Colludium's Forum Posts

  • Yes it is is the quick answer. I'm sure it's somewhere in the account settings but its been a while since I adjusted mine - I'll have a look around and see if my memory gets jogged. What you can't do is change your merchant account country or the keystore you use to update any apps you have already published (you could use multiple key stores but you then run the risk of confusion). If you want to change the login gmail account then you have to setup a second merchant account and then email their help from both stating that you want to swap, they will then refund one of the $25 fees. That's detailed in the play merchant FAQ.

  • Yeah, I know. It's late and I could have put more time to it - I'll have a look to tomorrow. The custom movements are great but if you want more or a different effect I don't think they play nicely together.. �

  • It appears that there is some sort of 8-direction key press buffer that is active even when the behavior is turned off. If I just tap the arrow up key very briefly so that the sprite moves into the gravity zone then it works ok, but if I long-press the up arrow key when the sprite is already above the line then the duration of that key press seems to have some impact on whether or not the sprite will misbehave.

    Not sure what the answer to the problem might be, though. If you just want to simulate falling due to gravity then using custom movement or taking more control of the default 8-direction behavior might be a better option (I've found that mixing movement behaviors like this is often more trouble than it's worth).

    Edit - link deleted: the solution is posted below...

  • Nice idea and I'd be up for it but I'm 'stuck' in Florida.... It's a bit of a drive!

  • Thank you for this KYATRIC. It's effective and simple enough even for me to understand!

  • — As far as i understand, the Height and Width values of a sprite don't exist in system memory if the sprite.count = 0. You can't access the Width and Height values until an object has been created, and C2 doesn't remember all of the past values of an object's parameters unless it's IID 0. Here's what I mean:

    capx

    I've always taken the need to create objects in a dump layout just to be a quirk of C2. I have made the mistake in the past of deleting instance zero of an object from my dump layout only to then have the 'default' spawn settings subsequently go crazy - size and angle - with no easy fix because I couldn't insert a new object with a lower IID than zero...

  • Here's a quick demo capx for you!

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • I don't think that setting the physics iterations to 0 will have the desired outcome.

    There may be a better way to do this but I would put my physics objects into one family that has the physics behavior to keep things simple. I would also give the family 3 variables - let's call them X_vel, Y_vel and rot_vel.

    Then, on the pseudo stop physics command I would do the following (aside from gravity = 0):

    for Each Family:

         Family.X_vel = Family.Physics.VelocityX

         Family.Y_vel = Family.Physics.VelocityY

         Family.rot_vel = Family.Physics.AngularVelocity

         Family : set velocity = 0

         Family : set angular velocity = 0

    I believe that things could still bump into each other if you move them while they're paused like this, but if the scene is static then this should work (works for a Pause function as well). Then, on resume physics (gravity = 10): for each Family set angular velocity to rot_vel and the x and y components of the linear velocity to X_vel and Y_vel.

  • I think it's because when you run the group after a delay there are no instances of Sprite available for the create instance expression to reference when setting the spacing using Sprite.Width and Sprite.Height. It works on layout start because there's an instance that can be referenced before it's destroyed at the end of the start layout tick. Or something like that.... <img src="smileys/smiley4.gif" border="0" align="middle" />

  • AnD4D,

    It was a bit of a long shot - I'm sorry it doesn't work. I presume you're using cocoonjs - is asm.js is out of the question? I haven't tried it for a couple of months because of the teething problems but the use of a different and better optimized engine might work...

  • It's taken me some time to type and things have already moved on, but here goes what I think might be happening. So, as I read the events - and apologies if I've misunderstood - the following could happen:

    • People overlaps HumanMove and People is on same layer as HumanMove

         - People var Path = "Top" - do some stuff including put var Path = "Middle"

         - If Path = "Middle" then do some other stuff (it always will be true because of the previous event.

              Next events: half of the time tempY will be set to 465 which will then put People on layer "PathTop", which is the layer we were in at the start (and from the debug 465 is the People y value now).

    As there's a trigger once in the first event, at the next tick the People parameters will not have changed so the event won't fire again. The only thing I'm not sure about is what the relevance of the MoveTo target position is - so all of this could be me describing what you already want the code to do!

    If you're looking for the human to move in a +y direction to avoid the obstacle, could it be something like the crate's origin is off so that y=465 doesn't look like the human is avoiding it?

  • Understood - thanks ASHLEY.

  • capx

    Steps to reproduce:

    1. Run in browser, cocoonjs zip export or crosswalk apk

    2. Touch one coloured block with one finger and hold

    3. Touch the second coloured block with another finger

    4. On each touch an event moves a touch indicator to the touch co-ordinates

    Observed result:

    On the second touch the second touch indicator moves to the co-ordinates of the first touch so that both touch indicators become superimposed.

    Expected result:

    The second touch indicator should be positioned to the co-ordinates of the second touch.

    Browsers affected:

    Chrome (android): yes (I don't have a multi-touch laptop)

    Firefox (android): yes

    Internet Explorer: not tested

    Operating system & service pack:

    W7 x64 SP1

    Construct 2 version:

    r158.2 x64

  • Perlin, can I suggest that you to edit the OP title to something like "Possible multi-touch bug, please verify" to get some of the old and bold in on this?

    Thanks!

  • Perlin,

    OK, I've done some testing and I'm not sure whether or not this is a bug in C2.

    I even went as far as creating a crosswalk app using a modified version of the capx I linked to above to isolate the cocoonjs wrapper (if you download it again you'll see that I've tried to isolate which touch event is registered by holding 2 touch IDs in variables).

    Sometimes C2 is not registering a touch's end trigger - this is indicated by one of the touch variables appearing stuck at the last setting until a new touch is initiated to replace it (my events might be untidy and causing it, but I think it's ok). This seems to happen after a touch has been stationary for short period of time - a few seconds. Coupled to that, there does appear to be some cross-over when a new touch starts on the black square and a touch event on a colored square has been stationary. Here the co-ordinates of the colored square touch event seem to get passed to the wrong touch event.

    Just as you initially reported it, but I'm at a loss to explain it.

    Best submit it as a bug report, I think, and let the team take a look for sure. I can do it if you want, but you found it.... <img src="smileys/smiley16.gif" border="0" align="middle" />

    Ed