Ashley's Forum Posts

  • I'm not sure exactly what you mean, but opening a new browser window should work fine.

  • I'm not sure, I think it would be best to ask Microsoft. Construct 2 doesn't specify any particular version AFAIK.

  • All Construct 2 games have always worked offline: https://www.scirra.com/tutorials/70/offline-games-in-construct-2

  • I can't really debug anything from screenshots. A minimal .capx is always the most helpful thing to provide.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • You can, but it will stay in Free edition mode because the licensing systems are different.

  • So we really want to drop WinJS if possible - it means shipping a few megabytes of extra resources with the C2 installer and this overhead gets bundled with every Win10 app, which increases file size, memory use and startup time. It also creates a bit of a distribution headache because technically we should keep our bundled WinJS updated with the official version, which we hadn't bothered to do previously (so there is the risk of bugs/unsupported versions as well).

    I think pubCenter/Microsoft Ad SDK is the last thing relying on WinJS. I didn't find this while testing the changes in r228 since WinJS is not directly used by the pubCenter plugin, it's an indirect dependency. I've got in touch with Microsoft and they say the Microsoft Ad SDK is going to publish an update in the near future which drops the WinJS dependency, which should then fix the pubCenter plugin in C2 as it is. I think I would rather wait for this to happen rather than re-bundle WinJS, which adds overhead to every game. The Ad SDK update should happen before the next stable update (or if it's close we will delay the stable release for it).

    There are two workarounds for the time being:

    1) roll back to r227 stable. Note we have always advised that beta releases are just for testing, so you should not upgrade important projects without making backups and having a rollback plan in case there is a problem with the beta release. We cannot promise that problems only occuring on beta releases will be fixed any time sooner than by the next stable release (if indeed we decide it needs fixing). The hack-your-project-backwards trick only works if you don't use any features that were added in the intervening releases, which it looks like you did do, so it won't backport until you delete those changes. It looks like you started using the new "Set back button visible" action which was new in r228, which is why it won't backport to r227.

    2) add WinJS back yourself. Start a new blank WinJS project in Visual Studio, then copy-paste the WinJS library stuff to your existing project including the referencing tags in index.html. That is basically what C2 does when it bundles WinJS, so adding it back yourself will get the Ad SDK working again.

    I'll leave this report open as a reminder to others (and to me to test it works again before the next stable release after the MS Ad SDK updates).

  • The original report is over a year old, and it seems unlikely that the original report is relevant to any issues that are happening now, especially if they only occurred recently. Please file a new report following all the guidelines if you still have trouble.

  • This thread has become quite long and confused - I'm not really clear on whether the report is about the Chrome Web Store, or Google Play. As of recent builds I made some changes and collaborated with Kyatric who was testing the publishing side, and it was a bit fiddly to set up, but worked. Kyatric's tutorial covered the setup to help others get it configured the right way. So as far as I am aware, it is working and documented. Given these changes since this post was originally made, and the lack of focus in this thread, I am closing it and recommend to file a new report (again following all the guidelines as closely as possible) if you are still having trouble. I would also recommend making one report per store, so there is a clear focus in the bug report, e.g. one report for any Google Play issues, one for any Chrome Web Store issues, etc.

  • .capx link is 404, so closing (as per bug report guidelines this is required)

  • Thanks for the report, this should be fixed in the next beta. Note however it appears to work the same as Safari, where audio is muted until the first touch.

  • Thanks, should be fixed in the next beta. Good catch.

  • Closing as not a bug. I think this is just what R0J0hound described: if the behavior is disabled it does nothing so the runtime doesn't even redraw the game, so everything looks faster because it's not actually doing anything at all (a variation of the "my game goes slower when a sprite starts moving, why is moving a sprite so slow" type of report we've had in the past).

    I guess it's a minor problem a non-moving enabled platform behavior causes redraws, but I don't think it's a serious problem. The platform behavior is also very complex and has to do a lot of stuff like checking for moving platforms, slopes, gravity direction etc. so some of that could be triggering the redraw.

    Still, there is no evidence this is a breakage, it looks like C2 has always worked like this. As somebody actually proved there have been no significant changes to the platform behavior. So there is nothing new here, it's just reporting the way things have always been.

    If you have trouble with v-sync accuracy, that is an entirely different issue and depends on the browser code. You should probably file a bug at crbug.com with your system spec and v-sync stats to get that investigated, probably referencing crbug.com/422000 (the main vsync bug).

  • Thanks, should be fixed in the next build.

  • Closing as not a bug. This is a common misunderstanding of OR blocks.

    The OR block runs because the system expression is true. No Tilemap instances matched the "Tile (1, 0) = 0" condition, so there are zero instances picked. Testing a condition with zero instances picked cannot find that any are true, so the event does not run.

    If it worked like you wanted, it would mean running actions on instances which did *not* meet the event conditions, which entirely contradicts the way events work.

  • Can't reproduce on a 2nd-gen Moto X running Android 6. Seems to work just fine.

    Android's sound playback has been a bit flakey in the past, but that's mainly because the OS audio stack has been problematic, which isn't something we can fix ourselves.