Aphrodite's Forum Posts

  • It seems there is not an expression for the overall velocity, well, pythagore still works:

    overall velocity = (velocityX^2 + velocityY^2)^0.5

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • Thanks for the quick replies!

    I don't think I'll have as much as 100, but even if it won't make much difference, I'd rather stay safe and optimize it.

    Thanks again!

    the point was "the optimisation at this scale is probably a waste of time, as something else would probably eat more ressources", I said 100 but it would not surprise me if even at 1000 it would not be a big deal compared to other things that might happen in a game.

  • Android is an easier target for monetising As far as I saw (you cannot expect people to pay for a "meh.." game on PC, and advertising on PC is a strict no go), and advertising makes it easier (and can be a little too easily added to Cordova or canvas+ based games).

    Making an android game with C2 is harder than making a web game with it (ref. all the post of people complaining about C2 exports to android), but with a web game you have the "oh no what if it got stolen?" / "I cannot monetise it (not in a easy way at least)" / "I do not want to rely on browsers butching my game" / "I do not want it to be played on both PC and tablets/phones".

    The PC community is more demanding than people that play mobile games (most people playing on the phone just want to have some occasionnal fun, a PC gamer however can be easily an hardcore gamer wanting a full fledged game), however browser based games runs fine on most modern devices and PCs and people playing them on either generally wants a good time (I never saw someone complaining about a browser game being too long before someone says that outdated argument).

    But web based gaming still has this reputation of "bad games" or "short time games" or "non serious games", even though this has been proven wrong a lot of time (after all, all of C2's games that do not rely on native features can be played inside a browser, sometimes it is even a better experience inside said browser, and even with the native features issues it is still doable to have a browser fallback coded in most cases.)

    So I can see why mobile rather than PC, but why NEVER web, I cannot see (Only reason I can think of is Money, but surely that would not be that only)..

    PS: do not tell me about the ones doing html5 games to sell them to html5 publishers, all I saw was not video game creation but money absorbtion via template like games.

  • A behavior sine with a magnitude of 0 still takes up calcul time, to be exact, even disabled behavior take up calcul time.

    however, that will matter only when a lot of objets are concerned ("a lot" depending on the target platform, but I would say under 100 you should not see much difference, but test it.)

  • My guess would be that it (the for each element) has been designed to be used to to change values inside the array not based on the values of other cells, still a valid request IMO.

  • — I would like to run smooth game on dual core CPUs (like 14.000 in Antutu) however my last phone is broken, so I can test only on new one (26.000). It's going smooth but my friend has some older tablet, and he can see jittering. I'm using CocoonJS (I tried one time Chromium from CocoonJS, but it just gave me white screen).

    or maybe I should ask Ashley - what exactly means "collisions"? Does 1 object generate 1 collision check?

    As far as I can tell, a collision test between object A and object B will create A.count*B.Count collision checks, minus the ignored instances via collisions cells or filtering.Behaviors benefits too of the collisions cells which will really help.

    a collision check in C2 terms is not yet checking the collisions polygons overlapping, but the bounding boxes (aka the smallest non rotated rectangle that fits the object inside), if those are overllaping, the polycheck (aka collision polygon check) will be tested, this one is far slower so keep an eye on it.

    Some behavior combinations (solid, platform, jump thru, LoS, bullet with the bounce off solid) will increase the number of collision checks, but keep in mind that they can benefit from the collision cells, which means they will ignore too far off objects (under some circumstances that you must follow, aka not using different parralax rates for collisions).

    while it is true that keeping the number of collisions checks low is nice, testing on the device you are targetting (or ask someone to do so) will be decisive, as there is never a "universally recognized" number of collisions checks to not go over, each device being different.

  • You do not have permission to view this post

  • I tried on my tablet, I had the same as ramones, without webGL hiwever, I had only white instead of the brown object.

    did you intalled the directx websetup after exporting? (even though there is not much hope for that I would say)

  • Time to lerp!

    unlerp(340, 480, Sprite.Y) will return 0 if sprite.y is 340,1 if sprite.y is 480, and interpolates with a affine function between those values (0.5 will be for sprite.y of 410)

    lerp(0.8, 1, A) will return 0.8 for A =0, 1 for A =1, and interpolate with an affine function between.

    combined with lerp, you can do:

    lerp(0.8, 1,unlerp(340, 480, Sprite.Y)), which should return 0.8 for sprite.y being 340, 1 for sprite.y being 480, 0.9 for sprite.y being 410.

    hope I am correct

  • you are reffering to "v.1.190", where did you see that version number?

    last time I checked, C2 only had either versions like "r190" or "Version 19000" for references, nothing like v.1.190., so it is normal to wonder (unless the installer mentionned it?).

  • export using the "cordova" (or "android", depending on the release you are using) should remove that warning.

  • the app added size should always be the same (like, if it adds 13MB on a little game, it will always add 13 MB when the game become larger), due to the fact it ships with it an entire browser engine to read the game.

  • alexhui without talking about breaking changes (as some might say it is a breaking change, which would basically negates entirely the demand) with no reason, an invisible object just means it is not rendered. As for the manual, it does not state either that non rendered object do not respond to touch and mouse inputs.

    however, having a "respond to inputs" kind of state for objects in C2, could simply do that trick, As low priority as it can be, I am more into the "respond to inputs" state addition rather than modifying what "invisible" means really.

  • Yes.

    Now i have code like so:

    on start layout => Activate Group "Group1"

    condition ^ once => Deactivate Group "Group1"

    So, a group of events/actions included in Group 1, are only active from the beginning of the layout until a condition holds.

    I think it would be nice if an activate or deactivate condition could be entered in the group itself. Right now, the only condition available is active when layout starts.

    Dan

    IIRC, you can also include an event sheet under a condition, which means it will be read only when the condition is met, not sure however how that works with triggers. That may help you organise things a little.

  • well, they could, but in most cases the capx posted have issues in them or are incomplete (I personally recommand to post a description of the issue, and the capx with altered graphics, or another capx with the same issue so there is no stealing risks), however to make profit out of it, they would need a C2 license and working on your codebase to complete the game, which takes sometimes more time than doing the game themselves.

    however be wary if your project is something you do not want to share the entire source.