The Pre-Production Problem - How to Improve the Planning Process in Game Design - Extra Credits
ฝัง
- เผยแพร่เมื่อ 28 ก.ย. 2024
- Pre-production in game development has an unfortunate side effect of hurting programmers and killing productivity. We can look to the animation industry as an example of how to improve pre-production practices in game studios.
Subscribe for new episodes every Wednesday! bit.ly/SubToEC (---More below)
(Original air date: July 27, 2011)
Check out Molly's artwork here!
www.mollymalone...
_______
Get your Extra Credits gear at the store! bit.ly/ExtraStore
Play games with us on Extra Play! bit.ly/WatchEXP
Watch more episodes from this season of Extra Credits! bit.ly/2n8QcIn
Contribute community subtitles to Extra Credits: www.youtube.com...
Talk to us on Twitter (@ExtraCreditz): bit.ly/ECTweet
Follow us on Facebook: bit.ly/ECFBPage
Get our list of recommended games on Steam: bit.ly/ECCurator
_________
Would you like James to speak at your school or organization? For info, contact us at: contact@extra-credits.net
_________
♪ Intro Music: "Penguin Cap" by CarboHydroM
bit.ly/1eIHTDS
♪ Outro Music: "Gypsy Jazz" by Adrian Holovaty
ocremix.org/rem...
Programmers should be doing prototyping during pre-production! Give us the freedom to create one feature at a time, learn what works - and then the freedom throw out that code entirely, try again and again, until we can do it RIGHT during production. Of course nobody wants to pay a programmer to write code that's not going into the product. But it reduces the number of bugs we have to fix later and improves the overall final quality.
there lives a problem....the ideas from pre-production maybe don`t match with programmers ideas. So, its necessary a well established communication and a lot of time (tell me if Im wrong)
For that matter if you've decided on an engine just let the programmers go wild familiarizing themselves with the engine and middleware as much as they can. Demo projects, study time, whatever floats their boat.
Your suggestion reminds me of a kangaroo's reproductive system. At any given moment, a female kangaroo in her reproductive prime will typically have three joeys in development - an older joey who has left the pouch but still sticks his head in to nurse, a newborn joey who is perpetually nursing, and a fertilized egg that is paused waiting for the right time to implant. By the time the older joey weans, the newborn is ready to leave the pouch. At that point, the fertilized egg is unpaused and the pregnancy begins. Immediately after giving birth a few weeks later, the female mates with another male and gets another egg ready.
+Ettina Kitten Not at all the metaphor I would've expected, but it seems to work! XD
+Ettina Kitten So kangaroos are eternally giving birth? why arent we being invaded by kangaroos yet?
+leonidas56 They don't have the military prowess of Emus.
+leonidas56 Because Australlia.... It's sort of isolated....
+leonidas56 Because young kangaroos also have a lot of predators after them? Generally when a species have a high birth frequency, it's to compensate for something.
Ironically, I've been using a lot of my real estate investing tricks in starting my first game. I'm in the preproduction phase right now. Using your videos to help brainstorming, using a format I developed to develop blueprints and things-to-do lists, and while I'm still unsure of what engine I will use, or what art style will be chosen, I'm already talking with other producers to get quotes and make arrangements so that when it's time to hit the ground running, I can throw jobs at people who are well suited for the gig, and I can focus on my tasks without getting bogged down. As a small fry, I do have a personal rule to never undermine or lowball my team. Anyone who is worth their salt deserves respect and remuneration. And there are a lot of creative ways to structure the finances of an operation.
I like the way Pixar does things. This is definitely worth a try. This is a problem I don't have to deal with personally. I don't have the kind of time or money pressure large companies have. I do a lot of preproduction. It was fine at first, because it helps me learn. Then I get to the point where my ideas are as good as they will ever be. Then I am ready to build it. For example, I explored so many ins and outs of a battle system, I am ready to write all the rules down together. It haven't done this yet because I am still working on creature design. Still the time will come where I can write the battle system. Then I can go from there. I can develop story, items and the geography.
Here's my thought:
The difference between games and animation or film is that the product lives and changes after its release. Bug are discovered, weird story dead-ends that players discover have to be addressed, black hats hack into a game and it needs defending.
My idea, then, is this: After a game is finished, designate some time for the original programmers to head up customer support and debugging. Don't put the leads on these tasks, or if you do, pick other people to be leads on the next project. Create a window where the original programmers are the ones handling not just the programming corrections and bug fixes, but the contacts from customers.
While this is going on for say, 2-3 months, begin pre-production on the next game with a handful of lead developers.
Customer support can be stressful, but it's a different kind of stress versus the initial development process.
The problem here might be te cost of having a software developer be... Well... Not developing software. Not that it wouldn't be great to do strong and fast debugging, but I'm guessing it isn't done because the cost is too high.
Also, if I were a company I would have pre-planning done on the next game by the time the current one is done, like EC suggests. It would not only increase output quality, but even give a chance to train the programmers that stood out in the previous project to lead future ones.
Still, debugging exists. Does anyone know how that mind of post-release work is done currently?
Manuel Perez But that was the point of the piece. You can't do pre-production at the end of the production cycle because there's too much work to do, and you can't do pre-production after the game is launched because you end up with idle programmers.
My idea gives idle programmers something to do. And while they aren't programming something new, they *are* programming. And likely creating more satisfied customers in the process.
Eliot Hochberg The problem with that is that you're assuming the company has that type of money to spend debugging software past their requirement time. No matter what, the programmer is getting paid. And if the studio is paying a programmer without themselves getting paid, they are actively losing money on a programmer who is, in essence, making them no money at all. Unless it can be demonstrated to the studio that the cost of debugging past release would be worth it (Battlefield 4) and would garner more sales, they will more than likely be unwilling to do so.
