Aside from some further tweaking, Cloudburst is officially done. And with two weekends to go! And, yes, there are pics in here.
So, let's start off with a quick pic.
Ah, yes. Here we go. The doubles are working correctly, and everything was going swell. I had the first three levels complete, and decided to go for four. I had always imagined it would be a four or five level game.
Well, it ends at four.
I stepped up the timers so that they generate clouds nearly constantly. One random timer goes off between .75 seconds (remember, to keep the clouds from colliding with their own clones, I needed at least .65 seconds between appearances) and 2.5 seconds. The other runs between .75 seconds and 2 seconds. I put them in and started it up to see how it was working.
It was then that I realized just how cruel the game actually was.
Keep in mind, while testing it, I was doing pretty good, but didn't think it was all that difficult. Until a couple of beta testers gave it a whirl and bombed out in the second or shortly into the third stage. The fourth level, however, turns all the elements into a real battle for players to manage the screen.
Keep in mind the following design element: once you shoot a cloud once, shooting it again just wastes water, you don't get it back. My vision was that, as more clouds appeared, players would be more likely to squeeze off a well-placed shot at a cloud high on the screen. As it fell, it would take others with it, clearing the way, if you will, and replenishing all the water contained in the clouds. There was also the fact that I had fixed the collision detection so that, if the player was standing directly over a cloud when it generated, the player's shot wouldn't just pass straight through it.
That turned out to be a problem.
In level 4, there are so many clouds being generated that there was a good chance that you would line up a shot, time it just so the combo would take down five or more clouds at once and free up 2/3 of the screen, and just as you fired, another cloud would generate, taking the shot. Your original would drift offscreen, and you would lose a few other clouds as they drifted away. Probably the biggest sin a designer can make with games is not keeping the player in control. Few things will suck them out of the game faster and turn a fun experience into a frustrating one. I decided to use the collision detection issues to the player's advantage, and reset everything. Now, you might miss a cloud if you're quick on the trigger in the early rounds, but in later rounds, you appreciate it.
This was so complex, the nature of the game changed. Now, I was "juggling". When my remaining water got low, I would let clouds develop until almost all my water was gone. I'd then shoot the one at the top, setting off a chain reaction as the clouds collided and returned with their water. While they were up, with no water in the pit, no new clouds would generate, giving me some breathing room. Of course, one miss or if the clouds generated before I could line them up again, and I was screwed.
I could just barely get past level 4. Usually, I would have between 1 and 3 units of water left, based on the math the bonus counter uses. A fifth round became out of the question. This is a survival shooter. One more level with faster generation, the obvious way to go, would make it impossible.
So, I realized the game was nearing the finish line and began working on the closing screen. Given how rough it was, I didn't just want a simple "YOU'RE WINNER!" image, I felt some sort of credit sequence, however meager, was more appropriate. I'll be redoing the instruction screen at the start of the game to remove the credits from there. They are now at the end. I drew up a background, complete with a dry river that eventually fills as the credits roll. I really should have given more consideration to the end sequence when I was designing this, but I wasn't sure it would have an ending. I had though about just letting it keep getting harder until the player lost like in the old Atari games.
Before long, it was finished. All I had to do now was dial in the specifics. As I mentioned, I already fixed the collision detection so that it was faulty again (I know, but trust me, that phrase makes sense to me). I also gave the sprite, who I've started referring to as "Misty", a tiny extra bit of speed. Before, clouds drifting away before you could chase them down were pretty common. The little bit of speed, plus the fix to the collision detection, makes completing the game far more doable.
After that, it was putting in the reset so the game can restart without having to exit to the OS, removing the last variable display counter from the upper left corner, selecting the game over music, things like that. I'll be testing this out during the week, adjusting point totals and speeds and so on. Maybe make faster generators on the last level? Maybe. Have to see if it makes the game impossible.
Victory pose, as you can see above, and game over poses are done as well. When the game ends because you lost all your water, Misty falls to the ground crying, as a "WAH WAH WAH" plays and the sound of a toilet flushing is heard. If you make it to the end, you hear the "Hallelujah!" from Handel's Messiah and Misty strikes the pose above. I have compiled the build, the game is officially gamma, and with luck, will be a release candidate within two weeks. On Genny, my dev computer, you can see the new program icon here:
Nope. No self-congratulatory hoagie yet. I'm now catching up on the mountain of things I put off while I worked like crazy to get this done. Comics to organize, vacuuming to do, comics to draw, lots of stuff. Besides, good as this feels, the hoagie will be if I actually get this sold at the ChicagoCon in three weeks. Keep them fingers crossed for me
By the way, I spellchecked this before I posted it. I had spelt "vacuuming" as "vacccuuming". LJ's spellchecker, under suggested correct spellings, has the word "backgammon." WTF?!?