Ludum Dare 30 | Ludum Dare 23 | Ludum Dare 21 | Ludum Dare 19 |
Ludum Dare 18 |
 That 70's Award Awarded by GreaseMonkey on December 20, 2010 |  Most appropriate use of "retro" in the interpretation of a vague theme Awarded by Spooner on December 18, 2010 |
Charter, our ISP, has decided that we should have intermittent internet that’s only up about 25% of the time. And that this period of flaky internet should coincide exactly with the 48 hour LD window.
Bit of an impediment when developing an online-based game that makes heavy use of an externally hosted server.
Okay, so I finally got my scraping script completely done. It scrapes all of the LD #30 entries, and scrapes the comments from each entry’s page. I have it running as a cron job that grabs comments from a batch of 100 entries every half hour, so assuming we have about 3000 entries for LD #30, it’ll get through all comments once every 15 hours.
Now to actually get something set up in Unity with spheres wrapped in entry screenshots, with comment connection lines going between them.
Okay, I may have been a little overambitious with this concept. I’m making a game-like thingy that scrapes the info for everyone’s LD #30 entries from the website, including comments, and makes a sphere (aka, world) for each entry wrapped in that entry’s screenshot. Then shows connection links between these worlds, based on people’s comments on each other’s entries.
I’m using LD #29 entries for test data. There are about 2500 entries (and I’m sure there’ll be even more for #30). I’m still dealing with the challenge of scraping all that data, and haven’t even started on the actual game app in Unity. The PHP script takes about 15 minutes to get it all, and my hosting provider kills the process after 10 minutes. And since new comments are continuously added to entries, it has to scrape it all again at regular intervals.
So now I’m changing it to do it all in phased intervals, and manage it all in a MySQL database (previously, I was just using an XML file). I’ll probably have it scrape 100 entries every hour (so my hosting provider doesn’t hate me), which, assuming about 3000 entries for LD #30, means it will take up to 15 hours for new comments to make it in (but 7.5 hours on average for any particular comment).
Here’s hoping I’ll have enough time to actually make something in Unity that uses all this data.
Does anyone else find it suspicious that the first 13 (out of 20) themes listed in the final round only make it up to the fifth letter of the alphabet (‘E’)? This seems extremely unlikely!
Clearly, this is lazy, hastily thrown-together FAKE data by administrators abusing their power to hide the fact that “Potato” is actually the highest-voted theme by a wide margin. (And yes, they also faked the results in Round 2.)
The Potato Union demands a public audit of the raw data!
(The Potato Union should not be confused with the Onion Union, a loose collection of language reformists working to standardize the sensible pronunciation of vowels.)
Going to see if I can get some sound in there in the last couple hours. If not, I’ll go with what I have.
I’ve been using Jing for taking screenshots and posting them, and I have to say, I like it. Much quicker than Alt-PrtScrn, pasting into Photoshop, saving as a file, and uploading. With Jing, you just press your hotkey, and it takes a screenshot and uploads it to your Jing account (the account and app are free). Then you can just grab the URL of that image to put in your post.
Have the nested hierarchies of pieces, with targeting lock site thingy, and removing other sibling pieces when you successfully lock on (because it only needs to zoom in on that piece after that).
Got the nested pieces spawning, rendering, spinning, and zooming.
Not sure how this will work out, but here’s what I’m thinking:
An infinite zooming game (yeah, I know; we’re probably going to see a lot of those) which has multiple small, whirling objects on the screen. The camera automatically zooms in, and the player has to match her view rotation and position with that of one of the pieces so she “locks on” as she zooms in to it. The challenge will be that she’s controlling her spinning and panning at the same time, and panning is relative to the current (changing) rotation of her view. As she zooms in to the small piece and matches its rotation, it will of course contain more, smaller pieces, one of which she zooms in to, ad infinitum.
We’ll see how this goes.
Just a little postmortem on my LD #21 entry, Anthony’s Psyche: Escape. (This is mostly a cross-post from my blog.)