Eliot Hochberg You're assuming that they don't already test games with Alpha's and Beta's. The issue is that Publishers will often have set dates for a product to be finished and if it's not finished they lose like millions of dollars. So you basically have to ship before you're ready which is why we get shipped games that are super buggy on release but get cleaned up within the first few months. I also think people misunderstand the proces of debugging its not like you find a bug and there's just some magical fix. I think that if people better understood the process with which games are made they would be more sympathetic to developers.
I don't see how my solution is prevented by your response.
Even with alpha and beta testing, bugs get through, precisely because of the schedules you reference.
So why not have the bulk of the programmers continue to handle the bug reports and fixes, and peel off a select group to begin work on the next project while the initial bugs are identified and corrected?
FRC taught me something this year that may solve the "what do programmers do" ordeal. The truth is, _anyone_ might just have a grand idea, so getting literally everyone involved in the project (or at least everyone that's involved before hiring the masses) to brainstorm and grind out preproduction ideas is helpful. I myself am a programmer for my team, and I nearly had the final idea for our robot this year. My point is, programmers don't jist have to program, same for anyone else. In my humble opinion, *_everyone_* should be involved in the very first planning stages
wundrweapon I like your idea (was on FLL and a lot of my friends are in FRC and FTC), but I don't think that system will work for the game indestry. you see- the indestry is what it is- an indestry. a proggrammer in FTC is still a kid, who did not do any graduations or things like that which means they are in (sort of) equal skills for this. Whatsoever, in the gaming indestry, people actually spent years getting degrees in these things, and some people with some other degrees do very different jobs, for better or worse.
wither way, valid point and nice thinking!
btw, not native english speaker, so sorry for any mistakes....
Your idea carries truth, but is not practical or efficient.
Imagine a school that is ran by all members' idea, including the cook's view on how kids should be taught physics?
Small teams work fine in this system, but as you develop an enterprise with clear-cut jobs and a massive pipeline, having too many people speaking their minds start to hinder production past the healthy point of brainstorming.
Marcotonio you do bring up fair points, but there's another clear line you're missing: the game devs vs. most other industries. in game development, everyone gets involved in it in some way. the programmers, designers, artists, etc. are all making the same thing. a school chef is not a physics teacher, so they're not involved in the same closed-loop system. that's why robot building, game development, etc. have this advantage - nobody's job is too different from the end goal.
and besides, not every view needs to be _incorporated_ but rather considered
Fair enough! I agree every branch of the pipeline has some grasp on all other areas. But I still think it's not feasible to have this much interdisciplinarity in all departments. Some workers are bound to be codemonkeys, some will be drawhorses, and even though some of them might have good insight, some of them will also have poor tact for the overall balance in things, being good only at the task they are paid to.
I'm an cinema/animation major with some dabbling in videogames and I found out most visual artists and programmers I've worked with had really poor self-evaluation when it came to brainstorming. They love the media, they consume it and even produce it frequently, but they lack critical sense that a person more focused on, say, storyboarding and screenwriting (for a movie) might have. That's what happens in the Japanese industry of anime, for instance, where the working force (animators) has low standards in matters of content: i.imgur.com/BlHeP67.jpg
If these guys' opinions would be taken in account, hours and hours would be wasted convincing them their original character SHADOW_MAGEXXX666 the Hedgehog is not a cool addition to the game.
What can be done is having "chiefs" of each sector working close together on the pre-production phase, but that still leaves the main bulk of more task-oriented workers out of the equation.
The solution to that, I believe, would be better training these skills by promoting internal projects and letting them go wild where there isn't money at stake, but that leads us to another problem.... which I'm discussing on this tree, if you'll join me: th-cam.com/video/ukADFPuscG8/w-d-xo.html&lc=z13qifugjoy4fpaxr04cfpvhevi3y14g0zk0k
I just wanted to pipe in and add that I agree with the OP to a certain extent. As a dev myself, I run into this problem a lot when I'm doing contract work on a project that had all the preproduction done by non-programmers. Sure, in a large studio, the people working in preproduction can have enough experience to know how to consider the technical side of things, but I have NEVER worked on a project where preproduction was done entirely by creative folk that worked out well financially. The whole team wastes way too much time and money and it hurts the game. On the flipside, the projects I follow (never worked on one though), that started with all programmers tend to run smoothly, but don't seem to break barriers or innovate on creative gameplay that well. There's something to be said for dreamers who won't take no for an answer.
A little long-winded, but basically I wanted to chime in and say that bringing people in on all levels for preproduction could solve a hell of a lot of problems for small to medium size studios. And in large studios, bringing in a select few proven individuals from all experience levels and departments might not be a bad idea either.
..Can't we just get a seperate team JUST for pre-production? You have your animators and modellers working on the game models/animation, you have the programmers working on the engine and mechanics and such, etc. etc., why can't we make a group of 'pre-production'-ers to do pre-production work in the meantime, for any upcoming project? You don't need anyone to do any other work, everyone can specialize on their own thing, and the pre-production-ers can work on the pre-production for an upcoming project. What I'm basicly saying is, hire people specifically for pre-production WHILE you know there's enough time left before release/starting a new project.
Likely because production groups need to be able to consult on the viability of stuff introduced in preproduction: what if something is planned that would take much longer than estimated by the preproduction staff or if something is flat-out impossible?
+RhythmGameAddict Those problems still come up without preproduction and cost in terms of lost work (unused code and assets) is dramatically higher.
