HemingwayGames

Members
  • Content Count

    30
  • Joined

  • Last visited

About HemingwayGames

  • Rank
    Advanced Member

Profile Information

  • Gender
    Not Telling

Contact Methods

  • Twitter
    @HemingwayGames
  1. Hi everyone, I've just published Postbug on Kongregate. If you have a Kongregate account, feel free to fire it up. Any support would be appreciated :-) Postbug is a HTML5 game which plays in Chrome or Firefox on desktop computers: http://www.kongregate.com/games/HemingwayGames/postbug This new version includes a Premium mode which includes new bugs to play:
  2. Hi Pixi team, I just wanted to say thanks for the fantastic effort behind the latest Pixi.js version. I recently upgraded from 1.6.2 to 3.0.7 and everything is running smoothly. Thanks also for xerver's help on the forums while I was upgrading. I've also just released my game on Kongregate. Feel free to check it out if you play on Kongregate, and I'd appreciate any support http://www.kongregate.com/games/HemingwayGames/postbug
  3. Thanks for the response, AshleyScirra. xerver, I'll keep an eye out on issue 1738. It would be good to fall back to canvas or even drop support for the browser if WegGL fails. Cheers
  4. Hi there, I'm having some performance issues with IE on win7. As a result I've dropped IE support for my game. It appears IE 11 is disabling webGL and rendering with a software renderer. IE 11 console: WEBGL11258: Temporarily switching to software rendering to display WebGL content.WEBGL11135: getContext: Context attribute premultipliedAlpha:false is not currently supportedMicrosoft has reported this issue: https://connect.microsoft.com/IE/feedback/details/812301/ie-11-uses-software-rendering-for-webgl-after-installing-update-kb2884101 This issue was also discussed on html5gamedevs.com back in 2013: http://www.html5gamedevs.com/topic/2563-internet-explorer-11-webgl-bug/ Is anyone else experiencing this issue (or is just me because I'm missing a patch)? I heard that win8 IE11 is OK so maybe this is a win7 IE11 issue. It seems the best way around this is to force canvas rendering for E11. Since we're on the topic of IE, I'm curious to also know what IE versions you are supporting. I'm thinking of just supporting IE11 with canvas (if I can get reasonable performance)... Thanks
  5. Sorry I believe the JS needs to be corrected instead of the c++. Even though sqrt() returns a float, the value should be cast/converted to an integer value so that the decimal component is removed (NB the value will need to be truncated instead of rounded). Javascript: i_sqrR = Math.floor(Math.sqrt(i));
  6. That's an interesting test you have there. I'd however prefer to see an algorithm that didn't use system libraries (eg sqrt) to ensure the implementation is the same. I received different results from your code. C++ (Visual Studio 2010) I found 665111 prime numbers between 1 and 10000000 in 7952 micro seconds. Javascript: There are 665026 prime numbers between 1 and 10000000. Execution time: 12909milliseconds A couple of things to consider: - Firstly, both programs did not find the same number of prime numbers. I haven't had a chance to debug, however my initial guess is the javascript version is not handling types correctly (eg i_sqrt should be cast back to an integer/unsigned long). What results did you receive? - More importantly, the clock routine reports on 'clock ticks' instead of milliseconds. http://www.cplusplus.com/reference/ctime/clock/ http://stackoverflow.com/questions/1516659/how-do-i-count-how-many-milliseconds-it-takes-my-program-to-run
  7. My game uses a fixed frame rate for movement and animation. i.e. it assumes that each frame will run at 30fps regardless of how fast the game is running. Just posting this here in case anyone else finds this helpful. Opt-out of the auto ticker before your application runs: // opt-out of auto tickervar ticker = PIXI.ticker.shared;ticker.autoStart = false;ticker.stop();ticker.speed = 0.5; // run at 30fpsCall the ticker and increment the time with a fixed value. var ticker = PIXI.ticker.shared;var forceCurrentTime = ticker.lastTime + (1000.0 / 30.0); // + 30fps//var currentTime = performance.now();ticker.update(forceCurrentTime);This is working fine for now, however this may cause problems if the ticker ever compares ticker.lastTime with performance.not() because ticker.lastTime will not be inline with the clock...
