Does it make a sense to use a build server in small (2-3 developer) game development team? I like the idea of continuous integration, and the team members are not in the same place, so it might not be a bad idea.

  • 2
    \$\begingroup\$ I think the question is more: why not? What's stopping you? \$\endgroup\$
    – knight666
    Jun 18, 2012 at 10:08
  • 1
    \$\begingroup\$ No, I'm pretty sure you don't need CI for game development at that scale. If you are looking for a real answer, add more details because your question is way too open-ended. What problems are you running into now that you think might be solved by CI? (Team members not being in the same place isn't one of them since that should be solved by an online issue tracker, version control and Skype/IM/IRC.) \$\endgroup\$
    – Eric
    Jun 18, 2012 at 10:48
  • \$\begingroup\$ Well I would not say that we could not progress well without it. Occasionally work stops because untested code gets committed into the repository but not too often. As I mentioned I just like the idea, and I was curious if a build server can make development somewhat easier besides wrong code gets revealed sooner. \$\endgroup\$
    – nosferat
    Jun 18, 2012 at 11:30
  • \$\begingroup\$ Is it normal for developers to write code without running it? Maybe I'm misunderstanding what Continuous Integration is, but it seems like the point is to make sure the code runs, in which case I can't imagine not constantly running my code while I work on it. \$\endgroup\$
    – jhocking
    Jun 18, 2012 at 13:03

2 Answers 2


The main benefits of CI are:

  • Constantly checking if the code is broken
  • Making an "official" build which ships to the customer.
  • Running the tests after you make a build and passing the build if all tests pass

I just completed a small game in a one-month cycle that shipped. We didn't use CI; even if we did, we would not have needed those above two benefits because:

  • Our code-base was interpreted (Ruby) code
  • We coordinated and integrated often
  • We didn't have automated tests.

Your situation may vary. If you don't need any or all of these benefits, it's not worth the additional overhead to:

  • Setup the CI machine
  • Maintain it when things go bad
  • Figure out why things work locally but not on the CI machine.
  • \$\begingroup\$ I would argue that 'figuring out why things work locally but not on the CI machine' is not overhead but valuable information, a common and nasty bug category is 'works for me, and just for me', CI sometimes helps find these bugs. Great answer anyway :) \$\endgroup\$
    – Roy T.
    Jun 18, 2012 at 21:27
  • \$\begingroup\$ @RoyT. I couldn't agree more. Usually it's overkill though in the context of gaming. Depends on your game though :) \$\endgroup\$
    – ashes999
    Jun 18, 2012 at 22:18

It makes sense to use whatever tool gives you more benefit than it costs to use.

In the small team I am working in, one person suggested that we have a CI server, but upon discussing it, it seemed that it would take a significant degree of effort to get it working with the tools we use. In our situation, we couldn't spare that amount of time so we chose not to use one.

The best answer for you will depend on how quickly you can set one up, and how much benefit you think you'd get from it. Personally I think most of these systems are only worthwhile for big projects with large numbers of developers, but you may think differently, based on your own experiences and on how much benefit you expect to see compared to the difficulty of setting it up and maintaining it.


Not the answer you're looking for? Browse other questions tagged .