GDC Day 2 started with more tutorials and summits. I attended the Microsoft Developer Days as they talked about DX11 and Windows Phone 7. Microsoft demonstrated how easy it was to create a quick accelerometer enabled game. It used a windows logo that reacted to tilting and was created in around 5 minutes. It showed the trivial amount of work needed to get something on the screen, granted that it was Michael Klutcher (Dev lead for XNA) who made the program. The most frustrating part of it was the limited amount of information they were allowed to tell us because they had to wait for MIX10 next week. Even the developers I talked to were frustrated that they could not reveal anything to us during GDC. 

GDC Day 3 kicks off the start of the main conference, this is where the madness begins. I expect for things to get this sign in front of them :)

The first day of GDC concluded with a very interesting trend. I have seen the social game tutorials and summits fill up, with around 400 people listening to industry leaders talk ab out social games. That is is a big change from the last GDC where there were social games summits but nothing quite this big. Also I have been able to talk to someone from the Microsoft lobby who explained that Silverlight and XNA will initially not mixable in the same mobile 7 application. Today I will be doing the Microsoft Developer Days tutorial and hopefully learn alot about XNA 4.0 and how it will work across all three platforms.

I will be demoing Scribble Defense during GDC, but for those who won't be at GDC to see the game I posted up a video of the Build that I will be showing off. Here are the following changes since the last video:

1. Added tower placement UI, the towers are not hard coded to specific keys/buttons anymore and now are handled through the new UI
2. Added support for up to 3 separate creep paths. Creeps can come from all over the place! :)
3. Added a Area of Effect Tower that can harm all creeps within its range. Good for whittling down enemy health en masse
4. More tower/creep specific animations
5. Added support for high scores

After weeks of trying to find a name for my new tower defense game I have settled for the name Scribble Defense. I chose this name because the game reminds me of scribbles I made when I was younger. Mixing contrasting white lines on a black background, as well as having colorful towers to accentuate and add colors to the game.

Currently I added some new concepts to the game. First of all the game now supports 2 player co-op. Having two players in the game will increase the challenge level to account for the added player. Also I have added a money drop system, money no longer automatically add to the player's bank, but has to be picked up by either player to be added to the bank. This makes the game more dynamic as the player has to constantly move around in order to get money he needs to buy more towers.

General Thoughts:

1. This is my third game to be published on XBLIG
2. Code base was derived from a unpublished game which had the framework setup
3. I just wanted to create a simple shooter game
4. Attempted to do cross marketing between my games
5. Development time took 2 months

What went right:

1. Very fast development time because I knew I just wanted a simple dual stick shooter
2. The code base I have been using has been mature enough that I did not spend alot of time scratching my head
3. Simpler menus allowed me to concentrate on the gameplay better

What went wrong: 

1. The cross marketing seemed to have ticked some people off, even comparing my game to the Yaris game that came out of XBLA
2. The game was not quite as impossible during the first release, I had to fix gameplay for a patch but that probably hurt my sales
3. I did not try to innovate as much with this game to differentiate it from the dozens of twin stick shooters out there 

General Thoughts:

1. For this game I tried doing a clone of existing games like Galcon, and adding my own twist to it
2. I hired an artist to try out outsourcing some of my work
3. Worked with my roommate as he developed the algorithm for planet placement
4. Took me 6 months from inception to completion on this project 
5. The game was well received but the lack of singleplayer beyond skirmishes hurt the game

What went right:

1. Hiring the artist to do all the artwork freed me to do programming and gameplay. It also increased the visual appeal of the game
2. I was working off a established code base which meant I had alot of support functionality built in 
3. I used my own collision detection and physics. Primarily because I knew exactly what was going to be needed, this freed up CPU cycles to be used somewhere else
4.  I knew exactly the game mechanics and gameplay. Unlike Gum Drop Celestial Frontier, I had games I can compare against in order to judge quality and figure out what works and what does not

What went wrong: 

1.  I did not market the game as well, so I ended up with pretty low sales
2.  I should have added some sort of single player to increase gameplay time or even hook the people more into the game
3. Tutorials were not as clear, as well as the game mechanic harder to get into than most games out there

General Thoughts:

1. This was my first XNA game, in which I was totally un-aware of the publishing process, testing process, and even the framework. I did alot of stupid mistakes, but I generally learned from them
2. Development time was 1 year because I had to get up to speed with the XNA framework 
3. I had fun making this project because it marked my return to game programming, which I almost gave up.

What went wrong:

1. I had a grand vision of what the game should be. I added so many different concepts in the game but most of them were cool on paper but not on the screen
2. It took me a while to figure out the review process and I was a bit hurried in getting the game out
3. I used farseer physics for physics and collision detection. It was an awesome library but it had inherent limitations that forced gameplay decisions to be based against tech versus actually what was fun
4. Feature creep, as I got deeper into making the game I was seduced to add more features that were cut anyways, not necessarily a bad thing, but it did extend the time 

What went right:

1. Using farseer physics to handle physics. It put me more on gameplay side without having to deal with implementing my own physics library 
2. Prototyping early on with a simple concept allowed me to get a quick idea of what I want the feel of the game to be
3. Feedback from GDC, I actually had alot of people from GDC try out the game. I got so many good ideas and feedback on the game that it helped me refine it

I recently put up the alpha version of Neo Terra Zune HD here. It is a simpler version of my Xbox game Neo Terra (Xbox.com).  You can download the game from here: Neo Terra Zune HD