Kojima ponders 'pilot episodes' for next-gen games

Game development costs will increase even further for the next generation of consoles, or will amongst developers who want to use all the extra horsepower to push polygons aplenty and tremendous textures. Given that so-called "AAA development" is already pretty dang expensive and risky, this may prove troublesome. Metal Gear creator Hideo Kojima has recently proposed one solution: releasing "pilot episodes" of games to test the waters before plunging fully into development.

11

Game development costs will likely increase even further for the next generation of consoles. Given that so-called "AAA development" is already pretty dang expensive and risky, this may prove troublesome. Metal Gear creator Hideo Kojima has recently proposed one solution: releasing "pilot episodes" of games to test the waters before plunging fully into development.

"It's possible to make many things more realistic, but that doesn't mean you should," Kojima told Edge. "You have to prioritize, and that is what's going to separate the teams that succeed from the teams that don't. A very deep, 20-30 hour game might need a bigger team and take three or four years."

"I think there's a different way of tackling this problem: something similar to a TV series, where you can use pilot episodes to test the waters before you jump completely into the project," Kojima offered. "It can be distributed via download channels, so the player can try it out before production continues. Something like that wouldn't take that long to create, maybe a year, and if it's successful, you can continue."

A year of development is still a huge commitment, mind. And creating a slice of a game is no mean feat if you want a "pilot episode" with the level of polish we see in television pilots: you'd need release-quality art; UI; script; sound; and levels of not-being-all-buggy. All the same, it could stop companies from dumping too much money into projects which aren't working out, then shoving them out the door in an attempt to recoup any of the costs. With next-gen budgets, that could go from painful to disastrous.

An even cheaper alternative is to ship lighter, more prototype-y games, with placeholder assets and their core systems in reasonable shape to test the ideas. Goodness knows too many people already struggle to understand that an unfinished game is not finished, though. And if a game is story-driven or open-world--like Kojima's own upcoming Metal Gear Solid: Ground Zeroes--it may struggle to shine in prototype form.

Double Fine Productions recently experimented with pitching 23 games to the public, letting people vote on which prototypes they'd like to see made, and paying in return for getting to play them. This lets the studio discover what people want in the first place, then get feedback on how they turn out.

Kojima can see next-gen consoles encouraging "a social aspect" to development, he told Edge. "You'll get user feedback, and I think there'll be this back-and-forth between users and creators."

Anyway. Complicated times in the industry, huh? Companies funding video games will need to be a lot more careful with their money, not going hog-wild chasing 'realism.' Boy, those video games!

From The Chatty
  • reply
    March 12, 2013 12:45 PM

    Alice O'Connor posted a new article, Kojima ponders 'pilot episodes' for next-gen games.

    Game development costs will increase even further for the next generation of consoles, or will amongst developers who want to use all the extra horsepower to push polygons aplenty and tremendous textures. Given that so-called "AAA development" is already pretty dang expensive and risky, this may prove troublesome. Metal Gear creator Hideo Kojima has recently proposed one solution: releasing "pilot episodes" of games to test the waters before plunging fully into development.

    • reply
      March 12, 2013 1:01 PM

      Hmm. It's like they're looking for a new way to share these digital wares

      • reply
        March 12, 2013 2:43 PM

        I wonder what they'd call these "shared wares"?

    • reply
      March 12, 2013 1:13 PM

      The gap between the pilot and the full game would be too long for people to keep their excitement, i'd imagine. I am STILL waiting for SIN: Episodes Episode 2

    • reply
      March 12, 2013 1:20 PM

      PS3 / 360 were the era of AAA games. We need A inflation. I want AAAA games at a minimum for ps4/xbox 470

    • reply
      March 12, 2013 2:18 PM

      As a gamer, I absolutely love this idea. I honestly don't have time for multiple 40 hour game campaigns in a given year, so giving us a chance to get just a taste of a new franchise in a bite-sized installment is a winning prospect. Really, this isn't all that different from the "episodic gaming" experiment from a few years back (that didn't pan out for anyone except Telltale), but I hope more developers keep this approach in mind.

    • reply
      March 12, 2013 2:31 PM

      I ponder, is Kojima still relevant.

    • reply
      March 12, 2013 2:42 PM

      "It might only take a year." Versus, what, 2 years for a typical full-blown AAA title? And there will still be a tremendous amount of content generation needed to build up all the reusable assets (textures, models, etc.). I'm having trouble seeing a whole lot of savings.

      • reply
        March 12, 2013 2:46 PM

        Don't AAA titles take more than 2 years?

        • reply
          March 12, 2013 2:54 PM

          Depends on the game. I think the average is around two years. Longer is arguably worse, since you have to actually finish most of the game to release this "pilot", but you have to do it in 1 year instead of 2 or 3 or whatever years. See patd's post.

        • reply
          March 12, 2013 3:18 PM

          AAA is generally 2.5+ years @ 10MM+ budget. The definition varies and AAA is applied sometimes to even mid-level budget small team games in order to try to market it as premium.

    • reply
      March 12, 2013 2:48 PM

      I don't think this is as viable in action as it sounds on paper for "AAA" titles.

      To finish a chapter, lets say 1/4 of your whole game, you need to create and complete every system in the game. Your rendering, AI, weapons code, controls, cinematics pipeline, rigs, animation sets, hud, menus, etc. etc. all have to completed, tested, submitted for console verification, and be at final shippable quality.

