Hello,

I thought this issue was related to the Canvas plugin, but when I remove everything but the canvas it wasn't happening anymore, so trial and error to find out it might also be related to the platform behavior. But, I'm posting it here because the Canvas plugin is the part that isn't native (and seems like it may be the issue).

When a canvas and a sprite(with the platform behavior) are in the layout there is a notable decrease in FPS. Even if neither is doing anything. But ONLY when webgl is "on". Even though nothing is using webgl as far as I can tell (maybe I missed something).

Here is a capx example:

http://dl.dropbox.com/u/659906/canvas_test.capx

Sorry for the complicated button presses to show FPS. But I wanted to make sure the text was not part of the issue.

In this capx, you press "enter" to save the fps to a variable, then hold "space" to show it on screen. When you release space the text will be destroyed.

Enter- save fps to variable

Space (Hold)- display saved variable on screen.

Space (Release)- destroy text on screen

1- destroy sprite with "platform" behavior

2- create sprite with "platform" behavior

0- destroy canvas

9- create canvas

If you destroy the sprite or the canvas, the FPS picks up to 60. If you add more than one sprite, there is no change in FPS. If you add more than one canvas there is additional drop of FPS.

also, if you turn off webgl there is no issue with fps.

I would appreciate a second set of eyes here, as well as any suggestions on what I may be doing wrong.

Additionally, I'd like to say this is an issue that came about with the newest C2 release, but I've yet to uninstall/reinstall/test.

Hello,

Question:

Are you looking for a programmer to work within the current engines for both those platforms? Or are you looking for someone who can customize the engines?

A thought:

C2 and Unity platforms have very little in common, as far as porting goes. From everything I've learned you'd have to basically re-make the game on either end, from a programming standpoint. I'm sure you could reuse all the media, but programming would have to be re-done. Though depending on the complication of the game, this may not really be that big of an issue.

As someone else asked though: Why not just make it in Unity?

And the reason I ask this is because it seems you are wanting to target more platforms than just Web or Mobile. So if you want to hit up platforms like Wii or Playstation or X-Box (the least expensive for indie development being X-Box), you'd have an additional set of programming for pretty much no reason, other than the desire to work with C2.

And although I love C2, it may not fit your long term goals. Or at the very least create an unwanted development time to get to all your desired platforms.

I am having the same issue. In one event I will create the "one shot" particle emitter and the immediate next action is to change the rate.

I am using [r112], which looks like it would be the next version from the OP, though I'm pretty sure I've always had this problem.

Raymond - are you still having the same issue?

It could just be me ;)

Oh and don't forget to post it in the bug section

I guess that's what I'm trying to say, I don't think it's a bug.

Thank you for your input though, It's always nice to get a fresh thought process on an issue.

:)

I had another thought. I understand now that "every x seconds" is a system event, so it has to be separate from the object/instance events (or that seems to be the idea anyways)

Question:

Could we get an additional event added to object (sprite) events list?

-Every X Seconds

The idea being is the same as the system events, but for the objects instead. As well as, having it be instanced (otherwise they wouldn't be different from the system version).

I guess I've noticed that in every game I've made I've had at least one (but usually several) objects that I use in an "instanced" fashion, and giving them small randomness to their actions seems to bring a little life to the project. Not that it's hard to make a timer (now that I know how), but it'd be easier and more straight forward with an object based event option.

Just thinking out loud...

You can work with it for now ...

I'm sure the devs will find a solution soon enough

I hope I'm not coming off as though I'm looking for someone to "fix" this, I'm only trying to figure out if my experience is abnormal, which it seems to be.

I'm also hoping that if there are others with similar/same issues, they'll be able to learn from what I've been through.

Speaking of custom button mapping....

I don't have a stripped down capx, sorry, but I'll try to explain what I did.

I made a button that shows up when I hit the esc key. This button can then be clicked to start the "button mapping" process.

An on screen prompt will tell you what to do.

Push (A) button,

Push (B) button,

Push (X) button,

and so on...

Using the "any button pressed" gamepad event and the "last button" expression I record the button that is pressed and store it's code in a variable for each button.

variable: buttonA

variable: buttonB

variable: buttonX

and so on...

this way you can always use the variable for the button you want and it will use the corresponding "mapped" button code.

and to go one further with it...

I then have it go through the actions to map buttons to.

"Please press the button to map for Jump"

"Please press the button to map for Run"

"Please press the button to map for Shoot"

and so on...

I had to build in a check for each action mapped, so that if you hit a button that has already been mapped to an action it will clear the previous button.

So the first part maps my buttons correctly, and the second part maps the actions to the buttons I want.

again, to go further...

