jump to navigation

Game 1 for 1GAM 2014 – Asteroids January 14, 2014

Posted by Andor Saga in 1GAM, Game Development, Open Source, Processing, Processing.js.
add a comment

Asteroids

Skip the blog and play Asteroids!

Back in November, I picked up a contract to develop Asteroids in Processing.js. After developing the game, I lost touch with my contractee and thus $150. Soon after, I went on vacation and when I returned, I decided to polish off what I had and place it as a 1GAM entry. I added some audio, gave it a more authentic look and feel, added more effects and the like. So, this is my official release for my first 2014 1GAM!

Advertisements

Implementing PShader.set() October 5, 2013

Posted by Andor Saga in JavaScript, Processing, Processing.js, PShader.
add a comment

I was in the process of writing ref tests for my implementation of PShader.set() in Processing.js, when I ran into a nasty problem. PShader.set() can take on a variety of types including single floats and integers to set uniform shader variables. For example, we can have the following:

pShader.set("i", 1);
pShader.set("f", 1.0);

If the second argument is an integer, we must call uniform1i on the WebGL context, otherwise uniform1f needs to be called. But in JavaScript, we can’t distinguish between 1.0 and 1. I briefly considered modifying the the interface for this method, but knew there was a better solution. No, the last thing I wanted was to change the interface. So I just thought about it until I came up with an interesting solution. I figured, why not call both uniform1i and uniform1f right after each other? What would happen? It turns out, it works! It seems one will always fail and the other will succeed, leaving us with the proper uniform set!

Developing Shaders in Vim August 7, 2013

Posted by Andor Saga in Vim.
add a comment

I’ve been recently writing shaders in Vim and found some fun shortcuts to help in development.

Frequently I’ll want both vertex and fragment shaders open in the same terminal window, but I’ll want Vim to have multiple windows. In the man pages, I found a simple command line switch to open both files, one on top of another. This is known as staked windows.

vim -o vert.glsl frag.glsl

But I like seeing the shaders side-by-side, especially if there are a bunch of varying variables, so use an uppercase “O” switch instead.

vim -O vert.glsl frag.glsl

Making it Pretty

I finally decided to add some color to vim, so I looked around for a syntax file for GLSL and found one at vim.org. Download and install it to get syntax highlighting.

So after starting vim, I like running these lastline commands to set the line numbers and get the nice GLSL syntax highlighting:

:set nu
:syntax on
:set syntax=glsl

Make sure there isn’t a space between “=” and “glsl”, otherwise you’ll get an “Unknown option” error.

Navigation

Now that the windows are all set up, you can start navigating between them and editing your shaders. To manipulate the windows you use CTRL-W which enters the window command mode, then add an extra character for what you want. For example, to toggle the cursor between the windows you can use CTRL-W w. This involves holding down CTRL, pressing ‘w’ letting go of both keys and pressing ‘w’ again. To close the current window, you can use CTRL-W q in the same way. You can of course still use the lastline mode commands :wq, or :q!

If you ever need to split up the windows further you have a few options. You can use CTRL-W s to split a window horizontally or CTRL-W v to split vertically. You can also go to lastline mode and type:

:split

Or if you want to split the current window vertically,

:vsplit

I found some of the more useful window command mode commands are these:
CTRL-W w – Go to the next window
CTRL-W q – Quit the current window
CTRL-W s – Split the current window horizontally
CTRL-W v – Split the current window vertically
CTRL-W n – Create a new empty window
CTRL-W = – Make all windows equal size

But better yet, start Vim and type :help windows which will give you more information than you’d ever want about windows!

Experimenting with Normal Mapping using PShaders May 19, 2013

Posted by Andor Saga in GLSL, Processing, PShader.
4 comments

Normal_Mapping_PShader

Just over half a year ago, I wrote a blog about Experimenting with Normal Mapping. I wrote a simple Processing sketch that demonstrated the technique and I also wrote a hacked up Processing.js sketch to squeeze out some extra few frames/sec on the browser side of things.

This long weekend, I found myself with some extra time to hack on something. I remember that several weeks ago, Processing 2 introduced PShaders, which at the time I found exciting, but I didn’t have a chance to look at them. So this weekend I decided to take a look into this new PShader object. I haven’t touched shaders in a while, so I brushed up on them by reading the PShader tutorial on the Processing page.

