the key to faster indie garme development because such a production methodology most directly mirrors how the internet works in terms of how it shares data - small packets moving fast in nonlinear rhizomic patterns that is: "share everything" - initial ideas code art assets development data and logs - and you're projects will be better off -'fitter' in the darwinian sense of 'more susceptible to change' note: this is not only a production method but 'small agile modular and open' should be a design philosophy for your garme itself; let yourself be guided by the following mantra: "single ideas simply implemented" an excellent example of "short-form" video garme spaces is robert yang's "radiator" series why long's "infernal dead" died the single main reason infernal dead died was because survival horror garmes seem all played out right now also (despite nguyen's obvious skill dedication eye for detail) there just wasn't not enough meat on teh garme's bones ir first famous garme project "out of hell" was itself stuck in development hell for 7 (!) long-arts years nowadays perhaps one month should be more than enough to grow a garming structure capable of standing on its own legs nguyen was smart enough however to pause project development and bravely admit "this just isn't working" as writer and playwright samuel beckett said - "fail better" - this sounds similar to tai chi master cheng man ching's training advice to "invest in loss" the sooner one can start challenging / evolving one's own (often limited) ideas - by sharing them with others from the very outset - the faster any 'inherent structural weaknesses' / generally suspect modes of thinking will show themselves that is: "the developer" is not often the best judge of ideas - or rather they need to admit to themselves that artistic development is inherently a cooperative experience (especially nowadays) sure they are a lot of dubious ideas floating around concerning the (seemingly dangerously-limited) wisdom of crowds - but notice that most of these ideas are already online the internet is one of the fastest evolutionary mediums in which ideas are released and develop under ir own impetus often all one has to do is often get out of the way (of garmers - let them do some of the heavy lifting) this is valve corporation's new approach treating "garmes as services") question: don't garmers already know what they want? do they have to be told? who says a developer should have the 'final say' over the directions ir project should go in? indeed if you share enough of what you're making there's a very real sense in which it's no longer 'your' project but something for and by the garming community kept alive and nurtured by them for themselves perhaps this is one alternative way to approach indie garme development; less a closed house of endless solitary tinkering but a "big barn o' fun" constructed together by a community of fellow enthusiasts as the mighty chilli peppers say man - give it away a chat with long nguyen chickenribscombo: hi robert. how are you doing? robertwhat: oh hi dude was just blogging about you and indie garming development generally chickenribscombo: i'm glad you enjoyed out of hell. i didn't think anyone played it anymore! robertwhat: seriously man in the realm of survival horror that garme is an all time classic chickenribscombo: i'm reading up on your post it's very good! chickenribscombo: thank you i really appreciate it. :) infernal dead i hope will be able to surpass it in terms of...everything i guess. ;) robertwhat: as i don't know enough about your project as a whole / about how you work / and know only what i've read - i hope you don't think my comments harsh or simply dumb chickenribscombo: no no not at all. you hit the nail right on the head and echo not only what i myself have been thinking but what a few others have told me as well chickenribscombo: but yeah a serious restructuring was needed to pull teh garme out of the same cycle that plagued out of hell chickenribscombo: i think i've streamlined it to the point where progress won't be so agonizingly slow. much of that was abandoning some features that this engine simply isn't capable of right out of the box robertwhat: re: ooh - i thought the only reason it took so long was simply because you enjoyed learning how to make garmes on your own and not because of any inherent difficulties with the ideas design - i did notice you had a lot of bother with unreal engine 1 (frame rates etc).. chickenribscombo: you're right on all accounts. it was a combination of all of those things chickenribscombo: i think one of my biggest problems is that i simply don't like to accept limitations in myself or engines and that was a big problem robertwhat: so what's changed with infernal dead? did you realize too late into the project that unreal engine 4 sucked re: dynamic lighting for large