However, you make a good point about the diversity necessary in a good pre-production team: have a designer who codes, one who's done sound, and one who'd done art, etc. (preferably people who acquired those skills on your team). Then they can be an advocate for the different teams that need to make all those ideas happen.
Also, no one says the preproduction plan is set in stone, it's just complete enough that everyone can start work on the game, while the design issues that cause disagreement are modified.
+Cyler s, +Machin Shin, +Marcus Mattern, +RhythmGameAddict, +Cowfaic'd Realm, +Extra Credits.
For the smaller studios they could higher people who are just learning their craft which could either be hobbyists who participated in a game jam or someone who just finished their training/school. They would work back and forth between the Pre-Production and the programmers during the current project. In this situation these people can take on more simple roles as they are mainly there to keep communication going between the more experienced workers.
My reasoning for this is they are able to learn how games are made as well they can learn if they are truly interested in the position they were studding or if they seen something else that's more interesting.
For the Massive studios they can use people with more knowledge as well as some basic experience. They could have just come out of school or been apart of smaller games.
Lastly hen the current game is getting ready for publishing and the new one is about to start production this group can move up into positions they would like to work for. You could also cycle some of the other staff to be apart of communicating or Pre-Production if they feel burned out from finishing the last game.
+Cowfaic'd Realm It might just be me, but it seems likely that changing the head of a project from the lead pre-production designer to the lead production designer could easily lead to details being lost from the "vision". Trying to teach someone else to create something you imagined probably isn't as easy as creating it yourself. Besides, I doubt the people pre-producing something like _Bioshock_ or _Portal_ are eager to hand it off to someone else...
You kinda get two? Two main problems.
One is expertise and familiarity with a given project. To an extent you can transfer the time spent on it to new people but that's going to take time itself and it's going to lose a lot of the tangential ideas that weren't there for the final concept that could come up and be useful. Even then like the people who did the pre-production are still going to be the foremost experts on the game for a lot of the game's early development so that's just lost time.
Like you /can/ do it but you're bleeding man hours basically.
Two is ownership. It's maybe not talking about a lot in education but it comes up a good amount in progressive business in my experience at least. People who feel genuine ownership for their work rather than that they are just a cog in the machine tend to be more dedicated and more productive and produce better products. By switching out the core cast of people planning the game you really hurt your ability to get that ownership.
I really like a model that's similar to the pixar one he mentioned in the video where you have core teams rotating (Pixar specifically has a few sets of core teams that rotate) but where instead you develop those cores from within your staff for each project. While multiple projects are in production you can have designers or other team members pitch projects and if any get approved they can shave off and begin pre-production, then as their pre-production ends you phase it into full production aggregating staff from either company growth or other projects slowing down. It means you can't have the full staff crunch time and you have to depend more on strong producers to wind and control projects but it gives every game exactly what it needs and with regular reviews on whether it needs to be stopped and the staff phased into other projects.
If there are people such as the programmers not working during pre-production, why not include them in the world design/story design process, in order to get them better situated with the possible mechanics they're going to be working with? After all, the games industry is based around the idea that good and specific communication ends up with more efficient work.
I'd be inclined to say that the answer to this may lie in DLC and perhaps accepting a bit more overhead in the design department.
DLC offers the whole team something to do post-launch but it can be scaled to avoid requiring some designers, programers and artists to split off and move into pre-production. This isn't the case with all DLC but if you look at Eugen's free DLC for their Wargame series I think it's clear that they're designed to avoid requiring a whole team while still giving some programmers, designers and artists something to work on. Obviously designing the scope and workload of each DLC product to fit the reduced workforce will be all the more important, feature creep won't just be bad for the scheduling of the DLC launch but possibly involve compromising the pre-production going on and thus the whole point of the exercise.
Given this imperative, and the fact that design staff are going to be spread thin during this period I think accepting having a few more Game Designers on staff would be needed so that there's enough to guide the post-launch DLCs and run pre-production effectively, and I suspect that the number of added bodies involved will be an order of magnitude less then the would-be idle programmers. I can't imagine that during the production run that it would be impossible to find enough to do for a few extra Game Designers, if nothing else having one or two work on laying the groundwork for DLC plans or for pre-production should do.
The side benefit of this is that a programmer-heavy DLC team will, in a well run Development Studio, be able to fix bugs and the like more aggressively then I imagine a pruned-down 'patches only' team would be able to so that games with a lot of potential but a buggy launch ( looking at you Rome 2 ) can be more easily brought to a reasonably good end-state. Also there's a lot of potential for free DLC aimed at providing/improving mod support for games well suited to that.
Someone should put every episode of extra credits on to blu-ray discs and ship them to every video game related person or company there is. developers, publishers, consumers... they could all benefit from these videos.
The whole "what do you do with the programmers" issue reminds me of my school's Robotics club... where I'm the lead programmer. We spent, like, three months in the design phase, where I did practically nothing. I looked up stuff on the internet when a teammate needed to know the dimensions of something, I did research on the sort of tools I was gonna be programming for... but those were all quick things that I tended to finish in five minutes and spent the rest of the hour that we met that week looking for the next thing to do. It wasn't until stuff actually began to be built that the team had any idea what they wanted me to code up and I actually did anything useful. And, believe me, I tried to help during the design phase, but I was put into my position because I'm just a natural at programming- I have little relevant experience with designing robots, meaning that I didn't understand the relevance of most of what they were arguing about. Whenever I tried to help, it was just me asking a stupid question in the hopes that they'd somehow missed that minor piece that made an entire idea work or fail.