After my refresher, I got to hacking and re-wrote my normal mapping sketch. Here is my complete sketch along with vertex and fragment shaders:

The sketch:

PImage diffuseMap;
PImage normalMap;

PShape plane;

PShader normalMapShader;

void setup() {
  size(256, 256, P3D);
  
  diffuseMap = loadImage("crossColor.jpg");
  normalMap = loadImage("crossNormal.jpg");
  
  plane = createPlane(diffuseMap);
  
  normalMapShader = loadShader("texfrag.glsl", "texvert.glsl");
  shader(normalMapShader);
  normalMapShader.set("normalMap", normalMap);
}

void draw(){
  background(0);
  translate(width/2, height/2, 0);
  scale(128);
  shape(plane);
}

void mouseMoved(){
  updateCursorCoords();
}

void mouseDragged(){
  updateCursorCoords();
}

void updateCursorCoords(){
  normalMapShader.set("mouseX", (float)mouseX);
  normalMapShader.set("mouseY", height - (float)mouseY);
}

void mousePressed(){
  normalMapShader.set("useSpecular", 1);
}

void mouseReleased(){
  normalMapShader.set("useSpecular", 0);
}

PShape createPlane(PImage tex) {
  textureMode(NORMAL);
  PShape sh = createShape();
  sh.beginShape(QUAD);
  sh.noStroke();
  sh.texture(tex);
  sh.vertex( 1, -1, 0, 1, 0);
  sh.vertex( 1,  1, 0, 1, 1);    
  sh.vertex(-1,  1, 0, 0, 1);
  sh.vertex(-1, -1, 0, 0, 0);
  sh.endShape(); 
  return sh;
}

The vertex shader:

#define PROCESSING_TEXTURE_SHADER

uniform mat4 transform;
uniform mat4 texMatrix;

attribute vec4 vertex;
attribute vec2 texCoord;

varying vec4 vertTexCoord;

void main() {
  gl_Position = transform * vertex;
  vertTexCoord = texMatrix * vec4(texCoord, 1.0, 1.0);
}

The fragment shader:

#ifdef GL_ES
precision mediump float;
precision mediump int;
#endif

#define PI 3.141592658

uniform sampler2D normalMap;
uniform sampler2D colorMap;

uniform int useSpecular;

uniform float mouseX;
uniform float mouseY;

varying vec4 vertTexCoord;

const vec3 view = vec3(0,0,1);
const float shine = 40.0;

void main() {
  // Convert the RGB values to XYZ
  vec4 normalColor  = texture2D(normalMap, vertTexCoord.st);
  vec3 normalVector = vec3(normalColor - vec4(0.5));
  normalVector = normalize(normalVector);

  vec3 rayOfLight = vec3(gl_FragCoord.x - mouseX, gl_FragCoord.y - mouseY, -150.0);
  rayOfLight = normalize(rayOfLight);

  float nDotL = dot(rayOfLight, normalVector);

  vec3 finalSpec = vec3(0);

  if(useSpecular == 1){
    vec3 reflection = normalVector;
    reflection = reflection * nDotL * 2.0;
    reflection -= rayOfLight;
    float specIntensity = pow( dot(reflection, view), shine);
    finalSpec = vec3(1.0, 0.5, 0.2) * specIntensity;
  }

  float finalDiffuse = acos(nDotL)/PI;

  gl_FragColor = vec4(finalSpec + vec3(texture2D(colorMap, vertTexCoord.st) * finalDiffuse), 1.0);
}

Performance

I found using PShaders very exciting, since I could place all this computational work on the GPU rather than CPU. So I wondered about the performance vs my old sketch. I’m on a mac mini, and after running tests I found my original normal mapping sketch ran at 30FPS with diffuse lighting and it ran at 21FPS using diffuse and specular lighting. Using PShaders, I was able to render specular and diffuse lighting at a solid 60FPS. Keep in mind the first sketch is 2D and my new one is 3D, so I’m not sure if that comparison is fair.

No Demo? 😦

Sadly, Processing no longer allows exporting to applets, so I can’t even post a demo running in Java. The perfect solution would be to implement the PShader in Processing.js, which is something I’m considering….

Game 2 for 1GAM: Tetrissing May 17, 2013

Posted by Andor Saga in 1GAM, Game Development, Open Source, Processing, Processing.js.
add a comment

tetrissing

Click to play!
View the source

I’m officially releasing Tetrissing for the 1GAM challenge. Tetrissing an open source Tetris clone I wrote in Processing.