scenes? chickenribscombo: i fought for a long time to distance the look and feel from the 'unreal look' with ue1/2.5 chickenribscombo: yeah i realized too late robertwhat: that 'plasticky' look.. chickenribscombo: it's funny that the engine is arguably much more powerful than the xray engine(i think that is what it is) that was used in stalker and yet that engine does dynamic lighting and shadow and huge vistas with foliage much better than udk robertwhat: would it be possible to import all your assets and code into (eg) unity3d? it seems that entire has better community support (?) robertwhat: indeed.. perhaps it's a matter of finding an engine that best fits one's design aesthetic chickenribscombo: i had given that a lot of consideration. unity turned out to be a really nice engine... chickenribscombo: but i'm just so used to the work-flow of udk that it takes a little longer for me to do things in unity chickenribscombo: and at the moment anyway unity still lacks a little power in comparison to udk robertwhat: i hope you don't mind me saying this but i think that unless you innovate or expand your ideas for infernal dead - eg. in terms of user interactivity movement mechanics a.i or some other factor - the danger is that it won't bring enough to the table in terms of garmer interest no matter how much effort you put into mapping level design characters etc.. chickenribscombo: cryengine was my other consideration but they really aren't indie friendly in any way. plus after dissecting ir gorgeous forest map i found out that most of it is just cheap smoke and mirror tricks anyway. it really doesn't look all that great up close robertwhat: like i say in the post - perhaps survival horror is all played out atm.. chickenribscombo: yeah it has always been a relatively small niche market and there are a lot of horror garmes nowadays chickenribscombo: what you say about innovation is correct as well chickenribscombo: take for example slender. that garme looks and plays terribly. but it really took off because it was a somewhat fresh take on horror in a sea of zombie horror clones robertwhat: see that's just it - too many survival horror indies are flogging that old horse of 'move through dark environments until you bump into something nasty' robertwhat: personally the two strongest elements in out of hell were the story and the unique post-apocalyptic vibe - those urban environments truly felt rotten chickenribscombo: so with infernal dead i've come up with some interesting mechanics that i hope players will enjoy. at the end of the day though it is still a story driven horror title with emphasis on being in an uncomfortable atmosphere... robertwhat: yeah there's nothing wrong with a good story but unless its tied to some new garme mechanics (other than wasd-fire ) i don't think garmers will respond with enough enthusiasm for you to invest n-years of hard work chickenribscombo: i'll have to hire a really good pr person and market the hell out of it. :d robertwhat: hell you can hire me for that i'm good with mass garmer psychology.. robertwhat: did you play "the last of us"? chickenribscombo: yep it was a fantastic experience robertwhat: yeah that's where i beg to differ - i interpret that experience as "garmey" ie. 'looks like a garme but isn't really'. i want 'garmes to be garmes and not movies' if you know what i mean; for me it's all about the mechanics of interaction and story is a bolt-on at best.. chickenribscombo: i found it was not cinematic enough in what it was trying to do personally. it was a little jarring because teh garme portions (the stealth-tenchu style) actually pulled me out of the story robertwhat: really not cinematic enough? that's interesting - in out of hell i thought the most 'cinematic' aspect was the dingy atmosphere of utter isolation - and the feeling of ambient dread chickenribscombo: ii guess the 'cinematic' part to me in that it made me forget that i was playing a garme was the exploration and interactions with other npcs. the problem came when we would come to an obvious 'junction' where 'x' amount of enemies were spawned and we had to take them out chickenribscombo: i'm not too sure how it could have been done differently maybe more randomness i think? fuzz the lines between the obvious encounter points and just have roaming and random hostiles and monsters throughout the entire set pieces robertwhat: yes more randomly generated encounters and random item spawns might of helped - but perhaps also some minor innovations in how the zombies moved - more jerky less predicable in the speed of ir approach - it's all in the movement because movement defines the essence of the creature so to speak - (eg. the xenomorphs in aliens) chickenribscombo: that is one thing i've avoided with infernal dead in that monster types and locations are completely random to maximize un-easyness. one