Run/Jump/Flip - http://www.youtube.com/watch?v=vqDDmStOVAc

Awesome, thanks a lot!

Are we talking about hosted solutions (e.g. Google Code) or applications you install yourself on your own server or both?

I've had good experiences with Redmine, although it was a pain to set up.

Rock Paper Shotgun is organising a meetup on Friday night.

No idea, that looks fine to me. My code is:

Code:

var p:Number = (this.loaderInfo.bytesLoaded / this.loaderInfo.bytesTotal);
text.text = int(p * 100) + "%";

which should be essentially the same thing.

http://www.youtube.com/watch?v=6jzcAKmodRA

I took a break from animating and am arting up the next level, video this weekend of course.

I love everything about how this looks but hate what the gameplay looks like. That really doesn't seem like it would be fun to play, and if that's the case then it would be a terrible terrible waste.

stage is null until the object is added to the display list. Since you create the Main instance before adding it, you can't access the stage property in the constructor.

Add an ADDED_TO_STAGE event listener:

Code:

addEventListener(Event.ADDED_TO_STAGE, onAddedToStage);

And then you can add your key handlers in that function.

This is a really cool thing you're doing here, great job.

I'd love it if you wanted to do a video of one of my games. Either Mindbender or Run/Jump/Flip or Terrible Tiny Traps maybe: whichever one interests you the most.

Made Ten levels.. Tell me what you guys think.

http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAAAAAAAAAAAAAAAAAAAAAABBAAAAAAABACBAAAAAABDABAAAAAABAABAAAAABEAABAAAAAABBBAAAAAAAAAAAAAAAAAAAAAAAA
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAAAAAAAAAAAAAAAAAAAABBBBBBAAABEAAADABAABEAAAAABAAABADABBBAAABAAAACBAAAABBBBBAAAAAAAAAAAAAAAAAAAAAA
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAAAAAAAAAAAAABBBAAAAAABEEEBAAAABAAADBAAAABADAABAAAABABDBAAAAABAAAABAAAABAAACBAAAAABBBBAAAAAAAAAAAA
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAAAAAAAAAAAAAAAAAAAAABBBBBAAAABEEEDABAAABEBBAABAAABAAADABAAABAADABAAAAABBAAABBAAAABADACBAAAABAABBB
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAAAAAAAAAAAAAAAAAAAAAABBBBAAAAABAEEEBAAAABAABBAAAABBDDBAAAABAAADABAAABAAAACBAAAABBBBBAAAAAAAAAAAAA
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAAAAAAAAAAAAAAAAAAAABBBBBBAAABAAABAEBAABADADAEBAABAADADEBAABCBAAAEBAAABBBBBBAAAAAAAAAAAAAAAAAAAAAA
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AABAAAAABBABCBAABBAAABABABEEAAABABABEEAAABABBBBBDBABAAAAADABABABDBABABABAAAADAABABAAAAAAABAABBBBBBBB
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAAAAAAAAAABBBAAAAAABAAABBBBAABCDDBBAABABAADAAAABAABBBBAAABAAAAABABBAAAAABBAABAAAABEEEABAAAAABBBBAA
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAABBBAAAAAABEEEBAAABBBEEEBAAABAAADAABAABADDADABAABCABBAABAABBBADDABAAAABAAAABAAAABAAABAAAAABBBBAAA
http://jonathanwhiting.com/coding/ldw/sokoban.swf?level=AAAABBBBAAAABBBAABBAAABADCEEBAAABABAEEBAAABABDBBBAAABABAAABAAABAAADABAAABBADAABAAAABAABBBAAAABBBBAAA

My thoughts:
Level 2 isn't as compact as it could be (which was one of the stated aims for the sokoban challenge).
Level 4 is less challenging than level 3.
Levels 7 and 8 can be done by moving one block at a time which makes for an uninteresting puzzle.
That throws the difficulty curve a bit so it's an unusual shape.

Don't have time to study in detail right now, but general text-dump first impression..  It felt very difficult, I was drawn over towards the bottom right, and then got to a state that I couldn't progress past (big moving platform, followed by small moving platform and killer 'minefield').  Part of the problem was I found the jump felt very unnatural, but I wonder if the level design wasn't a bit unforgiving too.  I suspect some of the other equally reachable areas were probably easier, but I wasn't drawn to them first.  The relative accessability of areas may therefore be worth considering (make sure they nail the basics before they can get at the advanced stuff).

