Ashley's Forum Posts

  • Prominent - it's probably not anything to do with C2, it's probably some other software on your PC.

  • BTW, Ashley, Is not it better to create a eclipse or Android studio template with crosswalk webview in it? So, a user like me just need to replace the folder "www" and compile the files into APK easily. Moreover, Android studio or even Eclipse can handle all the signing process too.

    Because we target primarily non-technical users. Asking non-technical people to "just" install the Android SDK doesn't make much sense IMO, so we tend to guide people to publishing services like the XDK that handle it for you. Of course if you know how to do a custom Cordova build yourself, go for it.

  • This thread is from 2012. We added support for loader layouts over three years ago

    https://www.scirra.com/tutorials/318/how-to-use-loader-layouts-to-make-custom-loading-screens

  • fldr - we can't test every device in existence. The problem is always with the graphics driver so it's not our fault and is almost impossible to do anything about - check our extensive blogging on the subject through our blog archives for more information. We have proven that this type of error is the fault of the graphics driver time and time and time again, they really make crap drivers.

  • You can just delete the manifest="offline.appcache" attribute from the HTML tag in the exported HTML file. But I'm not sure why you'd want to disable offline support?

  • The latest beta update has a fix for a bug which caused choppy playback in Chrome. We should have a stable update out soon anyway that will include the fix.

    There isn't one audio or video format that plays in all browsers, which is probably the main compatibility problem of HTML5. Apple and IE/Edge only play AAC audio and H.264 videos, but all other browsers can play Ogg Vorbis/Theora & VP8.

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads

  • Any service that uses the Android webview should work fine on Android 5.0+, since the webview supports WebGL, Web Audio and auto-updates through Google Play for the latest features. However for Android 4.0-4.4 you really need Crosswalk to get a decent browser engine to run the game - the webview sucks real bad on those Android versions.

  • I'm thinking of doing it later this year, since it would be interesting to include Edge on Windows 10 Mobile (which hasn't released yet), as well as Cordova on iOS 9 (which should be able to default to WKWebView). That should get close-to-60FPS all round I think, even on devices as old as the iPhone 4S.

  • You can't. It's a limitation in iOS Safari.

  • iframes on iOS are weird. Add the attributes scrolling="no" noresize="noresize" and they generally work better.

  • No, for that error, the mistake is definitely in their plugin. As it says, their plugin defines two actions with the same ID, which is obviously not allowed.

  • — - running from disk like that has never been supported - if it says file:/// in the address bar then it will never work, because browser security around local files prevents the game loading. You need to visit chrome://extensions and use the "Load unpacked extension" option. This is all documented by Google.

  • Can't reproduce and nobody else is reporting the issue, so it must be a problem specific to your system. Sorry, not sure what we could do about this. Maybe check your graphics drivers are up to date? The "File - New" or "Open project" steps are the moment the editor first creates an OpenGL view, which is where most buggy graphics drivers then throw an error.

  • Thanks, should be fixed in the next build.

    Note there's no reason at all to use a C:\ path - if you just use "Test_1.webm" without a path it loads it from the project file.

  • Thanks, good catch. Should be fixed in the next build. Note for me it did loop the correct number of times; it is just that the Chrome dev tools log would collapse identical logs in to one line with a number next to it. The issue was loopindex("a") was returning the least nested loop's index (event 2) rather than the most nested (event 4). From the next build, when identical loop names are used it returns the most nested first, which fixes it.