I am doing a platformer game this time, I thought it should be easy because i saw a lot of similiar examples. But I was wrong, I did not check their code, just thought it is easy. And now, I feel tired out, nealy two days passed, I finally get my player walk, crouch, jump in my own way. Even though it is not flexible operations, at least it works. English is still my problems, I am worried about that other people can hardly get the story I am telling, tell a story is not easy, especiall tell a game story. Also technical problems, 1) I want to creat anti-gravity workd,(it should SHOULD be easy, that is my main part). 2) I want add save and use json module, (still, I never used before, I just know other people use it). 3) I want add an replay, the computer memory the event and time, and then post it, this would be a innovation, however, I do not know I can do it. OK, forget all about that, I should go to sleep first, tomorrow is another day.

Second day. The initial pandemonium of where to start is past. Yesterday I started on a level designer, some graphics and the basic mechanics. I might soon have something that will be worth uploading. The basic idea of the game I settled on is that you are a reclusive man trying to keep something out of your house with your swords! Naturally the thing you are trying to keep out is a clown :-) This will be a fairly straightforward plat-former puzzle game with maybe a time element.

Spent most of the day today working on the core engine. The game is about infiltration and so a key part of that is the enemy AI. That took most of the day. I had a behaviour tree implementation but I couldn’t get my brain around it and had to abandon it in favour of a hacked up monster function!

I have that working quite well now and so that should be the bulk of the technical complexity behind me (hopefully!)

Tomorrow will be working on the full gameplay. I still just have placeholder graphics in there for the moment and so I might start to work on those tomorrow also.

Mostly because I believe in documenting my process, I took the three tiny ideas and attempted to flesh them out.

The political statement game came off because I couldn't build a decent proposal document.

So the other two:

Security Maze

Play the other side of a stealth game: You're the bumbling guard. Early on, it's just you and your wits against a hidden enemy. But then you start getting tools that will improve your ability to track and catch your prey. Or simply keep him out until he runs out of time.

Primary Game Mechanics:

• Limited vision. Only see where you have eyes, right in front of you, or in range of any cameras you place.
• Limited mobility. Your prey is faster than you, which means it's only your planning that will keep them out.
• Security gear. Noise traps, cameras, barriers and more to track the intruder, slow it down, and help you stop them.

Symbiotes

Beat monstrous warriors and creatures, collect their battle spirits and combine their best qualities into yourself. Become the perfect warrior: the symbiote.

An idea that borrows from Shadows of the Colossus, Titan Souls, and Megaman and combines them.

• Symbiotes and equipment: Equip your warrior with symbiotic spirits and conventional equipment to take down bigger and badder beasts.
• The Beasts: Each encounter is a specially designed beast with unique attack patterns and skills. You'll need to master the combination of skills the symbiotes provide.
• Shops: The symbiote is a terrifying creature, and depending on the spirits you're most attuned to, certain citizens will keep you from their businesses. Good luck getting rest when the inn stops letting you in.

I have some basic game play done. If anyone would like to try it, I'd like to hear any preliminary feedback.

I found a nice looking sprite sheet over on open game art.org, so I wired up player walking animations. I also made a simple map using Tiled. There is also a simple scene manager, which lets you move from the title screen to the first scene. Everything is pretty crude, but I find that I'm making a ton of progress due to the constricted time frame. I've rewritten a few standard processors that I've written many times before in other games, and was happy to realise that I was finding new tricks now that I'm not obsessing over the code too much.

I feel that the theme for this pyweek22 is sufficiently imprecise to allow me to use an idea that I wanted to put in a game for a long time. I would like to code my own procedural terrain generator, and I think that "you can't let him in here" is quite a nice theme for an exploration game. There is a guy. There is a big, big, big map. You have to explore the map in order to find the guy. Because you can't let him in here.
I coded the whole night to obtain a satisfactory result, but I'm happy now. I will add screeshots tomorrow if I'm less tired than now. I hope that I can finish to finish to implement the kernel of my game by tomorrow night, so that I can spend the rest of the week in appending content, debugging, gui-coding, and tuning graphics.

And they're off!

After some lengthly discussion we settled on a tower defense game. We suspect that given the theme -- "You can't let him in here" -- there are going to be a number of tower defense games, but we hope ours is different enough to stand out a little and be fun.

Envisioned core mechanics:

• During the day you prepare to defend you space turnips by planting them in safe places and surrounding them with a variety of lights.
• During the night, the space mould attacks and you have to defend your turnips by turning lights on and off.
• However, the station only has enough spare power for a few lights, so you'll need to be careful which ones you switch on and try not to overload the batteries.
• And the space mould might mutate or become immune to lights of different colours.
• Once you've produced enough turnips, you get to leave the station and be assigned to a more difficult station as reward for your success!

Progress today

• We have a lot of core pieces of art, including turnips and some mould.
• We have some generic pygame and pymunk infrastructure and a ray tracing engine that generates a set of convex hulls that describe the regions light from lamps can reach.

We've largely ignored the masculine, human, singular "him" in the theme -- you can argue with the sentient space mould who is stealing the station's turnips if you would like to use a different pronoun.

So, "You can't let him in here." I am so not happy to see this one come up, but I'll work with it.

So top level ideas: you're limiting a specific person or thing or class of things from entry. Either something about the player or a companion doesn't allow you into a place, or you're tasked with keeping others out.

I've got a couple of first pass ideas on how to take this:

I could make a political statement: I could figure out mechanics that help model the emotional resonance of being a trans woman in a world where "bathroom bills" are becoming daily news. This would be a game more about identity than game place, and I'd be limited to a short prototype rather than a functioning game. Set up mechanics that let the player choose the identity they'll be playing, then slowly clamp down on their expectations.

The second idea is kind of like the other, except someone you keep with you limits what areas you're allowed in. This could be a tiny action game? Maybe use a symbiosis mechanic.

Last idea is building a reverse stealth game with something similar to tower defense mechanics. A lot harder in 2D, but possible. Makes vision management fun, though.

I'll start drawing things up and make a decision by the end of the night.

I was going to create a dual-stick shooter in a dark cave where the only light is the light from projectiles. I got a functionally working prototype [youtube link] wrt lighting effects. Unfortunately, despite the smooth appearance, it's actually dropping FPS and there's only a few bullets on the screen and so it's not particularly viable. It seems the pygame blend modes are great for either one-time-then-cached image modifications or blitting small areas. But having a "shadow layer" that you blit light away from with gradient images...well...I'm not sure what I expected, performance-wise.

I'm going to go down to the next item on my ideas list and start over today.