I began working on the game during Ludum Dare 26. There were a few developers hacking on LD26 at the Ryerson Engineering building, so I decided to join them. I was only able to stay for a few hours, but I managed to get the core mechanics done in that time.

After I left Ryserson, I did some research and found most of the Tetris clones online lacked some basic features and has almost no polish. I wanted to contribute something different than what was already available. So, that’s when I decided to make this one of my 1GAM games. I spent the next 2 weeks fixing bugs, adding features, audio, art and polishing the game.

I’m fairly happy with what I have so far. My clone doesn’t rely on annoying keyboard key repeats, and it still allows tapping the left or right arrow keys to move a piece 1 block. I added a ‘ghost’ piece feature and kickback feature, pausing, restarting, audio and art. There was nothing too difficult about all this, but it did require work. So, in retrospect I want to take on something a bit more challenging for my next 1GAM game.

Lessons Learned

One mistake I made when writing this was over complicating the audio code. I used Minim for the Processing version, but I had to write my own implementation for the Processing.js version. I decided to look into the Web Audio API. After fumbling around with it, I did eventually manage to get it to work, but then the sound didn’t work in Firefox. Realizing that I made a simple matter complex, I ended up scrapping the whole thing and resorting to use audio tags, which took very little effort to get working. The SoundManager I have for JavaScript is now much shorter, easier to understand, and still gets the job done.

Another issue I ran into was a bug in the Processing.js library. When using tint() to color my ghost pieces, Pjs would refuse to render one of the blocks that composed a Tetris piece. I dove into the tint() code and tried fixing it myself, but I didn’t get too far. After taking a break, I realized I didn’t really have the time to invest in the Pjs fix and also came up with a dead-simple work-around. Since only the first block wasn’t rendering, I would render that first ‘invisible’ block off screen, then re-render the same block onscreen the second time. Fixing the issue in Pjs would have been nice. But that wasn’t what my main goal was.

Lastly, I was reminded how much time it takes to polish a game. I completed the core mechanics of Tetrissing in a few hours, but it took another 2 weeks to polish it!

If you like my work, please star or fork my repository on Github. Also, please post any feedback, thanks!

Game 1 for 1GAM: Hello Hanoi! February 27, 2013

Posted by Andor Saga in 1GAM, Game Development, Objective-C.
add a comment

hello_hanoi

My first iPhone game is available for download on the App Store. I call it Hello Hanoi!. It’s my “Hello World” for iPhone game development via Towers of Hanoi. My motivation to code and released the game came from the 1GAM challenge. A year-long event that dares developers to release 1 game every month.

When planning the game, I wanted to keep the scope very tight. At the end of it, I wanted a polished, tightly scoped game over a feature-rich, unpolished game. I think I managed to achieve that, but for my next game, I’d like to try the opposite.

The Pixels

I had a difficult time getting a hold of artists, so I decided to just learn pixel art and make all the assets myself. To begin making the art, I tried a few applications: Gimp, Aseprite, and Pixen. Gimp had issues with leaving artifacts on the canvas. Aseprite had problems with cursor position and I felt the UI was awkward. Pixen kept crashing. It was a bit frustrating so I restarted and re-installed them all. I launched Pixen first, and it seemed to work, so I stuck with it.

The result of making all the art myself shifted the release date dramatically. I should have released at the end of January and it’s almost March. At the same time, I had a lot of fun learning pixel art and learning about art in general, such as attention to color, lighting, shadows, and mood.

One particular level was very tedious to create and I soon realized I could generate the art instead! So, I launched Processing and wrote a small sketch to create a series of city buildings. It doesn’t look as good compared to what I could have done by hand, but it was a lot faster to create with this method.

The Code

The code was straightforward, but I did have to learn a few iOS specifics. How do I write a plist to storage? How do I use the new storyboards? Like the art, it was at times frustrating, but in the end it was worth it.

One mistake I did make was over-generalizing. I had the idea that it would be neat to support n number of discs. I wrote all the code to handle the rendering and positioning of the discs, but then realized it didn’t fit into the game. Players would get bored before they reached that many discs, and the necessary horizontal scaling of the disc assets would break the art style. So, I ended up taking the code out. Next time, I’ll try to catch myself over-generalizing.