  8. For my own future reference, this is the code which calculates deltaTime for the ticker: Ticker.prototype.update = function update(currentTime)...this.deltaTime = elapsedMS * CONST.TARGET_FPMS * this.speed;// Invoke listeners added to internal emitterthis._emitter.emit(TICK, this.deltaTime);Debug session: elapsedMS: 16.999CONST.TARGET_FPMS: 0.06this.speed: 1this deltaTime: 1.02
  9. Thanks xerver. This is exactly what I was after :-)
  10. I'm in the process of upgrading from 1.6.2 to 3.0.7 and have noticed that the animationSpeed is now time based instead of frame based. Pixi.js 1.6.2 (frame based): this.currentFrame += this.animationSpeed; var round = (this.currentFrame + 0.5) | 0; this.currentFrame = this.currentFrame % this.textures.length; Pixi.js 3.0.7 (time based): this._currentTime += this.animationSpeed * deltaTime; var floor = Math.floor(this._currentTime); Currently I'm running my game at 30fps and my game logic relies on a fixed framerate. In the old version, I'd set animationSpeed to 0.5 in order to update the animation every 2 frames (15fps). Now in v3, the animation speed seems to run at full speed. What should I now set animationSpeed to for this example? i.e. 15fps animation speed for for a game running at 30fps? I'm not yet sure how deltaTime is calculated and whether it is dependent on CONST.TARGET_FPMS (which I'm not using btw). Also is there any way of keeping the animation speeds fixed? At this stage I'm thinking of creating my own MovieClip or updating pixi's MovieClip to support fixed frame. Thanks for your help.
  11. Hi Oyed, You might want to try rendering your rectangles at position (0,0). My understanding is this position will be relative to the graphics object's position. squares.drawRect(0, 0, squares.width, squares.height); For example if the graphic object's position is (10, 10) and you render your rectangle at position (10,10), then the rectangle will render at global position (20, 20). Your new rectangles may actually be rendering off the screen. You might also want to check what the graphic object's dimensions are (ie squares.width and squares.height). I personally would use my own variables for the dimensions: var width = .. var height = .. squares.drawRect(0, 0, width, height);
  12. My new blog post on the Freeplay Independent Games Festival and exhibiting Postbug at the Freeplay Fete ‪ http://hemingwaygames.com/blog/postbug-freeplay-fete
  13. Sorry Gustavgb, I only just saw your post. This is now fixed. The touchscreen block can now be bypassed by pressing any key on the keyboard. See how you go with this, and I'd love to hear your feedback on the game. My brother says that this runs well on his Surface Pro, and I soon hope to begin testing on these these devices soon (with touch input). I'm having touble working out how to identify phones and tables, but not excluding devices such as Windows 8 at the same time. I want to avoid using the user agent (which is the general advice out there). This is probably a question I'm going to throw at the html5gamedev community sometime soon. I thought it'd be a good idea to detech touch devices, however as you mentioned this line is becoming blurry. Google Chrome has also started sending false positives on Windows 8 for devices which don't have touch support. I figure this is to suggest the browser can handle it if the hardware does, however it would be good to have better hardware detechtion in browsers. My solution so far is to detect touch devices, but then allow the user to override with a keyboard (as I mentioned earlier).
  14. A new blog post on my experience exhibiting Postbug at the Melbourne iFest Games Festival http://hemingwaygames.com/blog/postbug-ifest-games-festival
  15. I've just put together a new blog post on Postbug's game design objectives: Postbug is a game which encourages experimentation, exploration, and strategy building during gameplay. An aim of Postbug was to create an emergent system consisting of objects with a diverse range of attributes and behaviours. These objects include grass, dirt, rock, water, lava, bugs, eggs, spiders, fish, and crabs. In order to survive, players will need to experiment with their environment to learn how these objects behave and interact. More details here: http://hemingwaygames.com/blog/postbug-game-design-objectives