thegrind's Recent Forum Activity

  • I have had the same problem. Reached the same conclusion as you.

  • Ashley Would be nice and useful to be able to activate and deactivate most behaviours.

  • This is very basic stuff - it is just that I'm experiencing some weird results. Take a look at this:

    http://i.imgur.com/yJMJset.jpg

    If I have 2 visible sprite instances, I'd expect variable 'count' to go up by 2 per tick - right?

  • Say I have X instances of a sprite. What is the proper way to select circled object:

    http://i.imgur.com/El1PI8v.jpg

  • thegrind Sorry! I forgot to backup when I moved to different host, pmed you a code for reward, thanks for reporting.

    No problem, and thanks. Didn't get any code though.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads
  • Demo does not work.

  • That's a smooth little clicker. Gratz on the completion!

  • Problem Description

    Maybe this is a generic browser issue and cannot be fixed. 'On save complete' triggers even when a game is not saved, because of insufficient disk space. This is unfortunate and can cause great player wrath, because they will still get 'Game saved' notifications.

    I remember experiencing it myself once, but can't remember if it was in Chrome or Firefox. I've had 2 players reporting it, who both solved the problem, by freeing up diskspace.

  • Hi guys!

    Finally finished our free browser based incremental game (for desktop/laptop):

    http://idlekiller.com (with full hd gfx, audio, and story).

    Got a lot of good feedback (and traffic) from http://www.reddit.com/r/incremental_games/ today.

    A big thanks to all the nice people here, who's been very helpful answering questions And thanks to Scirra for a great engine (I'd really like if I could just get one tick/sec when tab is inactive).

    I'd be grateful for any help sharing the game.

    A couple of screens:

    Cheers!

  • Hi guys, would like to hear your thoughts about some trigger condition technicalities.

    In above example 2 trigger conditions are evaluated on 1 user action (mouse click). It should be a pretty typical example. If Sprite2 is over Sprite, we dont want to register a click on Sprite. But contrary to std. event sheet execution, objects are destroyed immediately, and the sub event in line 4 becomes true. So, the easy way to resolve the issue in this example, would be to just move line 3 below line 4.

    However in projects of considerable size with 100s to 1000s of events, it seems messy and difficult to keep track on the sequence in which trigger conditions can occur. I wonder why trigger conditions caused by the same trigger, is not handled like events in a tick? For instance handling object destructions after trigger condition stack. Maybe I'm missing some obvious reasons for handling triggers this way. Any thoughts or insight?

    I thought about this for a rather huge project. The cleanest solution I could come up with, was a wait for signal before actions in problematic triggered events. And then, end every tick with a signal. But it still bugs me

  • Bumbing, since a flood of spam just hit after I posted

  • If my understanding is correct, event sheets are run top down. If objects are destroyed, it will not be effectuated before the end of a tick. This means that you can't rely on stuff like object.count within the same tick. This is not a problem (one might argue though, that it would be more logical if the object.count variable was updated instantly, but of course you can always update your own variable).

    What seems odd to me, is that the destruction of objects in triggered events, seem to happen instantly, even if the user action cause triggers to fire multiple at places in the event sheets, resulting in a stack of triggered events.

    Below link is a capx with a very simple example. Description: If green sprite is clicked, it triggers a mouse click event, and +1 is added to a variable. Press 'b' to spawn a blue sprite at mouse position. If blue sprite is clicked it is destroyed. If blue sprite is clicked on top of green sprite, the green trigger event should't count a click, so a check is done to see if mouse is over green sprite. You switch the order of the 2 event sheet in the include, and it doesn't work.

    https://drive.google.com/folderview?id=0B6WdU53eEV7Jflkya2t6UWhCQ2NNWldlLU1Va3RUWWowNUhMLTNlSk9Tc2Fyd2FvM3F0dmc&usp=sharing

    I think synchronous triggered events should be handled like an event sheet more or less, meaning all the triggered events should be handled, and stuff like object destruction should happen afterwards, but before proceeding with other events. It is very difficult to order triggers in huge projects with many event sheets.

    I'd welcome any deeper insight, since I'm no expert in the underlying engine.

    Cheers,

    thegrind

thegrind's avatar

thegrind

Member since 22 Sep, 2013

None one is following thegrind yet!

Connect with thegrind

Trophy Case

  • 11-Year Club
  • Email Verified

Progress

12/44
How to earn trophies