It kinda helps to know that the issue of unused programmers is an issue in the field I actually want to go into, too, not just the field I'm in now. It motivates me a bit more to help find a solution.
What if...and this is a crazy idea...but what if you have short meetings or open-door policy on ideas so that while you work on smething, you could run down or call in to tell someone about an idea or thought about the next game?
they addressed that issue, hiring from project to project is a nightmare for both employee and employer.
(cont.) Skyrim uses top notch gameplay with modern stuff. If you think about it, since they already have some quests down and the main stuff, but not too much on gameplay, this might mean they're coming to a close on TES Online, and maybe doing Elder Scrolls 6, or a new Fallout even. This is just an amazing thought, knowing these guys are doing so well because they're doing what this guy here said!
Midway through the video, I had already conceived of a way to solve the problem; immediately afterward, you described how Pixar has done so, which was my plan exactly. I don't really see why it would be that difficult to set up a design company in the same way Pixar is but, granted, I'm not a designer, so I might be totally wrong.
Every time I end up watching an Extra Credits video I end up splurging on them for hours and hours. I want to go play some of these games but i keep learning.
You know what's cool? James (the "mastermind," although Daniel writes the scripts with him) *has* taken the initiative in taking these messages out to the gaming industry.
As someone who has studied communications and media production, with a program emphasis (not my emphasis) on film and documentary, it seems strange to me that the assumption is having all those programmers and unused workers would be paid to not work? Why can they not remain employed at X Company without being paid for their down time? Tends to be how in-house film crews work.
When film crews are in pre-production, even for entirely in-house productions, the production crew is not being paid because they are not working. Ideally, the pay they receive makes up for that down time. (Unless they are a jacks-of-all-trades crew of course, in which case they are probably giving their input and being paid for that input)
But I guess it makes sense?
I feel like my ideal situation would be to have producers or what-not working on pre-production for possible future projects while current projects are in development. Which I suppose is similar to that animation studio workflow.
+UnimportantHero I think this can't happen because of contract and this kind of stuff. It would be something too hard to organize, it's not optimal, so, not going to happen.
Lucas Couto I am guessing it has something to do with payscale. A lot of folk in film and television production can afford to work from job to job (though lower tier workers just barely so) because the pay is union rates and set with a 'job to job' lifestyle in mind. From what I understand, game devs are not really paid enough considering the money their products make.
Looked at those figures at 2:35. Gave me another good reason to become a programmer.
2012: I can't wait to be the next Blizzard!
2021: ...no comment.
What if you had specific people who would work on the game at the start and towards the end they would work on the pre-production? If they knew that they would do it, and did it every time, the problems with inefficiency and wanting to squeeze in more features might both be solved.
Another option is to have a new kind of company. A pre-production company that sells their IP to game studios and then consults on the development. Or they could hire a game studio to implement their IP. Once such a company garnered enough reputation, they'd do very well with this model.
I a coder in the game industry and have no problem with it. The developers would benefit because this way they are more insulated from the publishers. Publishers are already ruining the industry and something has to change. It was just a thought, not a proposition, but my idea is already in use in other industries. There is no reason it wouldn't work in game development.
Running a game company like an animation studio is a great way to look at it.
1:56 Good ol' XKCD.
the weakness of this is probably gonna be the inability to predict if the game that's being planned while another game is made is going to succeed neither will the game currently being made will succeed or not
Naughty Dog does this now. They have two teams, and like Pixar they have one team working on the current project while the other is working on pre-production. It's only started recently, but I think it's fair to say that The Last of Us speaks for how well this system works.
Adrian Holovaty for the outro! Woot!!
The name of the outro music is "Sweet Georgia Brown", not "Gypsy Jazz":
Wikipedia article: en.wikipedia.org/wiki/Sweet_Georgia_Brown
Example on TH-cam: th-cam.com/video/JkxrCttCcHg/w-d-xo.html
Extended pre-production may lead to over detailed planning which makes production process less flexible. When something goes not the way it was planned you can throw away the plan cause the rest of it depends on that thing. In other words best pre-production is minimal.
Thats why i love being an indie developer XD i got all the time i need hehe
ya know while i was watching this i remembered an interview about the dev time in the newest Zelda game (breath of the wild?) talking about the justification of the amount of dev they used (over 300 or something). if these figures are right then HOLY CRAP THAT WAS AN EXPENSIVE GAME! how the heck do they make that back!?
How have I not considered this before?
If I had a solid solution to the pre-production problem there's no way that I would share it with the rest of you. I'd keep it to myself and be on my way to becoming the next Valve, Blizzard or Bioware. You can learn it after I had turned out a few successful projects and gotten my foothold at the top of the industry.
Mentioning Games with a long Pre-Production Time and Half-Life 3 came to my mind immediantly... =D
"Gypsy Jazz" reminded me of Gladkov's themes for Plastilinovaya Vorona animated shorts :D
How about game companies start Pre-Production arms? The big game teams get the time they need to make a game as good as they can and all the while a small group of script writers, artists, software engineers, and if the company really supports this game, market salesmen work to build the foundation of the game. When the pre-production arms hit their deadline or think they've finished, that's when all the big game teams start to really develop the game in question, and since the foundation has been laid there should be less fighting within the team because they can't argue about what physics engine to use or what art style to use because it has been made for them. Game Teams will still argue over stuff like names of places and important characters, but if I am right newer games can be closer to being perfected right off the shelf.
