Blockshooter is a combination of Breakout and Space Invaders. It’s your job to pilot a little ship, shooting at coloured blocks that come your way. Shooting multiple blocks of the same colour will award you a point multiplier. Destroying entire segments will drop a “piece”, which can be collected and added to your ship. You lose points by getting hit or allowing blocks to slip past you.

That’s the nuts and bolts of it. If you’d like to give it a go, hit the jump for a download link and developer’s diary.

PS. Yes, this has come out on Monday instead of Friday. I’ve decided to shift the Game A Week schedule by two days, so I can have the weekend to playtest and debug. Don’t worry, I’m sticking with the whole “seven days to code” requirement.Download Blockshooter V1.0.

If you’re running Vista, you’re ready to go. If you’re running Windows XP, or get the error “App failed to initialize properly (0xc0000135)”, you will need to download the .NET Framework 2.0.

Blockshooter was by far the most painful game to develop. Not only did I have gameplay issues, but technical ones as well.

In the perfect world of game design, one should be able to come up with a concept and be able to implement it without having to worry if it can actually be programmed. Case in point: I wanted to code a shoot em’ up that combined Breakout with Space Invaders, but at the same time, not use graphics.

Impossible, right?

Fortunately, I had a solution – I would use controls in place of sprites. If you’re not familiar with the term “control”, it simply refers to any element of a Windows “form”. This can be a button, picture box, text box, label, etc. All you have to do is set a background colour, manipulate the control’s position during runtime and bang – you have graphics… of a kind.

The trouble is, controls aren’t really supposed to move during runtime, so I had significant performance issues. Even on a dual-core 1.8GHz PC with 2GB of RAM, early prototypes would slow to a crawl if more than 16 blocks were moving on screen. I had to find a way to either a) optimise the movement of blocks or b) reduce the amount of blocks. In the end, I placed blocks inside group panels, and just moved the panels instead of individual blocks. Of course, this broke the collision detection code, and I had to make a custom control to fix it.

Sadly, Blockshooter became a collection of these “hacks” to make the Windows GDI a gaming platform. As you can see by playing, it all works, but I get a little sick in my stomach when I look at the code.

I’m also not happy with the shipbuilding interface. It could be a lot more user friendly. I’m hoping everyone gets the hang of it, but I’m not expecting them to. I had to rush it, thanks to the technical issues I had with the GDI. I might do a revamp in the next few days, but I’m really keen on my next game, which I’ll tell you about soon.

Long story short – don’t use controls as sprites. It’s messy.

If you’d like the source, just hit me with an email.

Download Blockshooter V1.0.

Game A Week 2 is done. Done I say! Zafehouse is ready to be played by all.

As I said earlier in the week, this thing almost killed me. I had an idea that just exploded. I think the end result is quite awesome. Zafehouse has strategy, survival horror and, best of all, zombies. I love shufflers, I just love them to death, and I’m glad I got to put them into a completed game.

I have yet more plans for a version 2, which I’ll release when it’s done. For now, I have to get started on Game A Week 3. If you’d like to check out Game A Week 1, Wizkill, hit up this post.

Hit the jump for a development breakdown, hints and tips and unimplemented features, or download the game and give it a spin.Update: If you’re running Vista, you’re ready to go. If you’re running Windows XP, or get the error “App failed to initialize properly (0xc0000135)”, you will need to download the .NET Framework 2.0.

Update 2: Small bug with the raiding code’s been fixed. Please download the game again to get the updated executable!

Update 3: New version, v1.2. Here’s the info:

-Clarified help rules on buildings and resources
-Reduced the amount of meds found in a raid from 4 to 3
-Fixed the “Hours since.”.. label on the GUI so its value is never hidden
-Pressing “~” while a person is selected will heal them

Zafehouse began as a random name generator. I wanted to make something different to Wizkill, which involved one character (you), into a strategy game that had many. They needed a little personality, so I started by entering about 100 first names and 50 last names into two arrays.

Zombies had to be in there as well. I’m a fan of games that demand you manage tight resources and make important decisions on a minute-by-minute basis, and these concepts turned into the idea of holding off a bunch of zombies for days on end until help arrives.

Coding began mostly with the interface. A good UI was very important. Players would need to be able to assess resources at a glance, assign weapons and ammunition easily, and move survivors around quickly. I tinkered with about ten different designs, until I came up with the one in the picture above. The graphical resource counters were added last night.

About a day into the proceedings, I realised 120 days was too much. I reduced it to 72, which gives a good balance of longevity, and allows the gamplay to “develop”. Shortly after this change was made, I found I didn’t like the idea of days, as the intensity wasn’t there. So I changed it to hours.

The game is turn-based, with the actions of each hour affecting the encounters you have. I made a risky moving forcing the player to watch the results of their planning rather than being able to directly affect combat and raiding. It puts more weight on making good decisions beforehand, and makes your mouse hover over the “Advance one hour” button in tense contemplation.

Gameplay involves raiding for supplies, securing buildings for cover and resource bonuses and allocating weapons, barricades and ammo in an intelligent fashion. It seems basic at face value, but as the game progresses you’ll find yourself keeping careful check of who has your only shotgun, or if David has enough ammo to last the hour.

The amount of zombies you encounter and the chances of encountering them is determined by something I like to call the “Zombie Mass Index”. The ZMI is increased by two every hour, and reduced by one every time you win a fight. This way it regulates itself, and puts pressure on the player to be offensive. Sure, you can hold up in the House with two guys, but eventually you’ll have 50 zombies banging at your door.

A late addition was the “promotion” system. I wanted to give the player more control, and make them assign importance to some survivors over others. So, for every 9, 18 and 27 hours a survivor lives, they can take a bonus, such as more damage or lower ammo usage.

I almost put in a “goal” system, but the promotion system overtook it. Essentially it would let you trade resources for permanent bonuses, such as a few hours off your rescue time and higher resource generation from certain buildings.

Some other features that didn’t go in due to time were grenades, reclaiming weapons, a morale mechanic and zombie culling as a raid priority. While these would enhance the game, it’s very fun and playable without them. If I could implement one right now into the version here, it’d be zombie culling, as it gives you more control over events in the planning stage.

If you’re having trouble surviving, try only to hold onto one or two buildings during the day, focusing on those that give you resources you’re in need of. At night, don’t spread yourself out if you don’t have to. I’ve found the Mansion and the Warehouse are a good choice, as the former can hold six people (more guns means more downed zombies) and the Warehouse (a constant stream of supplies during the night means you can reinforce your buildings).

Again, if you want the source, drop me an email. Otherwise, give it a go and post your scores!

Don’t forget to check out Wizkill, the game from week 1, and make suggestions for other games I can make.