Unless you're using physics, which uses up a lot of CPU processing power, the performance gain from not checking AI conditions when there are no AI ships on the screen is marginal. However, for good practice, it's probably best to disable the AI ship behavior group if AI_ships.count = 0 and then enable it if AI_ships.count > 0

I use families for checking. This makes it easy to expand upon. For example, in the future I may add 'allied' ships to the game. Instead of adding condition checks for these allied ships, why not put them all in a family and check if the AI ship is near the family? Same goes for adding different asteroid sprites. Put them in a family and do the condition check on the family.

My first ever prototype made with C2 is called Icarus Wave Attack. You can play it here.

The AI behavior isn't perfect, but you'll notice the AI ships will tend to strafe around the player, occasionally turn to shoot asteroids that are close, and more.

I will give a basic overview of the behavior I created. It uses a combination of custom movement and pathfinding.

There are two major questions:

-Is the AI ship near the player ship?

-Is the AI ship near an asteroid?

The basic case, which is the easiest, is when the AI ship is not near anything. Basically I have a variable called 'closeRange' and if the AI ship is not within closeRange to the player ship AND not within closeRange to an asteroid, the pathfinding behavior is activated and the ship finds a path to the player ship and moves.

Another case is when the AI ship is not near the player ship but near an asteroid. In other words, when distance(AI ship, player ship) > closeRange but distance(AI ship, asteroid) < closeRange, then have the player ship STOP pathfinding and turn on custom movement to move away from the asteroid. I also use the turret behavior to have the AI ship shoot the asteroid.

Another case is when the AI ship is near both asteroid and player ship. Basically this is when both checks in the section above are true. In this case, I have the AI ship prioritize its own survival and accelerate (custom movement) away from BOTH the asteroid and the player ship. Using a couple of condition checks and the turret behavior, the AI ship will fire at the closest object (which will be either the player ship or the asteroid).

At any point in the game, the AI ship is close to the player but not an asteroid, close to an asteroid but not the player, not close to anything, or close to both.

Hope this helps. Note that AI is one of the hardest aspects of game development to accomplish well. If you're just starting out in C2, I recommend starting with just something basic, such as an AI that chases the player but also crashes into asteroids. In development, many things are often more complex than they seem at first, and this is especially true for AI.

Best learn to start backing up then. Good developer habits cannot be replaced by software improvements.

keepee, the original Shockwave game is truly a gem.

There are people who have expressed interest on the Internet for a modern remake of Spybot, and I intend to give it to them. It won't be the same game, but the mechanics will more or less be the same.

Also, here is the first video devlog for this project, which is now called CoHack!

[TUBE]rFFR2P2Q6Zc[/TUBE]

Whiteclaws, I was just about to update!

Yes, the optimization issues are fixed!!! It took a week and a bit of rethinking of how I was approaching the problem to come up with a easier/better method.

There are still issues when letterbox scaling is used, but when scaling is turned off, all the parts seem to connect seamlessly.

<img src="http://i.imgur.com/Bot40kv.png" border="0" />

I've taken a different approach and have thus far solved my issues.

Kyatric, Arima please close this thread.

This may or may not interest you: http://www.scirra.com/forum/optimization-job_topic75254.html

You may or may not be interested in my job here: http://www.scirra.com/forum/optimization-job_topic75254.html

Just curious, how long have you been using C2?

You are confusing innovation with gameplay quality. "Bioshock" and "The last of us" might not be anything new, but they have a fun gameplay, sure, the story helps to avoid noticing how repetitive they are, but that's all. Gameplay is very polished and it'd be fun for Bioshock fans even without the dressing.

So in "Bioshock" and "The Last of Us" the story helps but there must be a gameplay better than just decent.

I agree that in those games the dressing is way above the gameplay, that's why I'd prefer to have them as a TV show, book or a movie instead of a game.

BioShock honestly bored me (and hundreds of others based on Internet searches) because fetch quest followed by fetch quest is not exciting, even with enemies to shoot. To this day, I have yet to find someone who says they can play BioShock for more than a few hours at a time (compare this to Mass Effect series and Deus Ex: Human Revolution, where many people were publicly admitting to no-life-ing the game).

Please don't make claims about the gameplay quality of BioShock unless you have actually played the game. No, BioShock would not be what it is without the dressing. The gameplay quality of BioShock is exactly decent, nothing more. The developers basically banked on story and setting, and they succeeded.

The Last of Us, I will admit, is not as dull as BioShock, but it still isn't anything better than the quality of a Call of Duty game, which for most people is acceptable since CoD is still highly popular.

.D.: Chess doesn't have a story but it's one of the best games ever made. Its core is so good that it doesn't require any dressing at all.

Chess is also a game that is centuries of years old, was required to learn for millions of people of many different positions of power and wealth, and is today being integrated in many schools as mandatory learning.

The success of Chess is only loosely applicable to games we could develop as indies.

Same here - I'm taking a lot of units this semester and it is interfering with my C2 work :(

Dull gameplay becomes acceptable if the story is engrossing. This is proven financially.

BioShock is hailed as one of the greatest FPS games because of its story and engrossing atmosphere. But play the game and you will realize it the gameplay itself adds nothing really new to the FPS genre and that the majority of the game is the player performing 'fetch quests' to gather some items.

The Last of Us, acclaimed by many as possibly the best PS3 shooter to be released this year, has a very engrossing story. However, when it comes to gameplay, there is nothing it adds to the FPS genre (reminds me of Call of Duty), and many of the fights occur in cut-scenes where you have to press some button within a short amount of time in order to win, which is a criticism that is often acknowledged but passed over because the story is just so good.

Tomb Raider also falls into this category, but I will not include it here because the new Tomb Raider was practically a guaranteed sell due to the massive TR fanbase that has been hoping for a relaunch for some time.

There are some exceptions, but even those can be argued against. Minecraft, for example, has no built-in story, but because it is a sandbox game where the player can literally do/build anything then there is a purpose created and re-created by the player himself.

It's personal preference as to whether a story is needed, but in terms of financial success, a great story with average gameplay will pay off more than great gameplay without a purpose.