Aphrodite's Forum Posts

  • Maybe the "force own texture" layer parameter set to yes

  • I think you are thinking of a virtual joystick, to do this, you will need two sprites, and to get the distance and angle between the two (using the distance and angle expressions) so you can use that to control the player

  • Based on timers, and so it is just an indication rather than a precise mesurement, also, the browser by itself may take a part of the cpu used

  • Kyatric : for the next time, if there is a lot of commercial plugins that pops out, could they be in a particular subpart of the "what you have missed"? So it is easier to track them (If you think it is not worth the effort, obviously it is okay, you already take a lot of time writing these posts, which are really helpful btw)

  • Increasing the gravity will make the object fall more quickly (it s vertical downward acceleration is bigger), when it fall down like you think it should, change the jump strength, so the jump height is good, the max speed, acceleration and deceleration will help you controlling the horizontal aspect of the character, a little acceleration will feel heavy and hard to move, a little deceleration will make the object slide before stopping when no input is acting on him, and the max speed is the maximum horizontal speed.

  • Aphrodite

    Great solution!

    I had used canvas to draw the fan shaped. It is the easiest solution, but the every tick drawing in cocoonJS iOS would drop the performance (the same as changing context of text object).

    My method requires a layer to force it's own texture, and uses blends mode, while I know PC will have no problem at all, I never tried on iOS itself, if you need a little help understanding something I had done, go aheahd and ask ^-^ (I am beginning to love blend modes)

  • Just using the platform behavior, jumping seems to come and go. Ill some times have to press the jump button 3 or times before the player jumps while running... this is a big issue for me as the game is a platformer so jumping needs to be precise. Any words of advice on why this isnt working?

    https://www.scirra.com/manual/100/platform

    Implementing reliable platform movements

    For the most reliable platform movement, it is recommended to use a invisible rectangle sprite with no animations with the Platform movement. Then, the animated player object can be positioned on top of that. Otherwise, the changing size and collision polygon of the object as its animation plays can interfere with the Platform movement's floor, wall and slope detection, causing a shaking or glitchy movement. For more information and a complete guide, see the tutorial How to make a Platform game.

    If you didn't do it, I suggest applying the platform behavior on a rectangle, then pin the graphics onto it (if you have multiple instances to work with, containers will be of help)

  • The persist behaviour can I think be of help (it maintains the exact state of the object after leaving the layout).

    Or you create it at runtime if a variable is equal to something, thus letting you not creating it if you already killed him.

  • A license is bought once, and will work after updating C2 (if for some reason you lose the license file, you can still re download it from the scirra store, just keep in mind of the limitations of a personnal license)

    More infos here :

    https://www.scirra.com/store/construct-2

    https://www.scirra.com/tutorials/57/how ... enses-work

    In case this is not enough informations, "Feel free to email licensehelp@scirra.com if you have any more questions! We'll do our best to reply as soon as we can!"

  • The savestating is huge, and while it works, I always find it too overkill, you are saving te entire game, in some cases, a simpler saving does the job really well and can also help not having the issues of saving unwanted things (sure there is the nosave behavior, but still)

  • To think of it, html5 itself isn t responsible, the node webkit export is (I still don t see why C2 even tried to promote wrappers, really the html5 export inside a browser just works in most cases without any issue compared to them, but the wrapper just tends to have more issues, just for distribution's sake, not even talking about the future when the browsers will interpret the game in a better way that the wrapper that would have stayed to it's current version), just saying that something should be done about the wrappers, we cannot just throw them in the bin (even though that is what I would have done, they serve no other purpose than a psychological executable)

    Edit: as for the high end graphic card, I can say for sure it is more false than true, even though the graphic card should be recent, my crappy laptop actually runs well done Html5 games fine.

  • You can add the browser object to your game, then use the request fullscreen action.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • Both should be the same really, so do what suits you the most.

  • This explains pretty much why abd how to worry about this, and I think it is the first time that I actually saw the fact that a 512x512 = 2x256x2x256 = 4 × 256 x 256 calculation is implied, I hope people will understand after that than designing a game with oversized assets is non sense, and that compressed image is not an image by itself, and so must be fully uncompressed.

    Also it is basically a rule of thumb: the less memory you take, and the less calculations you do, the better your game will run, and the difference between good design and bad design is that a better design will do the same job with less, while a very baddly designed game will run in "theory", but will fail in practice when everything comes into play (this is also a reason why people should try to learn those things, I don t think they need to know everything about hiw it works in practice, but at least what is a better design, and for exemple inspire themselves of old games, which were far more limited).

  • If you want however an indicator not starting from 0 (for exemple, if your bar min lenght would be 10)

    You could use unlerp(minlenght, maxlenght, curlenght)*100 to get the percentage