I also set up other buttons on a menu that allow me to see what button is mapped for what action, as well as clickable buttons in the menu that let me map each action individually.

So each clickable button just calls the one specific event related to mapping that action.

This was all incredibly time consuming, but it's really helped me through development (cause I can change what my buttons do instantly), and I think it will be best in the long run for the end user/player, so you know that everyone will be able to play it (as long as their button presses are recognized by chrome anyways)

in the future...

Eventually I will flesh it out to be more polished. I'd like to include it as part of the options menu, and then also separate the mapping from the assigning of actions to buttons.

If I can figure some way to strip down a copy of my capx without having to redo the whole thing I'll post it here for reference. I don't have all the buttons in variables, etc, but enough to get the idea.

I used the newest driver on both computers. My OS (win7 x64) installed built in drivers when i plugged it in the first time on both computers, but I then went through and installed the newest software package (includes drivers).

I only seem to have problems with the button being off on the one computer, but I built a custom button mapping process into my game so it works no matter what (once I map the buttons of course).

Support Page:

microsoft.com/hardware/en-us/p/xbox-360-wireless-controller-for-windows/jr9-00011

Software:

download.microsoft.com/download/6/9/4/69446ACF-E625-4CCF-8F56-58B589934CD3/Xbox360_64Eng.exe

Use a instance variable instead of "every X seconds".

Add an instance variable to enemy2 and call it "timer". Then replace your event 12 with these two events:

enemy2 : timer > 0

------- Enemy2 : Subtract dt from timer

enemy2 : timer <= 0

------- enemy2 : Spawn plasmaBullet

------- enemy2 : Set timer to enemy2.fireRate

Thank you for this. I was having a problem using the "every x seconds" as well. It's a shame you can't combine this with (or after) a trigger that selects a single object (like "overlapping an object", or "health <= 0). Was seriously driving me crazy.

Really though, it'd be a lot easier to use the built in timer (every x seconds) if you could have it reference each instance of an object individually, like most of the other conditions. I mean how often are you going to have multiple copies of the same object? Seems like every game has a few of those, and then to make them all not be the exact same you set some randomness. In my case I just wanted them to all jump a little different than each other, so a random timer would solve this perfectly. Too bad the built in one references all instances of the object, even if picking using a instance variable. I guess this just plays into the whole "not needing to program, template, reuse, repeat, instance based sprite system". I guess it's either been overlooked, or just isn't possible?

Interesting. I tried it on a different computer, just for fun. Same OS, same browser version, same gamepad, but the buttons are what they indicate, in Construct 2, on the other computer.

I'm not sure what is happening, but I re-installed the gamepad drivers just in case. Didn't change anything for me.

I'm sure I'm missing something, but lack of replies on this thread seem to indicate that this is not an issue for anyone else.

A good workaround I could think of was allowing custom button mapping. Though this seems to be a pain to set up in Construct 2. And I've not found any forum posts on the subject (though the forum search results are inundated with "help test the new gamepad object" when doing any search that includes the word "gamepad".

Anyways, ima work on the custom button mapping. I'll post my idea here after I think it through a few times.

For me I have not found the right analog stick's index. 2 and 3 are the trigger buttons, i started going up from there but didn't get a response so I stopped. Not sure if it's my controller (xbox360 wireless).

As for your original question (if you are still looking for an answer), Since the center of the stick is 0,0...

I was going to try getting the angle like this:

angle(0,0, Gamepad.Axis(0, 2), (Gamepad.Axis(0, 3))

also, i was going to trigger it with an "OR" block. So that it would only calculate it when the user was moving the stick beyond the dead zone. I had put in a custom dead zone for both axi. (rightAnalogDeadX) and (rightAnalogDeadY).

Condition:

if right analog y axis is greater than "rightAnalogDeadY"

or

if right analog y axis is less than "-rightAnalogDeadY" (negative)

or

if right analog x axis is greater than "rightAnalogDeadX"

or

if right analog x axis is less than "-rightAnalogDeadX" (negative)

I recently bought an xbox 360 wireless controller (the "for pc" version). According to what I've read it should have the best chance of compatibility (for wireless controllers). I tried searching the forums for anyone else having this issue.

I'm using chrome 25.0.13 and Construct 2 release 111

The button mapping does not seem to be accurate.

-------------------------------------

X in Construct 2 is actually A on my controller

A is B

B is X

Left shoulder trigger is Back

Right shoulder trigger is Start

Back is Left analog button

Start is Right analog button

Left analog button is nothing

Right analog is nothing

--------------------------------------

everything else seems mapped properly

As well as, sometimes it seems Chrome does not recognize the gamepad until i hit the a/b/x/y buttons.

Anyone having the same issue? Or can point me in the right direction for resolving?

Thank you,

Cym