Ashley's Forum Posts

  • I don't think our UI library supports breaking out the tabs in to their own window, so it would be difficult to do I think.

    However there's an obscure trick which I didn't even realise was possible until someone *else* discovered it: click and drag one of the tabs in to the main view, and it will create a split screen view. You can use this to view a layout and event sheet at the same time. With some careful resizing, you can make the main C2 window really wide, set the separator just between your monitors, and you kind of have a silly version of dual monitor support, heh.

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • No, this won't work on Construct 2, because generally servers need to know the correct file extension. Plus it's not really security, just obscurity.

  • Yeah, a lot of publishing methods require your own hosting. TBH hosting is dirt cheap these days and all you need is a static file host which should be cheaper than the ones that support server-side scripting. For a tenner a month you can probably get an acceptable file host (anyone have any recommendations?)

    You could also try plugging in a Scirra Arcade URL, but that might be against their terms of service since you're using another arcade, plus it will show our embed ad.

    • Post link icon

    Hi all,

    Mozilla and Google are working on gamepad input for their browsers. It's very much experimental, but is basically the coolest thing ever. I've been playing around with it and have got some simple input being detected.

    Unfortunately, the input is a mess. For one controller, the list of buttons and joystick axes can come back in completely different orders (sometimes even buttons counted as axes and vice versa) depending on the browser, the controller and the operating system! For example my XBox 360 controller has the buttons in a different order between Firefox and Chrome, and apparently (although I haven't tried yet) it changes again if you go to Mac, and so on and on.

    However if we do a lot of testing with different controllers, browsers and OS's, I can build mappings in to the gamepad object that makes it consistent everywhere. For example, regardless of whether you've got a Playstation 3 controller or XBox 360, Firefox or Chrome, or Mac or Windows, it should consistently work with "left analog", "d-pad", "face button 1" and so on. I need help testing this though! It would be very nice if this was working smoothly with all common setups when support arrives in the mainstream browsers.

    <font size="5">How to help</font>

    You can help by digging out any old controllers you have that plug in to your PC. Then, you need to install special versions of Firefox or Chrome which support Gamepad - it's not in the mainstream releases yet.

    Firefox 11 nightly with Gamepad support:

    http://people.mozilla.com/~tmielczarek/mouselock+gamepad/firefox-11.0a1.en-US.win32.zip

    Chrome 18 beta:

    https://www.google.com/landing/chrome/beta/

    Once installed, you need to go to chrome://flags and click 'Enable' for 'Enable Gamepads' to turn on support.

    Warning: installing beta versions of Chrome can sometimes overwrite your user profile. e.g. in the past I've had trouble going from a beta version of Chrome back to the stable one. Back up your profile first if you can, or just skip testing. The Firefox install is just a zip with the EXE inside it so it should run totally standalone.

    <font size="5">Run the test</font>

    It's here:

    http://www.scirra.com/labs/gamepadtest/

    Nothing will happen until you press a button on the gamepad. Then, if you're properly set up, a list of buttons and axes should appear, with the name of the gamepad at the bottom.

    One by one you can then test each axis and button, and see what order they are listed in and write down the results.

    <font size="5">Results for Windows/XBox 360 controller on Firefox & Chrome</font>

    Here are my results to show the idea - listing the OS, browser, controller name as it appears in the test, and button and axis mappings.

    Windows - Firefox

    "45e-28e-Controller (XBOX 360 For Windows)"

    Button 0: A

    Button 1: B

    Button 2: X

    Button 3: Y

    Button 4: Left shoulder button

    Button 5: Right shoulder button

    Button 6: Back

    Button 7: Start

    Button 8: Left analog button

    Button 9: Right analog button

    Axis 0: Left analog stick X axis

    Axis 1: Left analog stick Y axis

    Axis 2: Shoulder triggers (left up to 100, right down to -100)

    Axis 3: Right analog stick X axis

    Axis 4: Right analog stick Y axis

    Axis 5: D-pad left and right

    Axis 6: D-pad up and down

    ======================

    Windows - Chrome

    "Xbox 360 Controller (XInput STANDARD GAMEPAD)"

    Button 0: A

    Button 1: B

    Button 2: X

    Button 3: Y

    Button 4: Left shoulder button

    Button 5: Right shoulder button

    Button 6: Left shoulder trigger (up to 100)

