Mercury: The Simple 2D Game Library | >> BETA coming soon <<

[saucymeatman]

Although I don't think people should have to defend a threads existence (it ruins the thread), I come to check out this topic very often. Not sure how much I would come to this site really if it wasn't for this thread. Its interesting to me to watch this library develop. Why would it matter if others have already created better libraries? Better programmers than myself have made games of every genre before me, and it would be outrageous if I hoped to be more successful than them. And yet, we all still program games. And if anyone posted something discouraging the creation of a game because someone before them has done it better, people would be very upset. I don't see why its necessary to discourage creativity when people are going to make a tool for game development, its a very sad thing.

EDIT :
Dammit Kpars, you posted the same thing right before me XD

[atombrot]

Don't want to spill more oil in the fire, but I have some thoughts that I'd like to leave here.

I often just go to the "Recent posts since epoch" page and check out most of the updated threads. Because of this I have also followed your thread a bit.

It's nice that you have found something to do and that you find the motivation to keep working on it. That is something many people fail at (I have started enough projects myself, that I have abbondened after a while).

But I have to agree with HeroesGraveDev.

The signal to noise ratio on this project is just too low to be taken seriously. Keep that stuff on your IRC channel (*shudders*).

People might be interested in your project, but the amount of noise in this thread is quite high. You use this thread as a chat (or at least it seems to me to be like that). If I'm interested in this project, I have to go through 8 pages of spam to find some relevant bits.
If you want to appeal to people and provoke more (positive) reactions, maybe you should act a bit more professional. Update your opening post and add more information to it, so a potential user does not have to go through 9 pages to find out what all is about. write some more about where you are coming from (why do you do this) and where you want to go (what you want to achieve). Also most importantly, what can you already do? I look at the opening post and I cannot really say, were this project stands. what features are planned and what features are already there. put a changelog in there, so users see when new releases of your project are ready (at some point a homepage or wiki for this stuff might be a good place).

then also, you make a game library and currently the only image in the opening post is your logo. nobody cares about your logo at this stage. it's nice that you have one, but please, show some screenshots of your product in action

What would also help enormously would be to keep the development and offtopic chatter out of this thread. Have a forum, setup an irc channel, use an issue tracker, make new threads if you have a technical problem, but don't spam this thread. Others might not be interested in theese things. Keep the thread for posting new updates or for answering questions (like the one from jonjava). I tend to remember noisy posts and don't really look all too often into them, when I see there is an update.
 
And don't be easily offended on the internet Even if jonjava did not have any bad intentions with asking his question (I don't think he had), there are other people on the internet who just want to watch the world burn if you go on a public forum with your project, you're gonna be critizized. learn to deal with it. his question would have been a nice queue to drop some words about your motivation and wessles has given a nice answer to it, but the other responses were quite immature (and this reflects on your project).

So I for my part would like the see you take off with your library, make good progress and play some nice games made with it, but if you don't change your posting behavior here, I don't think I will be seeing it between all the noise

I hope no offense is taken, as this are just my thoughts about the visibility of this project. I don't want to discourage you, but I hope this helps you to reflect and improve.

[SHC]

@atombrot

What you said is absolutely right, I appreciate you. But if I'm in your place, I would have sent them a personal message rather than posting it here. I think that will make great difference - nobody likes these helping notices given to them to be visible to the public, isn't it? Just a word of my thought and not as an offence.

Games published by our own members! Check 'em out!

[atombrot]

@SHC: No offense taken. You might be right with your concern, but given the latest discussion in this thread, I felt the need to post this here. Then again, I think some other users might also benefit of my thoughts and reflect a bit on their visibility, and so now there is at least a chance those users might see my post (in contrast to a PM).

My suggestion would be anyways to abandon the ship (not the project, but make a new clean thread, ask riven to remove most of the replies or make a good homepage/wiki for the project). With this my nasty post would also dissappear *har har har*

[Agro]

I'm honestly offended right now. We have been posting updates about the project when we can, but whatever. Doesn't matter anymore. 9 months of life wasted because no one has said a single time that hey, this project is kind of cool, keep it up or hey, I've noticed how hard you have three have been working, its ok if you occasionally joke around and have fun.

Is it forbidden to have fun on here?

Go ahead, feed me to the chitchat monster. I'm done with this site.

not really a good thing if you get offended so ezpzily. you cant make everyone like something so fast(lol 9 months), especially when you guys treat it like a meme (imo anyway). the project looks neat, its just the way you guys treat it that depreciates its significance.

point is that the joking around and having fun part is way more occasional than occasional, so ya kthx

[ctomni231]

Honestly, it is kind of hard for me to blame anyone.

 It is obvious that this team is a group of young folks looking to write a game engine. Noise aside, there is an innocent energy that this project has that a lot of projects lack. It is fun, new, and exciting while trying to be competitive. Regardless of how much I shudder at the lack of organization, I would hate for this project to turn into another "business like usual" library. You guys may actually be onto something with such a relaxed structure.

Yeah, you can have a front page and try to organize it a little bit. But, making a library seem like a fun activity is no easy task, and I am feeling that vibe from this group. Try not to lose that flavor as you move on. It is very refreshing to see young teams get things done just by their whims alone. Keep it up!

[Riven]

This meta discussion has got to stop at some point, so why not right now?

[Drenius]

On a side note, the website is pretty close to completion; Jev has estimated about 3 days.

We will see (sorry Jev)

Nice too see that you can still keep your motivation on this. The only way to create a good library is to keep refactoring it.

[ags1]

Do you have a link to the main source file for this intersection checking in (presumably) Git?

[Danny02]

how about:

1   2   3   4   5   6

int v = 0; while( v < vertices.length) {     Vec2 l1v1 = vertices[v], l1v2 = vertices[++v % vertices.length];     //... }

Games published by our own members! Check 'em out!