I love the planning, story-making and worldbuilding. So I like making worlds :D
a search engine is a VERY different beast than a game like WoW, Warhammer, or literally anything else
So been playing games since ps2 and been craving to work in the industry for years most likely as someone who turns a creative persons ideas into reality so whats the industry like
What if you had your team or pool of programmers and two development teams, one keeping the programmers busy, the other one planning the next project? Or would that shift the downtime towards the development teams too much?
Where is that demotivational from at the beginning of the video? I want to save it somewhere, it's hilarious for me.
the biggest problem of companies is to improve things rather change things all the time.
they should think about quality first rather than talk about money and profit.generally they use dlc and reboot as an excuse to justify some problems inside game.
Remember when Michael jackson died in 2010? People were so unsure of the facts, google was inundated with searches like "Death MJ" and "Michael Jackson death" that google crashed.
I remember that it was mainly due to some 10 million poeple searching for two words; Not attempting to connect to connect to what is more or less a MMO all at the same time.
TLDR: Google has crashed, and as recently as 2010.
They could work on DLC and expansions for existing games
Instead of planning how they are going to chop up the game to sell in pieces before its released, they should get these guys with nothing to do to work on additional game content for another game already released
You realize i'm not able to see every response you get, right?
And there is a lot more information involved in just making sure everyone is seeing the same thing on screen at the same time in games than the definition of a word, or a lolcat video. There's physics, geometry, and depending on the genre, hit detection (FPS in particular). So no, I don't think the 2 things compare.
Bear in mind I don't know all the shit google has working behind the scenes or anything. Its a youtube comment section
Budget: Whatever you can pawn off kickstarter :D
I WISH all of animation worked like Pixar. Go to television animation, it's just like film. Sucks having to find something every time your project goes on hiatus.... especially when you're not in LA and so don't have the Animator's guild helping you with health insurance.
So *THAT'S* WHY INDIE GAMES ARE BETTER!
I don't think you can flat out say they're better.
*****
Just go with the joke, man. ;)
XerxesTexasToast I would love to but then I remember GUISE OF THE WOLF and DAY ONE GARRY'S INCIDENT exists and I forget to say indies are better because I am too busy trying to keep the scissors from my jugular XD
EDUCATION! That's your solution. Plan for training for your developers. There is ALWAYS something new to learn. Cross train people, train on new technologies, let them be creative with hackathons or brown bags. Problem solved.
Exactly, man. The game industry is very messy. We need some chances to learn how to be efficient workers.
In most cases proper education on most topics for all of the programmers would still cost quite a bit of money, not quite as much as the downtime though. It would definitely work but to save money I think the best idea could be found if you used that in tandem with other efficient but not truly complete methods of solving the problem.
L D B Well, I was thinking less like a college course and more like a game jam-esque situation.
I agree with the game jam idea. You always learn something new, be it about your specific role or the overall process of making a game. Plus its fun!
While I would really love this and the next project ahead will benefit from it, this also means that the downtime in productivity for your developers will be even more expensive ...
And we know our dear gaming industry, they don't want to think a few month ahead and see that this might be money well spent.
"The next Valve, Bioware, or Blizzard"
Wow, remember when they made good games?
(Half Life Alyx was cool though)
Yeaah the Blizzard part didn't age too well.
I think this explains why there's no Half-life 3.
+Olors SixtyFoar Wow, like, actually though. I heard that, the way things are set up over at Valve, anyone works on any project they like, and there has just been very little traffic on Half Life 3.
+X-Blade Wielder I need to get into valve
+Olors SixtyFoar no it's because valve is allergic to the number 3
omer shaik Gaben: "One, two, th-ACHSHITSNEEZE"
***** I guess you could say it took... *forever* to come out...
I'll be leaving now
I know I'm by no means someone to look to as an example as an unemployed college drop out who hasn't finished a single project yet due to scope creep, but here's something I'm trying with my development "team": I always have at least two projects in development that I drift between. Not necessarily two games, but two projects. I just poke projects and am basically in a constant state of pre-production for any given project, but always ensuring that at least one is active at any given time. Skill-wise, I'm low productivity on just about anything because of my lack of focus, but I'm pretty good at constant outpouring of ideas and good-naturedly pestering people. Maybe it'll work out well for us or maybe we'll crash and burn. If it doesn't crash and burn, then I guess that's somewhat of a solution to this problem. Problem is, I don't think that sort of model would scale well at the middling level of game development. Indie dev it might work, big developer it might work, it's basically just having a pre-production specialist who coordinates multiple smaller teams. Any mid-level company without the freedom of indie dev to take risks or the budget of a big developer might not be able to make this work. I'd be interested to see someone try though, cos people like me are probably unemployable in traditional roles of game dev and it'd be nice to see a field open up for people like me.
I'm always amazed that you compare yourselves to film, tv, and other forms of media. Why I don't disagree with video games becoming an art form in that it is a means of human expression, games are fundamentally software. Game Development is kind of an intersection of Software Development and various Arts combined into this interesting ball of creativity and ingenuity. However, with that said every single problem you laid out are not unique to games. They are common to all of Software Development. I cannot tell you how many projects have PMs pushing us developers to start making projects even before they've written the first requirement. But hey PMs and business control the money and these people never seem to understand game, technology, web development, software development or anything for that matter. Always seem to be people who are just doing busy work but never really understand the field and the people they work with. Very sad indeed.
I wonder how Nintendo manages their programmers. It works on so many projects at a time, so does it just shuffle its programmers to new projects after one is completed?
yeah i wonder this too... are they really working 12 hours for 7 days a week like in the other video???
man it must be hard....
A good bit of this is done by having several different studios for each game they have in production.
Japanese work mentality... work 15 + hours a day, party hard at night, power nap... repeat!
do it, do it right, do it fast!