      At that point you are at least 75% done with the total amount of work that goes into the game (even more if you launch with fully functional multiplayer). You are probably 3/4 through your budget, and looking at only a 25% ROI for an episodic release. If that episode tanks. a big publisher might kill the remaining episodes to save that final 25% of the budget, and we are left with a wasteland of unfinished games.

      I'd see it is a viable option for an indy/download titles, where a few hours of total gameplay is perfectly acceptable beginning to end experience. The team could continue to create additional chapters over time, continuing to provide them as long as there is a demand, but it's a much tougher prospect to apply this method to a full dev studio.

      • reply
        March 12, 2013 3:57 PM

        Isn't it better to have an episode tank at 75% of budget and 1 year of work, instead of a full game tank at 100% of budget and 2+ years of work?

        • reply
          March 12, 2013 5:37 PM

          It's not quite that simple.

          Let's assume a 3 year game cycle and throw guesses at the division of dev time.

          To go for the whole game you spend an initial 6 months in prepro defining systems and writing/concepting design and visuals. Follow with 12 months coding your systems and building core game content. Then another 12 months for additional content and polish, then off to bug test and submission to console manufacturers (6 months). You hit your 3 years with a single title, have 1 test and submission process, and 1 marketing and retail campaign.

          To do episodes (lets go 4 episodes) you spend the same initial 18 months up front for concept/systems/core content, then 3-6 months for polish, then off to bug testing and submission (3 months for less content). At that point you are already 2+ years in to a 3 year cycle. Following episodes will take 6-8 months, and you will need to go through the testing and submission process for each episode, and work on marketing and retail space for each episode. This becomes a 4+ year cycle for an initial 3 year cycle game, with the additional cost of multiple submissions and marketing campaigns, and the additional development cost for the extra year, making the game MORE expensive to create in episodes.

    • reply
      March 12, 2013 5:52 PM

      i know the japanese industry is out of touch, but is Kojima-san so out of touch that he isnt aware of, you know, all the huge failures of Episodic gaming over the past decade? it sure as fuck isnt going to get easier to make it work, if dev costs are going to increase even more.

      • reply
        March 12, 2013 7:39 PM

        Isn't the Walking Dead a success?

      • reply
        March 13, 2013 6:00 AM

        I can only think of three examples off the top of my head:

        Valve - not willing to work on it
        Ritual - company died because they decided not to lay people off when their non-Episode 1 projects were cancelled
        Telltale - has been successful

        Two of these three aren't failures of the model, they're failures of the company to produced product. I wouldn't be willing to condemn the model based on that. Are there any other examples that I'm forgetting?

    • reply
      March 12, 2013 5:57 PM

      Feel free to scale it back afaic, I got ridiculous value out of Minecraft and FTL, and hell I played Dwarf Fortress longer than I've played many AAA titles that I actually paid for.

    • reply
      March 12, 2013 7:46 PM

      I can't imagine this happening. However I wouldn't be surprised if more big publishers started releasing smaller cheaper downloadable games to test out the reception to certain gameplay ideas and mechanics, and then running with it in a full price game if it gains popularity.

    • reply
      March 12, 2013 8:20 PM

      Pilot Episodes of Policenauts.

    • reply
      March 12, 2013 8:24 PM

      Kickstarter for console games...

    • reply
      March 13, 2013 4:38 AM

      i can see them switching to a two team system to keep development time down.

      either running it as a team produces 1 'pilot' then hands it over to core team who have been working on the system side and would full take over project on the hand over.
      then pilot team begins their new game

      pilot team hand over to a newly hired team having developed the core systems themselves and saving costs without having more staff working in back ground and move on to a new game.

      both ways would have over lap times with pilot team finishing quicker then core team which could be used to bolster work force and pick up the slack since they already know the development

      worst case is a game gaining massive sales later on after the studio has dropped then project or laid off staff

    • reply
      March 13, 2013 6:07 AM

      That's just another name for episodic games. And most devs figured out the upfront costs are too high for that.

    • reply
      March 13, 2013 6:13 AM

      Wasn't this tried already as episodic content?

    • reply
      March 13, 2013 6:44 AM

      "Game development costs will increase even further for the next generation of consoles"

      why? we've already reached a graphics level that's acceptable and has been so for years. sure push the envelope however as many games have already proven you dont need crysis level graphics to make a game a hit. I'd much rather they nailed the gameplay and worry about graphics afterwards. almost all of my purchases for the last few years have been indie or non AAA games even though I have a pretty solid gamign machine. I dont need every game to have all the bells and whistles

    • reply
      March 13, 2013 11:49 AM

      Seems like this could be really terrible, or at least really weird.

      Imagine your favourite game is a pilot. You want more. You and everybody else votes with your dollars. The pilot leads to the episodes. The episodes are more polished, look better, play better, have better music... etc.

      Don't you think a fan is going to look at this and say "damn, sure wish the pilot had all this stuff"? Or worse, "why do I have to keep paying for this same game?" Which brings up the question, is this pilot going to cost us less, or just be less costly to the developer?

      Sounds like a good idea on the surface, but way less a good idea the more thought I think into it.

Hello, Meet Lola