With such a small space you don't have much scope to let the player mess around and learn the jumping feel etc. but I'd have certainly welcomed a bit more of an intro.

The bottom right was in fact the only part of the level you can get to initially: the rest of the level is locked until you've been there first. But it's not significantly easier than the rest of the level so your points still stand. I've tweaked that area slightly: nothing major but hopefully it will now be a less frustrating experience.

The difficulty of the whole level is in general fairly difficult: I found that with the pixel-resolution positioning there weren't a whole lot of interesting challenges between too easy and too hard, and (as I generally do) I leaned towards being too hard.

  I thought there was only one screen because trying to go up got the player stuck on an invisible ceiling. Didn't even try going down... (my theory for why is a combination of 1) the starting level having a floor and 2) the game mechanics mean you always want to be going up)

On a tangential note:
I think this is within the scope of this thread? I made a one-level platformer recently: Terrible Tiny Traps. I'd really appreciate any level design criticism that anyone has.

Being able to move the player in edit mode is more confusing than anything, it's not really any more useful here than it would have been for the previous platforming challenges.

The enemies only show up when you load from a level string: if you just add them from the editor they don't appear in play mode.

Here's my attempt. Not sure it does a great job at teaching anything, more it just has puzzles that use the mechanics one by one.

I agree, I don't think this is the simple solution you're hoping for.

My own solution would be to model it as the curved world that it is.  Store velocity and position for the player as normal and instead of applying a standard downward gravity force, I'd apply a force that pointed from the player's current position toward the middle of whatever planet she's closest to.

I'm not familiar with what physics stuff Flixel can cope with.  The main feature you'd need to implement my method would be testing for collisions between rotated rectangles.

This is what I would do, although I would probably make the collision shapes for the players circular to make testing collisions easier.

If you're using Flex 4 to compile, they broke font embedding. You want:

Code:

[Embed(source = '../bin/44v2.ttf', fontFamily = "n_font", embedAsCFF="false")]

Or you can tell it to use the Flex 3 behaviour with the compiler parameter -compatibility-version=3.0.0

Solution to level 6 is different to the Ludum Dare version? I can't seem to do it here (or I've forgotten what the solution was).

It'd be nice to have completion indicators on the level select screen.

You can release the source code for everything except the high-score component. Or for everything except a salt value for the high-score component.

For certain games, you might be able to send some gameplay data with the score to verify the playthrough on the server.

Is the player meant to be able to move in edit mode?

Sidenote: I'm getting that damn error 2046 again and have spent most of the afternoon trying to find a workaround for it. It seems to be Linux only, which is why I'm the only one who's brought it up here. Fortunately I've just discovered that Google Chrome comes with a different version of Flash Player which mysteriously works.

The thing is, you never really "leave" your ideas.  You can put them aside for a while, but pretty much any project you start, even if you don't finish, will stick around in your head and make later projects better, either because you can take parts of the old one that you liked and work them in, or because you learned a terrible lesson in what not to do.

I agree with this completely.

I'm really bad at finishing games off, but I'm constantly finding that I can take old unfinished projects and turn them into something totally different, that I wouldn't have thought of if I hadn't made the original thing first.

I felt like the score tokens disappeared very quickly. Maybe give them a longer life, or make the attraction distance larger, or both.

Once after I died a few times, it went to a blank grey screen. Can't reproduce it again though.

I'd really appreciate some way of being able to cancel the teleport after you've pressed X, or for it to cycle between minimum and maximum aim distance. Currently there's no way to re-aim if you hold it down too long and end up aiming too far away.

I don't think having lots of lists is necessarily a problem, as long as you've thought about how and where you're going to access/modify those lists. E.g. having an add() and a remove() method which decide which lists to edit rather than manually adding items to each relevant list.

I think the approach you've gone with makes as much sense as anything: insert new objects at the correct index and use bubble sort for frame-to-frame consistency. If you're having problems with this being unwieldy then you probably need to refactor it.

My solution would probably be to do it in a bitmap-based engine like FlashPunk or Flixel where you render objects in the traditional manner (blitting to a screen buffer). Then you know that after sorting your objects you can just iterate over them and be sure they'll be rendered in the right order. Of course, that would involve rewriting your game so isn't massively helpful to you right now.

I am almost definitely coming up for this, will be good to see you all.