I had a fun time making my first 1GAM game and I look forward to trying something new for the next one!

Automating my WorkFlow with TexturePacker’s Command Line Options February 17, 2013

Posted by Andor Saga in BNOTIONS, Game Development, Shell Scripting, TexturePacker.
add a comment

At work I use TexturePacker quite a bit to generate sprite sheets for UIToolkit from within Unity. Each sprite sheet ends up corresponding with a layer in UIToolkit, so the sheets we use are typically named “layer0.png”, “layer02x.png”, “layer0.txt”, “layer02x.txt”, where the .txt files are the metadata files.

During development, I’ll be interrupted every now and then with new and improved assets that need to pushed up into the Unity application. Once I have received the asset, I must take a series of steps to actually use that asset in Unity.

I open the TexturePacker .TPS file and drag the asset in. The sprites get re-arranged then I would perform the following:

  • Set the sheet filename for use with retina mobile devices, adding “2x”
  • Set the width of the sheet to 2048
  • Set the height of the sheet to 2048
  • Set scaling to 1.0
  • Click publish

I would then need to do the same thing for non-retina devices. UIToolkit will automatically select the appropriate sized sprite sheet to use based on the “2x” extension.

  • Set the data filename for use with non-retina mobile devices, removing “2X”
  • Set the width of the sheet to 1024
  • Set the height of the sheet to 1024
  • Set scaling to 0.5
  • Click publish

Once TexturePacker creates the 4 files (layer0.png, layer0.json, layer02x.png, layer02x.json ), I would rename the .json files to .txt to keep Unity happy. This process would be done over and over again, until insanity.

Automating Things

This weekend I had some time to investigate a way to automate this tedious process. I wanted a solution such that after editing the .TPS file, I could simply run a script that would take care of the rest of the work. I began by looking into TexturePacker’s command line options. After some tinkering, I came up with a short script that reduces the number of click and edits that I need to make.

I placed the script within the directory that contains all our assets. However, since the output sheets go into a folder that only contain the sheets and data files, so I need to reference these paths relative from where the shell script lives. So, this would be a script for one layer for UIToolkit:

#!/bin/bash
TexturePacker --sheet ../Resources/layer0.png    --data ../Resources/layer0.txt    --scale 0.5 --width 1024 --height 1024 --format unity layer0.tps
TexturePacker --sheet ../Resources/layer02x.png  --data ../Resources/layer02x.txt  --scale 1.0 --width 2048 --height 2048 --format unity layer0.tps

Note that I can omit changing any of the options that are already set in the .TPS file such as padding, rotation, algorithm, etc. This helps keep the script short and sweet.

The options all correspond with the changes that I mentioned previously. One interesting thing to note is the –format option which prevents needing to rename the .json data file to .txt. You might ask why I didn’t just set this option in the TexturePacker GUI. The reason is, I had just learned about this option after looking over the command line help!

I had created a script that I could run through command line, but I wanted the script to be easier to use. If I had just edited the .TPS file, I would be in Finder, and being able to double-click the script would be nicer than opening a terminal to execute it. Running the script through Finder would simply return an error since the terminal would be in my home directory.

To fix this issue, I had to modify the script a bit more:

#!/bin/bash
DIR="$( cd "$( dirname "$0" )" && pwd )"
TexturePacker --sheet "$DIR"/../Resources/layer0.png    --data "$DIR"/../Resources/layer0.txt    --scale 0.5 --width 1024 --height 1024 --format unity  "$DIR"/layer0.tps
TexturePacker --sheet "$DIR"/../Resources/layer02x.png  --data "$DIR"/../Resources/layer02x.txt  --scale 1.0 --width 2048 --height 2048 --format unity  "$DIR"/layer0.tps

When the shell script runs, we get the directory name where the script lives, cd into that directory and call pwd and assign that value into our DIR variable. This part took a bit more time as I learned spaces on either side of the equals sign will confuse bash, so I had to leave those out.

Now, if a new asset is sent to me, I open the .TPS file, add the file and save. Then I can run this script by a simple double-click. Tada!

Next steps

Using this method, I need to create a shell script for every UIToolkit layer. This isn’t nearly as bad as it sounds since we typically only have 2-3 layers. But, what I’d like to do in the future is investigate AppleScript. AppleScript can help convert a shell script into an app that allows files to be dropped on that app. If I did this, I could drop the .TPS file onto app, then the the script could extract the filename and do the rest. This would prevent needing a script for every layer.