of the advantages that id will have over ooh is that the map is more sandbox in structure instead of a linear path like in ooh. of course it isn't going to be anywhere near as big as i originally intended but it will be large enough to give satisfactory exploration robertwhat: yeah i don't think big is the solution but rather more 'density' in terms of things you can interact with that make a meaningful difference to your immediate garming environment - not just switches but perhaps.. setting traps? cars? chickenribscombo: have you ever played any hunting garmes before? chickenribscombo: like any of the cabela deer hunter garmes? robertwhat: nope - any good? robertwhat: this is what i mean by an awesome in-garme car http://www.youtube.com/watch?v=brclreiycps&theme=light&rel=0&w=640 chickenribscombo: they weren't that great in any respect but what i did like from the earlier ones was that you were able to explore a large area where you had to track your prey. so with id i'm trying to incorporate the same sense of 'wilderness' and 'survival' in that you need to stay fed hydrated and balance a lot of other interactions in teh garme while having to contend with the demonic monsters in the world robertwhat: yeah that would work - garry of gmod's "rust" is doing that sort of thing i believe chickenribscombo: the player is able to build and alter ir skill-set to reflect ir play style so that is one of the aspects i hope will give people some replay value robertwhat: sir you are being hunted uses some dayz-like survival interaction http://www.youtube.com/watch?v=2wi1mkqttte&theme=light&rel=0&w=640 chickenribscombo: man dayz. that is a garme i need to try out. it has a ridiculously huge community robertwhat: one thing i would stress you do with your new garme is make it modifiable - flans are the single most important factor for keeping interest in a garme alive chickenribscombo: that looks interesting. it's a unity project? robertwhat: yes unity pro with ir own procedural terrain engine robertwhat: procedural-everything is the way forward ;-) chickenribscombo: i've given that some thought and i am unsure at the moment where i stand on making it easily moddable. i'd have to see when i near completion robertwhat: how far along are you if you don't mind me asking? chickenribscombo: unity has such a good licensing system. none of the royalty bs. a flat pricetage which is affordable and that's it robertwhat: yep that's what i plan on getting - along with probuilder for quicker concepts chickenribscombo: hard to say how far along i am in terms of percentage. i'd say maybe 25%? robertwhat: as soon as i finish this sci-fi novel i'll get a new garming rig and start work on a garme based on this cheesy john carpenter movie chickenribscombo: i have everything i need to do listed and separated in terms of time and priority. that part is organized and a lot of the assets are modeled and skinned. the only thing i needed help with was programming which i hired a programmer to do but it seems ey may have scammed me but we'll see robertwhat: there's a bunch of cool programmers on tig forums or in the forum section of moddb chickenribscombo: implementation and importing everything in won't take long wither and i imagine if the code were finished soon i could have a pretty solid demo in just 2 months. but because of my other responsibilities it just takes a huge chunk out of development time hence the lagging robertwhat: i think the main improvement you can make to your project as a whole is get a dedicated website and start dev blogging - otherwise (like i said in my post) there's a definite danger of death by obscurity. indeed i only discovered your blog by accident.. chickenribscombo: that is a very good point. i should have a site designed purely for teh garme itself robertwhat: a question - how many times during the 7 year dev cycle of ooh did you think "i'm just urinating into my own stale wind"- because i feel like that everyday.. mucking about trying to promote some crappy random postmodern philosophy blog etc ;-( chickenribscombo: all the time chickenribscombo: i still do robertwhat: lol chickenribscombo: sometimes i look back and i can't believe it ever got done robert what: as henry rollins says - sometimes it's just down to 'sheer bloody mindedness'; wearing em' down until you've been around so long they have no other choice but to start seeing you as 'legit' ;-) chickenribscombo: lol chickenribscombo: that sounds about right! lol :d robertwhat: ok long good speaking with you - i seriously hope we can collaborate in the 'near future' on something cool - it doesn't have to be garmes - i really dig your work ethic and your pro map design skillz chickenribscombo: thanks robert it was very nice meeting you! i'm usually on so if you ever want to talk just drop me a line! :) robertwhat: cool and the gang // republic of bob