Slashdot Mirror

← Back to Stories (view on slashdot.org)

Economics of a 2D Adventure

Posted by Zonk on from the put-the-hampster-in-the-microwave dept.

Thanks to The Grumpy Gamer (Ron Gilbert of Monkey Island fame), for his excellent look at The Economics of 2D Adventure Games. "First, this is only a thought experiment. This is not something I am planning on doing, or even have a huge interest in doing, so please don't feed the rumor mills. Second, this article contains gory and gruesome details about the games business and, in particular, marketing and distribution. If you'd rather remain blissfully oblivious to the horrors of what goes on behind the scenes, this is the place to stop reading. If you're one of those people that can't help but stare at a car accident, read on."

10 of 29 comments (clear)

  1. The first thing I noticed by loopback_127001 · · Score: 4, Interesting

    In the budget was this: Testing Subtotal: $30,000 In case anyone wants to ever wonder why the game industry releases consistently buggy, shitty product, there's your answer. Testing. ALL the testing for 1 year. $30k. Half the salary of a single developer. I would assume that's either for hiring a single real person at 30k to do the testing, or else to pay 30k to a contractor service for some amount of hours of testing. I think it's interesting, especially given that the stated numbers are from his experience in the industry. Perhaps in this specific case, with a known engine and no cutting-hedge technology, spending such a tiny amount of testing of the game could work out, if you also have the developers and everyone else playing the game / testing along the way as well. Overall, a really interesting breakdown of how things would work for even an extremely simple game idea, and how much money is involved.

    1. Re:The first thing I noticed by AltaMannen · · Score: 4, Interesting

      I don't think it was assumed testing is full time for the full time of the project.

      Until you have something "playable" the programmers and designers can do all the testing and that is going to take a certain amount of time.

      Also, going with the publisher approach may mean that they have an additional testing team that goes under "publisher overhead". Most titles today have way higher testing costs than $30k, but this is a low-budget project.

    2. Re:The first thing I noticed by loopback_127001 · · Score: 2, Interesting

      I both 'agree' and 'don't agree'.

      First, the agreeing! I think you're 100% on point about the publisher test team, since publisher's have their 'acceptance tests' that a game must pass before it can be declared gold, and their test team is a part of that overhead.

      The idea of test coming into the product to test something that's already playable is very likely a common approach (I don't work in the game sector, but it's a common attitude in other places in the software industry, so I feel pretty ok making that assumption). But as anyone who has had to sit through a presentation on test knows, bugs found at the beginning of the design/implementation phase are a lot less costly to fix than ones at the tail end of it all. Having test involved in design and planning meetings at the early stages can make a huge difference in the final quality of a product.

      With a 2d point/clickita adventure game specifically, I don't think test would necessarily provide much in the way of useful feedback, but let's consider that for a moment. If your game is going to be an extremely simple, single-solution / single-path adventure game, that 30k number is quite reasonable.

      But if that's the case, what do you need all 4 programmers for? you'll really only need the Lead Programmer (who can tweak engine issues) and probably 1 junion dev, and perhaps an intern or some other nearly-free work. Heck, have a junior-junior dev who is also your tester, and save money in BOTH areas.

      Now if your game is going to have multiple solutions to puzzles, multiple paths, 'ideal' endings, and all the other things that are often looked for in modern games, test's job (and development's) becomes much more complex, including reviewing the game's relative difficulty, if following unexpected branches break forward progression in the game, etc. And in that case, 30k is not going to be enough, and everyone would benefit from having Test involved at a very early stage to provide feedback on the design and planning.

      It's also worth noting that "testing" is just considered a monobloc expense, and not an actual person/position, which goes to the heart of the problem in the games industry. "testing" isn't a job, it's a commodity you buy. You hire programmers to make your product, but you pay for the testing of your product.

    3. Re:The first thing I noticed by cgenman · · Score: 2, Interesting

      Bugs at the beginning of the design / implementation phase are generally the responsibility of the designer / programmer. We're not generally talking about the more sophisticated test casing and automated test script generation of the larger software industry, we're talking about people in their early twenties working for 10 dollars an hour or less. This is not to say that's how it should be: I've been pushing for more automated test solutions from programmers at the company where I work, and it has been yeilding tremendous results. But the fact of the matter is, you don't need playtesters until you have a playable. Up until then extreme coding practices and careful design implementations are what you have to rely upon.

