oosyrag's Forum Posts

  • I do not know how to do it, and I haven't seen any guides on how to either. Never said it was impossible, just I expect it highly unlikely that anyone here would walk you through it. I don't see any reason why an indie developer wouldn't be able to if they were motivated enough to. Many games with this feature are made by indie developers.

    Costly not really from a dollar sense, as there are many free resources online that document WebRTC and JavaScript, but I would guess very costly in terms of time if you were to research and learn how to do it. Or costly if you were to hire someone to develop it for you then yes.

  • My understanding was that the loader layout was for downloading assets only, such as when loading for the first time off a web page for an embedded html5 export. An exported phone app should not require any downloading at all, as all the assets should already be on the local device.

    The black screen delay is likely loading assets into graphics memory. This can be mitigated by using a more powerful device, reducing the memory load per layout (or first layout at least), or by loading assets during runtime via load from URL (not recommended unless you are comfortable micromanaging your loads). Or all of the above.

    For reference, just in case you haven't seen it yet: scirra.com/tutorials/318/how-to-use-loader-layouts-to-make-custom-loading-screens

  • Use a Timer behavior on your player object. The [X] is an inverted condition - so the events only run if the "Cooldown" timer is NOT running. 

    + Mouse: On Left button Clicked
+ Player: [X] Is Timer "Cooldown" running
-> Player: Spawn Bullet on layer 0 (image point 0)
-> Player: Start Timer "Cooldown" for 0.3 (Once)

+ Mouse: Left button is down
+ Player: [X] Is Timer "Cooldown" running
+ System: Every 1.0 seconds
-> Player: Spawn Bullet on layer 0 (image point 0)

  • Has anyone ever made a local multiplayer LAN game with construct 2 ?

    By local multiplayer i mean like "mini militia" where two or more players in a room can connect to one player's hotspot connection without internet and play together locally.

    I'm going to venture out a no here.

    I searched about it im forums and found that each topic about this ended with uncertainity.... So i just want to clear this up for myself and future readers of this post who might me wondering too, whether this is even possible or not with construct 2?

    Technically it should be possible. AFAIK it comes down to integrating a local WebRTC signalling server into your app, which is probably beyond the scope of these forums.

  • What isn't exact - the position, or the time?

    Frame updates do not necessarily happen in sync with seconds.

    The position should be accurate, unless you have other events or behaviors affecting the object besides what you mentioned.

    + System: Every 1 seconds

    -> Sprite: Set X to Sprite.X+90

  • construct3.ideas.aha.io/ideas/C3-I-107

    construct3.ideas.aha.io/ideas/C3-I-15

    Basically there are existing third party tools you can use to check in/out your project file that should work fine as long as you're not making changes simultaneously. Source control for a single file should be very straightforward.

    You can try to do it with your project saved as a folder, but the chances of things breaking catastrophically are much higher, considering how many features in Construct reference each other.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • The linked post should still be accurate. Apply the effect to each layer you need it on. See if you can combine or reduce unnecessary layers if you have a lot.

  • Functions will not make your project file size noticeably smaller except possibly in extreme cases. They can save space in terms of number of events on your event sheet. If your 6 events are replaced with a function that also has 6 events, you will have the same number of events. However, if you use those 6 same events in 10 different places on your event sheet, you would only need to call the function once for each.

    Note that additional actions do not constitute additional events - you can have as many actions per event as you want. The main benefit of functions is when making changes, you can make the change once and have it applied to all instances where the function is called, rather than having to change it in every instance separately.

    As far as efficiency and system resources, functions have negligible impact in most cases.

    To make your project smaller in terms of file size, the first thing to look at and reduce would be the number and resolution of your sprite textures and animations.

  • Possibly loading all your sprites into memory for the first layout, if your first layout is heavy (not just the loading screen). Try loading into a blank layout, and then switching to your actual first layout on trigger and see if it makes a difference.

  • This won't work with the built in platform behavior. You can either use physics or set up your own platforming events.

    You could also possibly use a helper solid object. Upon collision, move the upper block up a pixel, and place a temporary invisible helper object of the same width below it, so that the fallen block is resting on this helper object rather than the moving box. When the moving box no longer overlaps the helper object, delete the helper object and the upper box should fall down. Haven't tested this myself though.

  • From my past experience with pong, I found "perfect" bounces happened because of the limited resolution of the collision box (octagonal). You can tweak it a bit manually rotating the collision box points slightly so that they are not perfectly in line with any 90 degree angle.

    I had meant to try making an ball out of several separate objects to increase the resolution of the circle but never got around to trying it.

  • You probably wouldn't notice any performance difference either way. There will definitely be an organizational difference for your workflow though.

    First thing that comes to mind is that if you decide to use a single object only and later you find you want to remove some frames from each animation to improve performance (or lengthen certain ones), it would be a nightmare to find and re-reference all the correct frames in your event sheet.

  • CSS pseudo selectors are not a thing in Construct.

    There are workarounds though. You can use a substitute sprite font or text object to display what gets typed in your text input box, which can be set to invisible. The first character in sprite font or text objects can then be accessed with the .left(1) expression and replaced with a suitable BBcode tag.

  • Construct really isn't the greatest for text heavy applications, but it should work fine. From my experience with text based applications, the key is to not update every text object every frame (say, to display an array) and only change them on demand as necessary. If you do that, filling the screen with text objects should be no problem at all.

  • You can check for collisions between picked objects only, so yes. This could take some work depending on your application though, because the solid behavior and the behaviors that interact with the solid behavior (if you're using them) do not support filtering groups of objects.