Anyways, I landed a job with Music Reports, Inc. in September 2010 as a Java application developer. It’s going well. My recreational projects are pretty much entirely Android applications/games/prototypes now, so for only-slightly-more-up-to-date information check out my Android project blog here.

Power Hour is doing awesome. Passed 2,500 downloads after about 10 months on the market, keeping a fairly constant (low) download rate, but it’s nice to have a little extra cash.

War with Friends released in August of 2010 and revealed itself as the brittle house of cards that was its code base. People really liked it when it worked, but there were tons of bugs and feature requests and for one guy it was way too much work. Shortly after I got my real job I was contacted by a publisher in Seattle and I sold the rights to WWF for cold hard cash and an excuse to stop working on it, so that worked out swimmingly.

I just released a game I’ve been programming since December called “180″ on the Android Market. It’s a port of a successful iPhone game and I think it actually has some potential for success. *crosses fingers*

“Power Hour for Android” was released to the Android Market on July 6th, and I have to say I’m fairly pleased with how it’s doing so far. It’s clearly not a must-have app, and there are several other [inferior] apps like it, so I naturally wasn’t expecting ridiculous sales out of it. As of this post I’ve sold 65 units, which is more than I was expecting for the first few weeks, so that’s a plus. The first comments are positive (5 stars) and the momentum is [very, very] slowly increasing, so we’ll see what comes of it. The highest-selling Power Hour app on the market right now is in the 1,000 – 5,000 downloads range; if I can get up to 1,000 downloads even in the first year I’ll consider it a huge success. So that’s that!

My Risk clone (which I think I’ve settled on naming “War With Friends”) is coming along nicely. I recently figured out how to upload a Web Application to Google App Engine to work as the networking server to communicate with, so right now I’m typing away as fast as I can to set up user creation, friend connection, game creation/joining, and continuous updates to interface with the game code I already have written. As always, Google is making this way easier than it could be, but this is definitely a new problem for me, so it might take a bit longer to get it working robustly.

I love Risk. I certainly can’t call it that if I ever release it, but it’s been really fun to code so far. The real challenge is going to be taking the functionality from multiple players on one device to multiple players on separate devices through use of a server of some kind. I’ve read some preliminary articles on the challenges I can expect and the approach I should probably take with it. But that’s a near-future addition. Right now I’m still working on getting through all the functionality of the game. Here’s some early screen shots:

And once again, all images are hardcore placeholders.

• I’ve written it before, so I know the functionality and design issues; therefore the learning curve resides in the platform, not the program.
• It would require several interacting Activities to achieve satisfactory functionality.
• It would require use of a background Service for music playback.
• It would require querying of the phone’s internal or external storage, providing practice using Content Providers.
• The ListActivity used for building a music play list required a custom ListAdapter to make toggle-able list items
• Saving/Loading play lists would have to run in a separate thread as to not lock up the UI, so threading issues were explored.
• Interrupting a constantly-updating application required reliable and robust state preservation.

I think it was a really good exercise to learn my way around the Android platform.  I don’t currently have an Android device to test on (I’m holding out for the HTC Evo for Sprint), but testing on the emulator has been very successful. I obviously won’t release anything that hasn’t been extensively tested on real devices, but until I can do that I am very satisfied with how much I’ve learned about developing for Android.

(Images are obviously placeholders until I decide what I want to display on the app)

I’ll be out of a job on April 30th, and am unfortunately going to hold out on getting an Android phone until June when the HTC Evo comes out for Sprint. Until then I’ll continue learning all I can about it and getting a framework in place for when I have an actual phone to run things on.  I will update this blog as that continues.

It’s not finished, and by that I mean there are several features from the original game that are not implemented in my version. However, I feel I’ve learned about all I can from this game.  I did find out why the old adage “Composition over Inheritance” exists, as all my characters derive from a single PlatformActor class, so when special functionality was needed for one actor and not the others it became a bit of a mess. Now I know for next time! I’m pretty happy with the way it turned out, though.

You can look at the source code, use the level editor, and play 4 levels of the game by downloading the distributable zip file.

It’s not flashy, but it plays well.  I wrote it in C++ using the DirectX 9 SDK for the graphics and input. The DirectX wrapper code was disgusting and offensive in its first iteration, but it got the job done.

Next I moved on to BreakOut, which had slightly more complex collision detection and a more frequent main loop update cycle (Tetris was written with a main loop also, but it really could have been event-driven and the frame rate was never, ever threatened).

Again, it wasn’t flashy or anything, but the point was to learn, not to make a top-notch distributable game.

Development Blog

The project was to implement a Half-Edge Mesh data structure to represent and manipulate mesh data. The ultimate goal was to implement surface subdivision techniques (Loop, Butterfly, Catmull-Clark), but the data structure implementation was fun and challenging itself.

We had to keep a development blog and do a final write-up, so I’ll just redirect you there for details.

Development Blog

Final Write-Up

So I wrote a program to download tagged images from CollegeHumor.com.  At the command line, simply input the tag you want (e.g. “fail”) and it will access the appropriate pages, gather the image references, and download all those pictures.

The hard part was getting access to the website and finding image tags. I’d never used Python, but I knew I’d probably want to use a scripting language to sift through the HTML of the websites. So I did what every self-respecting programmer does: I found example code and modified it to do what I needed. I supplied the appropriate pages to parse based on the simple URL pattern the site uses, and the scripts extracted all the image tags, which were supplied to the main program to download to my computer. Someone more well-versed in Python probably could have skipped the entire Java aspect of it, but it got the job done.

No screen shot. I never developed it into a usable application; it just ran from the IDE with manual command line arguments.