Zebbi's Forum Posts

  • > rexrainbow would this be useful for recreating Jill of the Jungle's knife? https://youtu.be/t7E-P4VJbdg?t=45

    >

    > Sometimes, her knife can't find it's way back to her and falls to the ground: https://youtu.be/t7E-P4VJbdg?t=23 but this wouldn't have that problem really, would it?

    >

    It might be interesting to try but you would be better off using the faster/simpler position check that game uses. I think the weapon getting stuck is a feature, but you can get around it it by making the weapon warp through solid obstacles or not even colliding with them at all during the return phase.

    Is there a way of having a proper boomerang system? I'm not keen on it getting stuck, I'd rather rely on the mechanics of one usage per return of the weapon rather than worrying about it getting lost. I'm not sure how to lerp it away from me and return with the behavior, though.

  • rexrainbow would this be useful for recreating Jill of the Jungle's knife? https://youtu.be/t7E-P4VJbdg?t=45

    Sometimes, her knife can't find it's way back to her and falls to the ground: https://youtu.be/t7E-P4VJbdg?t=23 but this wouldn't have that problem really, would it?

  • Ashley only the ones I listed are disabled, all others are working and fine; is there something else I can try?

  • > Multiple Raster Threads: Disabled

    > Native GpuMemoryBuffers: Software Only. Hardware acceleration disabled.

    > Video encode: Software Only. Hardware acceleration disabled.

    > WebGL2: Unavailable.

    >

    The key entries are "Canvas" and "WebGL" (ie. WebGL 1), are they not listed?

    Everything else is listed as working and enabled, including canvas and webgl.

  • > I really hope that scirras own service to build apps in C3 is alot better than phonegap.

    > Just tested phonegap with a tiny test game and the performance is embarrassing.

    >

    Again, I will keep repeating this, you have to check for GPU blacklisting as the very first step. Visit chrome://gpu in Chrome on the device to check. This only affects about ~5% of devices.

    I have:

    Multiple Raster Threads: Disabled

    Native GpuMemoryBuffers: Software Only. Hardware acceleration disabled.

    Video encode: Software Only. Hardware acceleration disabled.

    WebGL2: Unavailable.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • >

    > > I'm pretty sure C3 is going to be the best thing since the discovery of fire.

    > >

    > Burnt fingers.

    >

    That's why children shouldn't play with matches.

    Takes one to know one? Sorry, I thought we were coming up with childish retorts as opposed to humorous wit.

    How about this: speaking of the invention of fire, newt is certainly taking a Neanderthalic attitude in his many trolling barbs of late.

    Best just not to feed them.

  • Is there a way of having magicam "catch up" so the player is centred on screen, no matter how much lag is given? I think I may have asked this before, but I can't be sure. I did actually add some acceleration to it if the player walks for a length of time, but it's kinda fussy.

  • Pulstar seems to work perfik! Bit of a mess, mind:

  • I'm pretty sure C3 is going to be the best thing since the discovery of fire.

    Burnt fingers.

  • Zebbi

    You should be able to use a different image per tiledbackground instance.

    Sorry, yeah, it works for every instance of the same tile perfectly, just need to get an example made.

  • In your situation I figure spawn might be more convenient.

    I would remove the containers, otherwise you will end up with creating too many detectors and badguys. Bah, I am not sure if this would not create a feedback loop if you forget to remove the containers. Badguy spawns detector via event, detector spawns rest of container on event creating new badguy, badguy spawns new detector via event etc.

    Yeah, you're right, I think it would create an ungodly loop if I left them in.

  • >

    > Yeah, I read that was the case. So, should I give each badguy a variable, put the detectors in each badguy's container, assign a variable to all the detectors and badguys, and pick each of them and give them the nearest UID?

    >

    > Or should I just duplicate code?

    >

    Do something like below.

    On family "badguy" created

    For each "badguy"

    create/spawn detector

    set variable "detector.parent" to "badguy.UID"

    Then if the detector colides with whatever it is to collide with do:

    On detector collides with whatever

    for each detector

    pick badguy by uid [detector.parent]

    do events for badguy

    Should work IIRC. Let me know if it does not, I will be using a similar method to link inventories to NPCs and Loot Containers in my current project.

    Also family.uid=object.uid, you can also use that to check if a family member is a particular object type IIRC.

    Gonna try this, should I still use containers at all, or just remove the two detectors from the container and spawn them? Also, is that definitely spawn and not create (two different processes)?

    Also, how about turning each character left or right? Is that just a case of using foreach?

  • Is there some other way of being able to reference families whilst maintaining individual instance profiling? I'd have thought duplicate event groups would have been really really discouraged, but how else can I use the events I'm using?

  • blackhornet Magistross There's also this containerhelper behavior that allows for picking family children by UID, but I'm too stupid to understand how it could be useful:

  • Families should do what you want. Can you make a smaller example, so we can see what you might be doing wrong?

    Sorry, I used a large example to cover all the possibilities.

    Containers do not work with families. However, you can manually link objects together using a common variable value for instance (like linking the detectors with the sprite UID). You will need to explicitely pick objects though.

    Yeah, I read that was the case. So, should I give each badguy a variable, put the detectors in each badguy's container, assign a variable to all the detectors and badguys, and pick each of them and give them the nearest UID?

    Or should I just duplicate code?