Steve Jobs used to discard many projects and only work on a few (1-3) potential projects, which i think Nintendo should do.
Nintendo has many studios, I think that each studio rarely works on more than 1 game at the same the same time (and I suspect that when a studio works in two games, they have a very small team in a preproduction for the next game).
For example, Nintendo Studio 3 has released all the Zelda games since 2003, and all the games that Studio 3 does are Zelda, they took on average 2 years per game.
I like the idea of pre production teams. I also like the idea of smaller games peppered in between the gigantic AAA games. A smaller game would allow some of your Devs to still be busy while larger games are in pre production. and once the small game is out you could put that staff onto the growing needs of the larger game.
7:25 This did not age well.
PRE-PRO-DUC-TION!! PREPRODUCTION!! Put your planning skills to work...
Currently in pre-production with a fun and playable tabletop prototype. The hard part is having enough time to change your mind
Sadly, I can't hear this episode, all that is going on in my head is that damn "Reproduction" song from Grease 2, but with an additional P added to the refrain
Yo tengo una idea ,
Un universo gobernado por el sentimiento universal buscando la logica del infinito por medio del cambio de universos alternos hasta finalmente emprender el trabajo de un Dios,finalizando con la union de los 7 mil millones de dioses humanos que formaron su universo,destacando la culpabilidad,la logica del infinito tiene sus sentimientos pixelados,
La Era AstrometaETICA ha llegado.
THE GOD OF THE DREAM OFF!
Que os parecio mi historia.
:v
Mi jueguito tendra esa historia.
***** :D
I'm an aspiring developer and I DO looove talking about how my game is going to be xD I'm setting myself a deadline right now to be done an early access version by the end of this year at the very latest just because I don't know how intense college is going to be for me personally. is this too long? it is just me coding
TheSomeoneXD yeah that's kinda what I've been doing.
TheSomeoneXD u a dev too?
TheSomeoneXD I did for a while too. trust me you want an engine. I know unity in personal is free but I also know unreal engine is bigger on platforms other than mobile. the Cong isn't too much more different, it's just learning what the new functions do etc.
How'd that deadline work out, OP?
Here's a simple solution. If the problem with pre-production is that too much of the programming staff won't be busy during it, add more pre-production!
So if you have about 30 programmers on hand, and only 10 of them get an assignment, start two more projects to occupy them! Or find some contract work for the studio to do lol
As I said in the episode on buggy games, this is why I (usually) prefer the low-key and indie games to the AAA industry. Games should be an form of art and entertainment, not just an assembly-line carte blanche to print money; games like The World Ends With You and Cave Story have far more genuine passion put into their work then whatever they call the most recent Call of Duty. That, and I’m not that fond of many of the most common AAA genres, especially FPSs, and my restriction to the DS line, Wii, and iOS means I can’t play a lot of AAA games even if I wanted to.
What you do with programmers? Have them do everything that you need technically done that isn't a game. Web design, for instance. Bug-fix older projects that just got released (there are usually way more bugs right after release than far after release), So say you had 30 programmers. 10 are doing useful things for the current project, maybe 5 are working on web design. 10 are bug fixing the last project. 5 could be doing whatever else you need done with your technical systems.
would you tell an airplane mechanic to fix a submarine?
***** It may be slower work, but it's still work. Even if you have to make them work with actual web designers, it's better than break time.
TristanBomb the point he was trying to make is not all programming is the same; just because they can code doesn't mean they can do everything related to programming.
blaze595 yes, exactly!
blaze595 I'm aware of that. I've done some coding myself and my dad is the IT Architect at his work. He can work with webservices but if I asked him to draw something on-screen he couldn't do it. That said, it would only take him 10 minutes to find out because he knows the basics and knows exactly where to look.
Pre-production for a decade? Who ever heard of such a thing? Definitely not Starcraft 2.
Clearly the problem is having too many big wigs with different ideas, and a lack of clear communication of requirements, and then actually sticking to them. If you get a new idea, don’t you dare force it in at the last minute. Instead run a prototype and use it in a future game.
Food for thought: 2K has roughly 7 months between WrestleMania and the release deadline. So between October 31 and March 31, They could have their devs releasing DLC of new superstars and such to keep the player occupied until the next game can be put on the shelf. Not only would this occupy the programmers and character devs, this would also buy plenty of time to plan the next project. And by the time you're ready to mash all this stuff into a big ball, playtest and ship, You'll already have premade characters and everything because you really just needed to throw some extra sauce onto the last game and ditch some really bad mechanics that no one ever touched anyway. Most game studios can do this.
It's not unique to video games. The issue of pre-production planning is an issue in almost all areas of development. Exists in game development, software development, web development, you name it. More often then not business owners just want it to be done and don't care about the details. Hence the just go go go mentality.
It's the waste you see versus the waste you don't. If programmers sit on their asses because there's nothing for them to do, that's an overt, obvious waste. If they instead are busy doing work that later turns out to not be fit for purpose, that's a waste too, just not an overt one. In Lean business development, there's a saying: There is no worse waste than doing the wrong things efficiently.
Maybe you can hire someone to do the pre-production of the next game while you're finishing off the first one?
What about agile methodologies used for software development, do they apply for videogames?
The best recent example of this is project TITAN that was in the pre-production phase for 17 years after they finally gave up on it they used the remaining assets and some ideas to make OVERWATCH infact the game is soo much a comeback to life story that overwatch was supposed to be the task force in project titan
7:19 Being the next blizzard... That didn't age well...
