I'm creating a multiplayer game just for fun and exploring different techniques like isometric games, node.js and socket.io. The point is that I've tried to make an alternative approach on multiplayer gaming in a browser. Have a look here; http://ny.scsterallure.nl/ The second player connects to the game by going on his mobile device to http://ny.scsterallure.nl/index.mobile.html All though I'm not finished with the game and controls, I think the mobile device controller shouldn't be a webpage but an app. Since the webpage doesn't handle double taps and other events well. So my question is; are there already some generic (virtual) mobile gamepad apps or is this something new? Also what are your opinions on the whole controller interaction.

  • 2
    \$\begingroup\$ This question sounds very much like advertising without much research about the question. Asking "which technology to use" and asking for opinions are both off topic for the site. It sounds like you want to start a discussion. Take a look at the FAQ and see where to ask discussion oriented questions. You can also search the site for similar questions: gamedev.stackexchange.com/questions/10387 gamedev.stackexchange.com/questions/5384 gamedev.stackexchange.com/questions/32104 \$\endgroup\$
    – House
    Dec 29, 2012 at 14:29
  • \$\begingroup\$ It's not advertising and I did some research on mobile browser based controllers. Still the links are nice to read but are code examples. Still I'm more concerned about the question if a webpage based controller can be as effective as an native app based controller. \$\endgroup\$ Dec 29, 2012 at 21:08
  • \$\begingroup\$ This really does sound like you're trying to start a discussion, not to answer a specific question. This conversation would be more appropriate in chat, or on a discussion forum, rather than as a question to be eternally archived. \$\endgroup\$ Dec 30, 2012 at 5:49
  • \$\begingroup\$ Well I just wanted to know if a web-based controller could be feasible. I see your point about being more like a discussion with a subjective answer. Still I think there's only one good option and a lot of could be good. \$\endgroup\$ Dec 30, 2012 at 21:39

1 Answer 1


Such a thing already exists -- see Google's ChannelConnect API. I built a game for a client using this technology to do exactly what you're suggesting. I'd post a link, but unfortunately it was a 1-day banner ad game on Youtube. It was hooked up for iOS and Android, using a browser-based controller.

And TBH, I don't think it's the world's greatest idea, mainly for reasons of latency. As far as I -- and I think most networked game developers -- are concerned, a game isn't playable when the delay between your manipulating a controller and seeing the effect of that manipulation onscreen is more than a tenth of a second, let alone half a second. So the mobile network had better be damned reliable -- which you cannot guarantee with your players on so many different providers, at different times of day.

  • \$\begingroup\$ Thanks, that looks indeed like what I'm trying to achieve. Still my main question is that the mobile device controller should be a native app and not a browser based controller. How was your experience with a webpage as controller? Some negative points I experience (but feel free to refute them) 1) on android the webpage cannot be full screen 2) on android htc webbrowser zooming and scaling cannot be stopped 3) for fast input a webpage isn't suitable. \$\endgroup\$ Dec 29, 2012 at 21:05
  • \$\begingroup\$ @CasperBroeren We had problems with browser-based control. Fullscreen was one of the issues, I can't recall how we got around that but there was a way, I think. I'd definitely avoid the browser headache altogether and try to write a native app for control, if you're serious. Even better, use a decent sockets-based solution for your comms, because latency will be an issue. I'm not sure if there are alternatives to ChannelConnect for mobile, but I would start with this list (kudos to Lil Angel). \$\endgroup\$
    – Engineer
    Dec 29, 2012 at 21:12
  • \$\begingroup\$ @CasperBroeren Whether alternative networking solutions will help, I don't know. All I do know is that -- via the grapevine -- I heard that ChannelConnect is sockets-based. If you find out that it is TCP instead, say by contacting the guys who work on that at Google, then I would suggest avoiding it as TCP can be notoriously laggy for real-time applications. Otherwise, perhaps even sockets are bound to be slow on mobile networks. Though I'm certainly not convinced about that. \$\endgroup\$
    – Engineer
    Dec 29, 2012 at 21:16

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