First pyweek for me.

Decided to do the smart thing and use a lot of software I've never used before...

Planning on using (also learning) pyglet and possibly kytten.

Idea is to design a 3d game with 3d models made from blender and exported in the obj format (googling for simple formats yield this one).

First thing first. Learn more opengl...

Estimated chance of having a finished game: 0.1%

Decided on game genre - never tried making a tower defense game and mutate seems like it might make for an interesting twist to the genre. Well more "evolve/mutate" really.

The idea is that the player will be able to build different types of towers, (e.g. fire, ice, lightning, piercing, blunt) and the monsters who are blithely following their path will be faced with the existential crisis of either dying or evolving so they can deal better against the onslaught brought from the towers. In turn the player will be able to, err, mutate? his towers to deal with the mutating monsters. So sort of a game and cat type thing.

This sounds ambitious enough to keep me procrastinating from the homework due Monday.

Chance of finishing: 0.2%

(Or I could go for a Grow type game with a mutating a tree or some creatures - but that sounds crazy difficult, a.k.a., requiring at least an ounce of artistic ability.)

Yay, a crappy title screen has been added. Well, it's mostly an exact copy from the kytten example provided, but it's a start.

Kytten seems pretty cool except there doesn't seem to be any documentation outside of some docstrings. Absolutely love that google code (repo for kytten) provides nice search capabilities. 

Started switching to git from svn two days ago (seems to be the trendy thing, and who am I to oppose a trend). Seems pretty cool so far.

Also switched from vim to python tools for visual studio today. Have yet to figure out how to get intellisense working.

Onto coming up with some simple art for towers and monsters and monster towers and tower monsters.

Chance of completion: 0.3%

Spamming another diary entry - it seems to help me keep focused.

Finished the first mesh- possibly an archer tower (writing this makes me realize that there would also need to be some sort of tower attack animation which I hadn't considered). I have only a foggy idea of how textures work and practically zero experience texturing meshes, so that will be postponed.

Next step is to get the mesh to display properly using pyglet.

Chance of a playable product: 0.4%

Every time I start an opengl project, I consider it a milestone of sorts when I get the screen to display something other than black.

Stole (well, I did give credit) pygame's wavefront object loader. Figured time could be better spent elsewhere.

The tower mesh I made earlier displays properly. Now I need to figure out how to implement lighting. Spent a few hours reading the red book last week in preparation for this, and it seems doable considering the 3D normals get calculated from blender (i.e. won't have to figure out how to find normals).

Rather than keep working through the night, I'll take the prudent choice of an easy homework (watching Broken Blossoms for film class) followed by sleep.

Chance of game completion: 0.5%

I decided that instead of doing a traditional tower defense game where the player builds the towers and the monsters evolve to survive, it would be more fitting with the theme, and perhaps more fun, to flip the roles and let the player be on the monster's side with the decision of how to mutate them best for survival. This is a bit similar to Villainous (http://www.kongregate.com/games/Rete/villainous) but hopefully more fun (if a thousand times uglier).

Other thoughts:
Make the player be able to lose. More precisely, don't let the player redistribute mutations eventually leading to the possibility of him not being able to beat a level. From a player perspective that sounds terrible, so who knows.

Also an odd bug/problem I just ran into. Ran my program this morning and was getting 8 fps on the title screen, down from the 800+ the night before. Turns out silverlight (netflix) was somehow doing something to... idk what, can't be opengl or python, so idk, but killing it fixed the problem. Odd.

Chance of completion: 0.4%

 

Got a menu bug someone fixed... sort of. Spent maybe 2 hours on something which should have been really simple - and it still doesn't work properly. Slightly regretting using kytten as there is no documentation and it's not widely used.

Implemented a camera with keyboard movement. Yay, first interactive element.

Implemented lighting, something I had never done before. I've yet to figure out how to make it look like I want it to, but it works.

Mesh texturing will be a big problem to deal with code wise and a bigger problem art wise. Note to self: befriend an artist or two at some point.

Next, model a sample monster.

Edit: chance of finishing: 0.8%

Won't have nearly as much time and energy to dedicate tomorrow on this but I'll be optimistic and outline some goals:

1) Get texturing figured-out/working. This may take anywhere from an hour to who knows how long.
2) Get a monster to move in-game. Since I'm not doing any fancy animations, movement will be an alternating rotation on the y-axis + forward translation. Hopefully it doesn't look like utter crap.


As far as upgrades available to the player (i.e. mutations) I've decided to keep it somewhat simple (I think):

The upgrade tree will be a perfect binary tree and the player will be able to make up his monster army by choosing which mutation path to take:

Legs: runs on the ground
    Longer-legs: +speed
Stronger-legs: +health, +speed, +dash(maybe)
Bigger-size: +health, -speed, causes towers to attack him (from fear)
    Arms: +health
Arm-shields: +health, +block%
Healer: (you need arms to heal, right?) heals nearby units
Wings: flight
    Figure this out later


Also need to figure out which towers will be good against which monster.

To keep the game interactive outside of choosing mutations and monster composition, there will be a timing element of when does the player want to send in a monster. Might add tower cool-down times to increase skill depth of timing.

 Bluh, so much to do.

Chance of some sort of playable product: 8%