Okay, I realize this is 8 months late, but this is exactly the kind of thing that cloud computing is meant to solve. Blizzard naturally doesn't want to invest in the level of infrastructure they would need to support those first two days, because it'll be wasted money after that initial peak. But if they host their services on the cloud, they could simply provision a huge amount of nodes from the cloud provider for those first two days, and stop paying for them when they're not needed anymore.
Was waiting for you to bring up Pixar but when you did it was not in the way I was expecting, so I'm just going to throw out something Walt Disney would have his teams do before they start a major project. Creating shorts.
This practice seems to have been revitalised in recent years by Pixar who always put their short before a movie and in games I've seen it done by Quantic Dream and Double Fine.
The point being that it allows teams some room to flex their creative muscles and practice their skills, without having to pour millions into a project. (Google "The Goddess of Spring")
As everyone knows Pixar put their shorts in front of movies to add to the cinema goers experience and they generally work in warming up an audience before a movie. They can also be a testing ground for new ideas to see how consumers respond. Recently Disney announced that they were going to make a feature length film based off the animation techniques of their short Paperman. Which won them an Academy Award but didn't make any money.
So, for games, take a project (anything) and give it to all those people not doing anything during the pre-production phase (a month?) and then release it. The studio could put it on the disc of the feature game as a cool extra or it could put it in some download store for a dollar. It's a much cooler way of collecting metrics and it leaves your team stronger when it's finished. Of course it could also make a profit but that's not even the priority. Just breaking even means that that first month of pre-production was totally free.
That's all I wanted to say but here are examples; Portal began as a mod; Spacebase DF9 (now in early access) was released as a prototype; Heavy Rain the tech demo existed before Heavy Rain the game; Geometry Wars was for a long time only available as a mini game on Project Gotham 2 because the studio didn't feel justified in releasing it as a full retail game; and Ape Escape 2 had a mini game called Monkey Football which I have invested more time in as a party game than anything the main game had to offer.
That's why many successful triple-A studio's nowadays make their core-team produce a stream of DLC for a year or two and then bundle all that extra content in a "platinum edition".
In the 90's, early 00's it used to be about producing new games as quickly as possible so your games always have the latest graphics. Nowadays graphics isn't evolving as quickly as it used to, so it's more about milking the product cycle as much as possible.
I don't work in the game industry but I am starting to work in the planning industry (ok that sounded better before I said it out loud but I'll explain). There is something out there called the "Project Management Institute" or PMI in North America. I know the UK has their own version but there is some serious overlap between the two. Basically it's designed for non-industry specific best practices for planning and monitoring a projects lifecycle. Many of the core problems here are addressed in the training, though I'm not convinced that the solutions provided directly copy paste to be solutions for the game industry but hey might be worth checking out. Watching the video I was already using the PMI terminology (example senior production team in planning would be the "Subject Matter Experts" or SMEs on a project) and fitting it into a scheduling chart in my head (yeah I need a hobby leave me alone). It won't be a 100% fix straight away but it should help alleviate some of the problems. After all making a game is really just a project.
All they lack is some project management. Almost every business faces the same issues. In the end the current project always becomes "hot" and you need an infinite amount of workers on it (I work in automotive, where any delay can cost hundrets of thousands of Euros). Nevertheless it is better to grab a few experienced lead-employees to work on the details of the next project. They don't have to do this 100% of the time. Eg. in my company the Project leader will start organising a new project while the old one is still running. He will grab a few workers for about 20% of their time to do some planning and create the first task-packages for the next project.
This has only minor effect on the current project, but greatly benefits the next one.
If you don't know how to organise this hire a management-trainer.
Yes the suggested method has merit, because it's the Mark Cerny Method, the one which got him all those awards. The pre-planning and the production teams moving from game to game, that's how Naughty Dog and Insomniac games does it. :D
Game development is much like a team-sport. However, injured players or those sitting it out are similar to those in the production team with nothing to do yet or are less busy. The only difference is that players sitting on the bench can replace those currently playing in case of a switch or someone else gets injured; a production team is a limited resource of participants that a limited budget can't afford to be wasting much needed time.
It is as if everyone on the team is injured, tired and not much time left in the game to insure a win.
"Pre production feels like you have all the time in the world..." (BioWare - Anthem 2019)
Oooh, subtle jab
Game design studios really should look into hiring supply and logistics management professionals to engineer their workflow. Those of us in the field have a firm grasp of how to handle organizing a complex system to flow without bottlenecks; it's pretty much our job description to do so. If you can nab one of us who also has experience with project management, even better. And if they've also tried their hand at game development (or at least have a familiarity with video games), you've got yourself a winning situation on your hands.
(Yes, I fit all three categories. I am shameless.)
how about you do pre production in the middle of the current project since your not close to shipping date you can miss a few people to do pre production for the next thing
Starting from Monsters Inc., Pixar started teasing their next movie in their current one.
So, basically pipelining?
You essentially have two or three games being worked on with offsets in production. One game in pre-production, one in the first year of production, and one in the second year of production, due for release at the end of the year.
Year 1, you have Team A working on the second year of Game X's 2-year production, Team B working on Game Y's first year of production (out of 2 years total), and the tiny Team C doing pre-production on Game Z.
Year 2, Game X has shipped, Team C merges with Team A to start work on Game Z. Team B moves into year 2 of production on Game Y. Team D is created to start pre-production on Game A.
I have been in preproduction for a project for almost 2 years because I am still learning how to make games, and only when I finish the course I am right now I will start the production fase. My project changed a lot alredy
What if you just had 2 teams? One small team solely focuses on pre-production work and the second team is the game developer team. That way no one from the game development team is being taken off of a project. Also, if you have a team that only does pre-production, they should eventually become efficient and complete pre-production faster. Once they've finished pre-production, they can start planning on another game, regardless if the first team is done with the current project.