Sprite Sheet Guide Generator January 13, 2013

Posted by Andor Saga in Game Development, Pixel Art, Processing.js.
add a comment

sprite sheet guide generator
Click the image above to get the tool.

I began using Pixen to start making pixel art assets for a few games I’m developing for the 1 game a month challenge.

While pixelating away, I found myself creating a series of sprite sheets for bitmapped fonts. I created one here, then another, but by then I found myself running into the same problem: before I began drawing each glyph, I first had to make sure I had a nice grid to keep all the characters in line. Each font used a different number of pixels, so I had to start from scratch every time. You can imagine that counting rows and columns of pixels and drawing each line separating glyphs is extremely tedious. I needed something to eliminate this from my workflow.

I decided to create a decent tool that took away this painful process. What I needed was a sprite sheet guide generator, a tool that created an image of a grid based on these inputs:

  • Number of sprites per row
  • Number of sprites per column
  • Width of the sprite
  • Height of the sprite
  • Border width and color

I used Processing.js to create the tool and I found the results to be quite useful. After almost finishing the tool, I realized I could alternate the sprite background colours to help me even more when I’m drawing down at the pixel level, so I implemented that as well.

You can run the tool right here or you can click on the image at the start of this post.

New Year Habits December 31, 2012

Posted by Andor Saga in Habits, Personal Development.
1 comment so far

Last year I decided to begin exercising. Not for a few months or a couple years, but for the rest of my life. I made a resolution to do some form of physical or mental exercise for 15 minutes a day every day. 15 minutes may sound meager, but by making a small habit change, I could later on extend my time without too much difficulty. I managed to keep my commitment to bike, run, meditate, or do yoga every day.

I had some slips when I missed a few days while moving apartments and traveling, but this small habit change got me to start running, and I eventually ran my first half marathon (: So, it worked out well.

If you’re setting some New Year’s resolutions, here are some tips on forming your new habits:

Make it quantifiable

You must be able to measure your goal so it can serve as an indication that you are on the right track. Either you did it or you didn’t, there should be no ambiguity. Setting a number to your goal is the easiest way to prevent this ambiguity. My metric was 15 minutes. Either I spent those 15 minutes on myself or I didn’t. So figure out: How long? How many pages? How many phone calls? What time? Put a number on it so when you are done, you can tick it off.

Keep a log

‘Ticking it off’ is an important step in habit forming because it helps motivate you and, like I mentioned before, it shows your progress. Jerry Seinfeld has a productivity secret: Don’t Break the Chain. This is a great tool that I use to help me on working on my goals. Upon completing your task, tick it off your log to track your progress.

When you fail…Start over

Don’t be hard on yourself if/when you fail, simply start over. Last year I did some traveling, which threw my routine out of whack. I ended up missing some days of exercise. I could have beat myself up over this, but that would have been counter-productive. If you go back to your old habits, simply start over.

Do what works for you

Don’t set yourself up for failure by doing something you hate. Forming new habits can be difficult, so there’s no need to make it even more challenging. You know yourself best, so choose a task that is somewhat enjoyable. Personally, I love stationary bikes. It’s hands-free, I can listen to music, and I can enjoy the scenery while I pedal away. So, if you want to succeed in your new habits, make sure to do what works for you.

Good luck! (:

BitCam.Me December 25, 2012

Posted by Andor Saga in Open Source, Pixel Art, Processing.js.
add a comment

bitcam_me_asalga

Check this out: I created a WebRTC demo that pixelates your webcam video stream: BitCam.me.

I recently developed a healthy obsession with pixel art and I began making some doodles in my spare time. Soon after I started doing this, I wondered what it would be like to generate pixel art programmatically. So I fired up Processing and made a sketch that did just that. The sketch pixelized a PNG, taking the average pixel color of the nearest neighbor pixels.

After completing that sketch, I realized I could easily upgrade what I had written to use WebRTC instead of a static image. I thought it would be much more fun and engaging to use this demo if it was in real-time. I added the necessary JavaScript and I was pretty excited about it (:

I then found SuperPixelTime and saw it did something similar to what I had written. But unlike my demo, it had some nice options to change the color palette. I read the code and figured making those changes wouldn’t be difficult either and soon had my own controls for changing palettes.

I had a great time making the demo. Let me know what you think!

Enjoy!