[ags1]

You do handle triangle-circle collisions, or rectangle-circle?

[Danny02]

If you are using maven, which is a great choice. You shoul get rid of any project folders from any IDE, configurate everything in maven.
If you need help with a proper .gitignore file, go to: gitignore.io

[BurntPizza]

I may be flat wrong on this, but aren't most things limited by fillrate rather than vertices? Also, stating what hardware you accomplish those numbers on would be a good idea I think.

Otherwise, cool stuff going on here. 

[kingroka123]

Wow, I envy  that nvdia 660  . You have a really good machine. Right now I have about the same power in everything but the video card. I have a radeon 7750, it gets the job done but I need to upgrade for next gen titles and the 660 is what I've set my sights on.

Now about that stress test: Really nice. Do you plan on android and ios versions being released because that would be very nice. Also, since your machine is so good, try testing on a slower device like a laptop just to be sure of the. Performance. Really good job with this    keep up the good work!

[Longarmx]

That card is unlocked correct? I think I have the same one. 

[ClaasJG]

The link in the front post is down.
http://weslgames.github.io/MERCury

But I read the wiki, nice work.
-ClaasJG

[kpars]

The link in the front post is down.

Don't worry. The new site should be up soon.

- Jev

[kingroka123]

Ok so I'm using this code in order to rotate a texture. I feel like I shouldn't have to reload the image each time I want to rotate it so is there any other way of doing this?

1	sprite = Texture.loadTexture(Loader.streamFromClasspath(path),(int)deg,GL11.GL_NEAREST);

[Edit] Also this keeps popping up everytime i rotate the image in this way:

1	WARNING: The provided Texture is NPoT (non power of two). Non-pot Textures cannot use GL_REPEAT.

My texture is 64x64 pixels and I'm not using GL_REPEAT so shouldn't this not be happening?

[kingroka123]

I am also loading a 16x16 texture but I am not rotating that one

[trollwarrior1]

@Wessles
I was looking at your stress test. Does your stress test mean that I could render 100k moving quads at 60 fps?

From what I can see, you have static geometry, and your quads are really small.

[matheus23]

Make the checkboxes' texture change, when you mouse-over. That makes the UI look much better and make it much more usable

[kpars]

We're going to be including a default Look & Feel in the future, don't worry... Those buttons are just for testing really.

It should be really fun to make... For me. It'll most likely suck for Wesley.
I don't have an issue with that.

- Jev

[NegativeZero]

For the last three options, they are a state rather than parameters, meaning that it is more intuitive to use radio buttons rather than checkboxes.

[Coldstream24]

Looking good. That said, the title text of the window looks a bit off-centre for me.
Also, with the close button, I think that if you want it to remain a circle, it's too big currently. OS X-size buttons may make it look nicer, and captions that appear when you mouse over (like an X) would look nice and shouldn't be too hard to implement.

Are you planning on adding event listeners to components and so forth (like mouse over, etc?)

[kpars]

Heh, those textures are just temporary until I get the time to work on the default look and feel. The reason why I'm not doing that right now is because I'm in the middle of getting the MERCury website done, which has been quite time consuming.

Expect a large post from me later regarding those two things.

- Jev

EDIT: Also, the reason why the text looks off-center is because the width that the text is centered with takes in both the width of the window AND the width of the close button. Sorry if I didn't word that too well, but you get the idea.

[Riven]

I'd advise not to make up awkward terminology.

SubTexture.convertToTexture() makes much more sense.

[kpars]

Fixed.

- Jev

[StumpyStrust]

Well ignoring whatever was going on in this thread, I tried out the lib and it really gives a java2D flavor. That is good as doing basic things in java2d is very easy.

However I wanted to see how easy it would be to port my particle lib and well...the batcher was kinda bad or more of missing things. I couldn't figure out how to specify rotation, scale, mirroring, etc. So I never made t very far. It took me a few minutes to port to libgdx and I was hoping it wold be the same for this lib but things I guess are still too immature.

Keep working on this if you like. I made my own engine'ish stuff and it was really fun for a while.

Requested feature that would really help game devs and greatly speed up rendering. Automatically adding textures onto an atlas unless specified not to.

PS: Look into using a geometry shader for sprite batching. I got about 100-150k particles or in your terms like 400-600k verts with basic batching. With a simple geometry batcher I made it p to 500k particles. It also has the advantage of smaller bandwidth.

[StumpyStrust]

The batcher is just to batch everything into one draw call ideally.You can batch things like rotation, scale, mirroring, etc for sprites. The next thing that could hit performance is texture switches which really do not matter if you are using an atlas or modern opengl. The geometry shader allows you to just give a single point and color and then the shader creates the geometry. Let me put it this way. You should try and batch everything.

This is how my render method works in no particular order.

drawImage(x, y, scaleX, scaleY, rotation, color, mirror up, mirror down)

The x and y will be the center of where the image is drawn. So drawImage(0,0,4,4) would be an image at the origin and a half width/height of 2. This I think is the most logical way for people to render a sprite. You give the location in the world where the sprite will be centered at and then some attributes like scale and rotation.

I would really look into doing a spritebatcher using the technique riven poste on here a while ago. People may use the graphics object for some stuff but when they want performance, they use a batcher. Heck I would make the backend of the graphics object batch everything.

[Riven]

This I think is the most logical way for people to render a sprite. You give the location in the world where the sprite will be centered at and then some attributes like scale and rotation.

You typically want to rotate around the 'origin', which could be anywhere in the sprite. This is a good reason to create a Sprite class, to store all these variables, instead of having to deal with a method call that takes more and more parameters, as the library matures.