      It's also pretty safe to assume that if you hire competent people a remote tester is going to get 30% or less of the bugs of an in-house tester, and will be less able to oversee their correction. In case you didn't know, publishers will have a team of QA on their premisis roughly equal in size to the developer's QA team, devoted to each individual game. Perhaps it's the difficulty communicating with the dev team, or the generally 400% turnover rate, but remote publisher QA is terrible. I'm baffled as to why this practice persists, and why we haven't just moved to a publisher pays for in-house model (with, of course, requisite spot-checks). It's not like the publisher has a team devoted to checking all of the artwork in a project all day long... Take the butcher's word for it, and put those QA developers where they belong.

      Testing is starting to get the attention it deserves as a discipline. The company I work at is going to a one designer - one artist - one tester model for level development, which should help communication immensely.

      3 testers for the final 4 months of a project this size at 30k is reasonable.

      But if that's the case, what do you need all 4 programmers for? you'll really only need the Lead Programmer (who can tweak engine issues) and probably 1 junion dev, and perhaps an intern or some other nearly-free work. Heck, have a junior-junior dev who is also your tester, and save money in BOTH areas.

      I noticed this too, but remember that programming includes scripting, which in a game without level designers means that some of those "programmers" are essentially on world layout. You also have object generation, input, and placement, tools development, effects generation, scripting cut scenes, etc. etc. In most other places a designer would do much of the jobs listed above. It might behoove them to add a team member to the schedule for that very purpose.

      As a side note, be careful with your interns: They may be making 10 dollars an hour, but once you hit crunch they may be putting in 80 hour weeks... or the equivalent of a 52k per year salary. Better to pay them a reasonable amount up front.

      --
      The ______ Agenda
  2. other scenarios by JackBuckley · · Score: 5, Interesting

    This is an interesting article, but the situation is quite hypothetical--the odds of someone trying this development model for a 2D point-and-click graphical adventure are slim. For real fans of this genre (who also might want to make a buck) it might be more interesting to relax some of the assumptions about part-time or hobbyist programming staff and the infeasibility of internet distribution (and/or even a shareware model). Developer/publishers like Spiderweb Software (makers of the Exile series, among other Ultima-type RPGs) help with the back-end stuff for small-time or hobbyist shareware developers. Similarly, sites like Home of the Underdogs promote and help sell "scratchware" games. The real question to me is, can a scratchware/shareware/late night after work development team make enough money to break even (given the real and opportunity costs of creating the game)?

  3. I think he should rethink the "PC Game only" part by Thag · · Score: 5, Interesting

    In particular, the Game Boy Advance is the #2 console, and would be both cost-effective to develop for and well-suited to a 2D adventure format.

    You could easily do Monkey Island on it.

    You're going to have to go through Nintendo, but you'll have to go through a distributor on the PC side too.

    Jon Acheson

    --
    All opinions expressed herein are my own, and not those of my employers, who are appalled.
  4. Re:I think he should rethink the "PC Game only" pa by cgenman · · Score: 2, Interesting

    I'm sure you could get Nintendo on-board for an innovative, broadly approachable 2D adventure game for the Nintendo DS. I'm sure at this point they're desperate for games that could be ready by launch next year, and might be willing to take the larger risk to have an expanded launch lineup, especially with a proven team.

    --
    The ______ Agenda
  5. Interesting but probably not that useful by mistersooreams · · Score: 2, Interesting

    This must be a pretty encouraging article if you are a small company or single person about to embark on developing your first 2D adventure game, but honestly, how many people is that? The article is interesting for its insight on budgets (other people have already mentioned the tiny testing budget) and other nastiness in the games industry, for sure. But most Slashdotters interested making their own adventure game would be well advised to check out something like Adventure Game Studio, which has a friendly interface and very handy scripting language. Sorry if this seems like an offtopic plug for AGS, but really, there's a lot of good adventure game creation tools out there (not just AGS), and they're probably more useful in practical terms than this article.

    That said, it's nice to hear that Ron Gilbert still has adventure games on his mind...

    --
    apterous.org
  6. Why so expensive? by creaturespeaker · · Score: 2, Interesting

    You could reduce costs by a huge amount if you used something like Multimedia Fusion: http://www.clickteam.com/English/multimedia_fusion .htm You don't have to do nearly as much testing because programs made in Multimedia Fusion always work on a wide variety of computers. Heck, you could probably even use Adventure Game Studio to make a commercial 2D adventure game (although you would be more limited then you would be using Multimedia Fusion)! Try thinking outside the box a little! =)

  7. Re:I think he should rethink the "PC Game only" pa by antin · · Score: 3, Interesting

    Not only that, but the DS with its touch-screen would make adventure games easily playable. Playing on a d-pad or even with a control-stick would get tedious as you sent your cursor all around the screen looking for hot-spots. The touch-screen would be preferable even to a mouse for such a game.

    And as most adventure games seem to make use of large inventories, the second screen would even get a useful purpose.