To Multiplay, Or Not To Multiplay

Currently weighing the value and use of multiplayer in "Deep Blue". Does it make sense? What does it add (if anything)? What does it require? What does it take away? Is there a "gotta have" reason for it?

Ever since early on in the play testing process Connor made the request for multiplayer. However, since this was my first shot into XNA and C# the best I've been shooting for at this point was couch play with two player co-op in two MANTA subs. Couch play seemed only logical since, barring some incredible miracle, I don't see "Deep Blue" having enough concurrent game plays to support a XBox Live multiplayer mode.

But the domino effect of including simple couch play will be felt throughout the game. How do we allow two players in the water and still have a balanced game that isn't twice as easy with two people? Then there's the whole process of registering two controllers and possibly two XBox Live accounts. Some of the UI elements (like showing the current depth) become confusing with two players.

At this point I think I'd rather toss multiplayer this time around and possibly use that time to create an awards/achievement system and more story elements.

I guess I'll run it past the game design boss, Connor, and see if he'll let it go. I just don't want to add multiplayer if it doesn't truely add something to the game play. Having it just to have it, does not increase the fun.

October Video Update

Here's a quick look* at how things stand. Obviously the new collision detection isn't in and I had made a noob mistake while rendering the latest rammer enemy which left the annoying light halo around the outter edge of the ship.

See the few previous posts for a short breakdown of what is new and/or changed.

Anyhow, progress is being made. More to come.

(* Facebook folks may have to follow the link: http://deepbluexna.blogspot.com )

Connor the Game Designer

Now that I've got a level system in place for "Deep Blue" I introduced Connor to the level editor, basically just a .txt file inside Studio Express. I laid out the rules for him of what each level needed to be valid and showed him how to do a build so he could test. So I put him in the role of both level designer and head of Q.A.

And here's Connor working on very intense level design.

....followed only moments later by Connor explaining to me all the things that were wrong.

Yes..."Deep Blue" still has a long way to go.

He's ALIVE!!!!

Sorry, I couldn't resist. I'm referring to the enemy ships that are now animated (swimming). I used the "Sample Platformer" game that came with Studio Express as a base for the sprite animation system.

Even better, using the level system I've got enemy placement in the level with multiple enemy types, one animated and one non-animated. So now I guess I'd better get to work on working in a few more enemy types to play with.

I've got Connor working hard on level design, he's working in Studio Express on a .txt file for enemy placement. He's going through all the create/test/alter cycle to get things the way he wants them.

Hopefully I'll get a video up soon to show progress.

Deep Blue 2.0

Things are moving pretty well with the restructuring. Here's a list of some of the changes, both big and small.

- (new) Logo screen with an image that may be swapped with a video before launched.
- (new functionality) Pop up windows ("How to Play" and "Credits") support multiple pages and an "open/close" state.
- (change) Projectiles are now stored in a List instead of an array to support infinite shots and stop the rubberbanding on the shotgun weapon.
- (new) Level object created, to support actual level design over the random spawn style game. Levels are designed and read in from a .txt file.
- (change) Enemies are now stored in a List instead of an array to support a variable number of enemies per level, as part of the level design change.
- (change) various cleanups of methods and classes.
- (change) and the big change was that the whole system is built in a modular fashion, instead of 90% of the code being found in one file. Individual classes for game screens, pickups, enemies, player, weapons, UI...etc. This was the entire reason for the teardown, so that the code could much more easily be modified for future games. It's also much easier to read through.

Still have a ways to go, but things are headed in the right direction now.

Making Progress

Mr. Montgomery's solution seems to be working. After commenting code and excluding classes throughout the project "Deep Blue" is up and running again. I've got a logo screen, splash screen, main menu, game over screen, and the beginnings of the main game screen complete with controllable MANTA ship.

The work ahead still looks pretty daunting if I step back and look at the whole thing in it's entirety, but plugging things in one little step at a time hasn't been too bad so far.

I just hope all the extra time it's cost me on "Deep Blue" will be worth it (both for the completion of Deep Blue and for the next game...themed around the same engine).

Back at it, with a new strategy

Ok, so although I've had a rush of other things that have popped up...like Cub Scouts and Scribblenauts...I have to say I've been a little discouraged with "Deep Blue". I had the goal of breaking the game and restructuring it, then putting it back together. Well, I managed to do the first part reeeeaaaally well. Maybe a little too well.

I've had trouble finding a way to approach "fixing" the code. I begun following the flow of the game from the beginning and fixing things that were broken along the way, but with the whole thing a wreck it made it impossible to compile and thus, test.

Then I tried looking at the debug code and fixing things one class at a time, beginning with the class with the least amount of errors. I found it proved very hard to tell if I was making progress or not. Again, without the ability to compile, it's hard to get a good feel for progress.

I THINK the restructuring was going pretty well. It definately cleaned up the main game class a ton, but I had no way to test.

So I spoke with a couple of friends at work skilled in programming and Mr. Jim Montgomery (honestly one of the most impressive sounding names at work, you really have to say the whole thing) suggested commenting virtually everything out in the main game class, just to get the game to compile and slowly turn things on one piece at a time.

So that's was the new plan. Even if I had to comment it down to a blue screen, I was going to get it to run again.

After my initial tests, things seem a little better than the worst case senario...after a ton of commenting I've returned to a logo screen, to a splash screen, to a menu screen (and the menu still works). And now begins the slow process of turning things back on and fixing as I go.