    Button 7: Right shoulder trigger (up to 100)

    Button 8: Back

    Button 9: Start

    Button 10: Left analog button

    Button 11: Right analog button

    Button 12: D-pad up

    Button 13: D-pad down

    Button 14: D-pad left

    Button 15: D-pad right

    Axis 0: Left analog X axis

    Axis 1: Left analog Y axis

    Axis 2: Right analog X axis

    Axis 3: Right analog Y axis

    <font size="5">Please note...</font>

    We only need each combo of OS-browser-controller tested once. If your combo is already tested in this thread, in order to keep the thread straightforward, please don't post it again. However, it's worth noting if yours is different to anyone else's test, just in case there's something funny going on.

    Thanks everyone and hopefully when Gamepad support comes it will be awesome :D

    Edit: if you want a quick test, try Space Blaster with Gamepad support:

    http://www.scirra.com/labs/spaceblaster-gamepad/

    Probably only supports XBox 360 controller at the moment.

  • To be honest it's their loss if they make it hard to publish Windows 8 Metro apps - there's a tonne of other places to publish HTML5 games. So let's hope it's a smooth experience.

  • You should use sprites for that too. Adding collision detection to the Particles object would be extremely CPU intensive.

  • You do not have permission to view this post

  • Why didn't you make C2 a web app?

    For a different reason: the editor is considerably more complex than the runtime - maybe 10-20x as much code and complexity. The games already are pushing the technology to the limit, so it's much more doubtful that a full blown editor on the web would work at all. Also, our prior experience is with C++/Windows, and to give our startup the best chance of surviving we stuck to what we know so we can work far more quickly than if we changed platform for the editor (you've seen our weekly updates, would probably be monthly on any other platform). The editor *would* be better and more convenient if it ran in HTML5 as well, but for now we can make the games better and more convenient by making them run in HTML5.

    ou see, I just might, once in a while, be offline.

    HTML5 games work offline too, we support that. Even if you're online, the offline cache means it only downloads once ever and never downloads again until you update it, making it exactly the same size download as if it were a traditional game - possibly less if parts of the game are streamed as you play, but we don't support that yet. There's lots of side benefits to that as well like auto-updating all your players at once, eliminating the chance anyone is stuck with an out of date downloaded copy.

    t's all about convenience, Ashley.

    I still doubt that. You're just one bookmark away on any platform. Browsers like Chrome make it even easier by having icons for Chrome Web Store apps which you can launch right from the new tab page. Windows are making HTML5 integrated in to Windows itself, and Mozilla are working on web apps for all platforms. I think the real issue is a social one. People aren't used to playing games on the web via a browser, or think of the web as little casual games. I don't think that's to do with convenience though, nor is it to do with the technology. It's a shift in perception to taking the web more seriously as a gaming platform. I think that shift probably has already started, given the fact games like Bastion are now running in a browser. Then there's more on the way like the Gamepad API so you can use XBox controllers and others to control games in a browser. It's 100% going to be a solid gaming platform in the future, it's just people don't think of it that way yet, and I think how people think about EXE games is the real argument behind us making an EXE exporter... it's just we don't have the resources, and I think people's perceptions will change with time too.

    Hopefully in future Mozilla or someone will make a good EXE wrapper for HTML5 anyway, then everyone will be happy in the end :)

  • I think you actually need a full install of Visual Studio to make Windows 8 metro apps, even if they're purely HTML + JS. Kind of a pain.

  • To be honest there are a million options we could add to the particles object. And if we added them all it would be unusable. I prefer to keep it simple and also high performance.

    If you need really custom effects I think it would be best to do what you would have done before the particles object was added - spawn sprites with the bullet movement.

  • OK, I think I've fixed this for the next build - neither particles nor sprite properly took in to account the object time scale. However you should be aware "Set object time scale to 1" is not the same as "Restore object time scale". The former means "use a time scale of 1 regardless of what the global time scale is", whereas the latter means "use the global time scale".

  • RacerBG - I wanted to wait a couple of days to ensure there were no serious new issues introduced. Does everyone think its OK to autoupdate this?

  • Fixed for next build, but the angles will still not work properly unless you use a perfectly symmetrical collision mask. Try using a bounding box collision mask in the next build.

  • Fixed in next build!

  • Oops, should be fixed in next build.