This was my third Ludum Dare. In each one, I’ve challenged myself to come up with a different interpretation of the theme than I thought would be typically done. For the Escape theme, I decided to go with the idea of psychological escape mechanisms, or avoiding painful thoughts and memories. This turned out to be a rather artsy, narrative-driven playable story of sorts. This is very different than anything I’ve developed before.
I spent about 27 hours on this entry. Friday night when the theme was announced, I spent three hours in the typical initial panic of trying to come up with an original interpretation of the theme. I settled on the psychological escape mechanisms concept, and that it would have something to do with words on the screen representing thought fragments. I was still unclear about the specifics beyond that.
On Saturday, I spent a couple more hours playing with ideas in my head, and settled on a design. I then spent about ten hours writing code and debugging. It took me much longer than I anticipated to get text with variable alpha per character working in Flashpunk. Probably five hours on that alone. I also spent a few minutes making the “art” for the game (the one stick figure) for a total of 12 hours on Saturday. By this point I had most of the basic functionality of the game working (moving a box of text around the screen and having the words fill in when over the character).
On Sunday, I spent about an hour getting Reason and my keyboard set up, and coming up with the short music loop and “thought complete” riff. I then spent several hours trying to come up with a decent story. I discovered that telling a story through first-person thought fragments is very difficult. When I started entering the text for the thoughts, it just wasn’t coming together. I also discovered some bugs in the way Flash renders text, so I spent a couple hours debugging and working around that. I finally gave up on the story I’d come up with, and about two hours before the deadline, I came up with a very different story that came together pretty quickly. I also wrote some more code for the title screens, ending screen, etc. That made a total of about 12 hours for Sunday.
The end result isn’t exactly a “game”, but I’m satisfied with what I came up with because it’s very different for me, and pushed me in a different direction. I like the overall feeling of the play. I’m thinking of developing something like this a little further.
Like my previous two LD entries, I created things as stand-in content (the stick figure guy, and especially the very short, repetitive music loop) so I had things to write the code around, but they ended up being the final content because I didn’t have time to do “real” art or music. The difference this time was that by now I’ve learned that when I create them, that’ll probably be the case. Ludum Dare is always a great exercise in game development (and a lot of fun) because it forces you to be ruthless in cutting features and calling things “good enough”.
I’m trying something really artsy for this LD. It’s a game about psychological escape mechanisms. A character with a mind tormented by guilt and denial runs around the screen, trying to avoid confronting the persistent thoughts in his head. It’s up to the player to move the “thought” (a fragment of text) over the character so that it fills in the words until all words in the thought are lit, at which point the character has come to terms with this painful thought. The player then moves on to the next thought, thereby piecing together the background story.
Had to modify Flashpunk a bit to be able to change alpha on specific characters in a Text object, which is critical to my game. After a couple hours research, managed to do it with only a couple changes to Flashpunk. Yay, progress!
Not sure if I have it in me right now (been pretty fried), but if the theme grabs me and I get any ideas, I’ll take a shot at it. If so, I’ll most likely be using Flash with Flashpunk.
Good luck everyone!
With each Ludum Dare, I am progressively more blown away by what people can do in a couple days. There are so many cool game concepts. I’m really fried for the first couple days after the compo, so I don’t get on and rate games right away, but just from the little perusing I’ve done, there are a lot of amazing entries. I’ll be back to rate and comment once I can see straight. 
I’ve got some basic playability in Disco?Very! now. You’re the guy on the right. Use left/right arrow keys, combined with shift, to do moves, and follow the other three dudes.
Click the picture to try it out. It’s Flash, so click inside the window once it loads (or keypresses won’t register).
Still have a lot to do (more music, more moves, levels, scoring, etc.), but the basic gameplay function is there.
Got my gameplay concept for my game worked out. The game is called Disco?Very!
Got the first still frame sprite done. I’ll need to make several animation frames for the different dance moves.
I also have part of the music track done. I have it looping correctly in FlashPunk, and confirmed with some tests that I can sync movement to the music. That was something I was concerned about.
4:30 a.m. here, so I should probably think about sleeping.
