And also a new gameplay video
<a href="http://www.youtube.com/v/3k9b7pAlnkE?version=3&amp;hl=en_US&amp;start=" target="_blank">http://www.youtube.com/v/3k9b7pAlnkE?version=3&amp;hl=en_US&amp;start=</a>

I'm working on the background...
I'm not sure about this stuff. The shapes are parallax layers and rotate similarly.

I agree with you. I try to write the simplest code possible without sacrificing speed.
But writing a new List implementation for an Updatable List require time, and I'm probably not going to get it perfectly at the first try. You may be able to do it on the fly, but that's not my case. That's extra work I could put on something with a higher priority.

I write simple code. Simple code is easy. Why should I want to make it more complex ? Dumb code = good code.
As long as dumb doesn't mean slow.

Because the point IS code quality, and because I haven't run on speed problems by now ?
If you really have to optimize speed, you should do your own test and check where the bottlenecks are. Because the data structure for your updatable list is probably not the cause: nobody gain 20 FPS by switching his LinkedList with two ArrayList.
But sure, it probably help if you want to run your game on a lower end PC.

As a matter of fact I did my own benchmark on this matter. Most of the time, the gain between using an ArrayList and a LinkedList is negligible. Even with a simple iteration (296 ArrayList / 315 LinkedList), which I just tested right now.
And by negligible I mean negligible on a modern PC, don't get me wrong.

What I haven't benchmarked yet is the potential speed gain of using a LinkedList to remove objects during the iteration.
I think the gain is so negligible I should'nt bother with it at all. Athough that's just a theory, as I said I should write a benchmark to be REALLY sure about it.

Clean code should be the main priority. The first rule of optimisation is "don't optimise if you don't have to".
But if you can optimise on the fly (using a LinkedList for an "Updatable" List for instance) without writing more code, what's the point of not doing it ?

I used a LinkedList for my Updatable List because that's probably faster than a single ArrayList (I remove a lot of Updatable entities and the list can get pretty heavy at times).
Two ArrayList are probably faster than a single LinkedList. But my code run fairly fast for the moment, why would I should optimise it now ? Premature optimisation may cause bugs, and in this case lead to ugly/verbose code.
I don't think data structures are really that important. In this exemple, how many FPS do you think we could gain by using two ArrayList ? Do we REALLY need to write the code (potentially more bugs and more loc) with two ArrayList ?
Though I'm not sure about this, I should write a benchmark.

You have to figure out where the real speed bottlenecks are. If you are writing a particle engine, chance are you could double the FPS by sorting the Renderable List using Textures to avoid unnecessary bindings.


You could write your own ArrayList implementation with a faster O(1) remove method (switch with the last element and decrement the size). In my case I don't care about the order.
Also I heard somewhere that using a for loop with an int and the get method is A LOT faster than using the iterator of an ArrayList. I didn't tested it but it could help some people here.

Yeah... That's definitely one of the thing I would have done differently if I knew about it sooner.

My main source of inspiration is the film "From Beyond", released in 1986. You can watch the trailer here: http://www.youtube.com/watch?v=zgZuCXY-vR8
That's where the pink palette come from I wanted to recreate that weird "feel" from the movie.

I don't want to make the numbers transparents (apart from the fade out) because translucent text doesn't seems like a good idea to me. I'll try it though !
Maybe some kind of "4*24" instead of four "+24"... I'll think about it.

To decode animated PNG !
Those will be used for animated sprites/decals, placed with the map editor (night moths around light, fireplace lighting, etc).



I'll try to make a short video to show visual effects I made lately. Mainly metaballs and refraction shader.

Thanks for the positive remarks !

What do you mean "planning" ?
Also I'm probably going to change the black background for a proper background, with geometrical shapes and parallax scrolling.

You make a good point here. It doesn't bother me ingame, but I'll try do do something.

LWJGL (OpenGL/OpenAL, Util), Slick-Util, JInput, JOgg/JOrbis and JavaPNG
I'll upload a screenshot, thanks

Hi !
I'm French so my english may sound bad, sorry about that

I'm currently making this game called Pink Point, since 1 year and a half.
It's a 2D scoring action game, playable in the browser (LWJGL applet).

The gameplay itself is pretty simple. It's a top down shooter, you have a weapon and you can cast spells. You kill monsters to increase the score multiplier. When monsters die, they drop a certain number of gems, which you have to take to gain score points. The game is playable with 1 player, or 2 players on splitscreen (coop or versus).
I'm trying to make this game as fluid and dynamic as possible.

I don't feel comfortable with releasing the game yet because I still have a lot of work to do.
Be sure you'll be the first informed about the beta though.
Here is a short gameplay video:

<a href="http://www.youtube.com/v/LQBzc73IFMI?version=3&amp;hl=en_US&amp;start=" target="_blank">http://www.youtube.com/v/LQBzc73IFMI?version=3&amp;hl=en_US&amp;start=</a>

And a screenshot 



A lof of sounds are placeholder and are taken from various games.
Feel free to tell me what you think !