What's tragic is that Pixar is now changing their pipeline. Now that they are under new management, instead of having 2 or 3 films in the pipeline, they now have 7-9. This is very unfortunate exactly because of what this video talks about: it cuts short time to develop the story. The movie Brave is a result of this. It wasn't a story of Pixar quality, even though it was visually impressive. It' sad to see this happening. It really is.
Each and every issue talked about here appears in ANY software project if the planning phase fails (or doesn't exist). But there is already quite a lot of books, seminars, authors and patterns regarding project management specific for the software industry... so why should game development be any different? Any Project Manager would tell you that without planning, your software is doomed... and another thing, if your company can afford teams of 30 programmers (not counting the rest of the staff) I think that they can hire a couple of PMs and "waste" 2 weeks of time planning their first project... and then have them manage multiple projects at a time. It's what big companies do, and works for teams with any number of members.
Naughty Dog spends about 6 months in preproduction, then almost a year in production. They juggle now. Working on two projects at a time, but I know the first time they started juggling projects was during Uncharted 3, they started working on The Last of US. I'm not sure how they did everything before then with 6 months of preproduction per game...
programmers can give feedback in the pre production phase. They are some of the smartest people in the industry. True their arts skills aren't there but they can give good feedback for game ideas and implementation. Its good to have everyone give feedback even HR lol. Every staff should be active at all times.
Solution: Have two preproduction teams. While one works on the current project, the other works on the next. This allows beyond necessary time for preproduction, and still having that high quality staff on the current game. This would of course cost more, but with significantly more time to plan, the odds of the project being well built and successful increase, meaning more profit.
What about just, you know, actually hiring more pre-production people, but also giving them some programmers who can do very early prototypes and such?
I used to be in a new indie team a couple years ago. All we did was talk on skype about 80% what the name of our team should be, 15% what the game should be like in the most vague sense (Steam punk sci-fi pixelated roguelike is what we eventually decided on.) and 5% one guy in our team talking about the lore and story of this world he's been building for years. What we never did is Pre-Production. So I kept telling them to make up some design documents and stop faffing around with generic ideas, and they kept telling me to get on to coding a game that I know nothing about. Next time, I'm starting my own team, and I'm going to be the lead designer, and a secondary Programmer. We'll plan out what we're doing before we start doing it. No wonder we never got much done.
Well if we go with the model of "two teams, one in pre-production and one in full production" and your problem is that a team might spend too long in pre-production if they have no deadlines, simply make the release of the game in full production the deadline for pre-production. that way, not only would the switch from one project to the next be faster and likely more easy to streamline, but it also removes the whole "no deadlines" dilemma. The pre-production team has a definitive deadline to get thins sorted out before they have to actually go into full production with the game. If a game's release date gets pushed out for whatever reason, then great! That means more time for the pre-production of the next game.
Though this also could be combined with the "let the unoccupied designers muck about" idea. What I'm implying is that you have a few people, not like high level people, just a few assorted guys on hand that either have lighter workloads than most full production staff, or they are simply another part of pre-production staff whose jobs are to experiment with the ideas proposed during pre-preoduction.
Say, for instance, that a particular mechanic or aesthetic were proposed for the new game, you could have someone readily on hand to say "let me see what we can do with that, and I'll get back to you." Not only does this save time during full production making sure certain mechanics work (because someone already has) but it also means that once that person comes back with the results, the planning guys likely have a new idea already lined up for what I think I'll call "pre-playtesting" since it allows for not only experimenting with how certain mechanics will end up working alone, but with each other, ensuring there are no weird control scheme hiccups based on what controls what, detecting bugs that might only occur when action X happens at the same time as action Y, which might not have even happened unless someone was actually testing two separate mechanics at once, which wouldn't have actually happened until waaaaaay later in production, and would have probably been much harder to fix by that point. They wouldn't even need to worry about game feel yet at that point, just how the mechanics interact with one another and how they might interact with the aesthetic being suggested.
WOW that got away from me... I didn't need to go THAT in-depth, did I?
This'll probably never be seen, anyway...
What about an entirely different solution? Instead of planning the ENTIRE game up front, why not develop the industry more around games that never stop growing and expanding. In that sense, a small arcade space shooter could grow to become something like EVE online. Developers could constantly ask themselves what the next best addition to their game would be, and remove one if it turns out it doesn't work quite as they though it would.
Pixar is the very very minority you know. These days 99% of the studio would hire, say, 20 concept artists during pre-production and fire them all to hire animators once they start production. Just an example.
Pre-Production time would benefit the gaming industry, but it will be hard to find a concept that works for them. Althoug, I must say, the Animation Studio Style of Pre-Production time might have the best chance :3
damn, no wonder we have been suddenly getting sequels to their films recently.
Why not require the pre-production guys to make rapid prototypes?
That way, they have goals to work towards in smaller intervals, and they can experiment with what ideas will or won't work after trying them out.
With this kind of performance, they could offer multiple possible game design ideas while actually making early progress on its design.
does not Nintendo do this with Zelda and Bethesda Assassins Creed? I have hird that while the guys who made Majoras Mask was working on it, the guys who made Ocarina of Time made Windwaker. I have also heard that there are three teams at any given time making the Assassins Creed series. Each team checks in on each other to see what new mechanics to add to the next game. Grappling hook. This could be a problem better solved for bigger teams. While smaller teams just have to force themselves to actually do pre-production and hopefully get into this cycle.