That is the exact thing I'd recommend you to do in your situation.

Let me guess: by pure Java you mean Java2D and that's why you instantly liked the idea of using Slick, right?

I only used them in school Ugh, those were not my favorite assignments. I see UML as a good learning tool when you don't have a strong grasp on OO programming yet and I also see some minor documentation benefits in it, but to actually use it in the design phase of an application... hell no. Applications are living and breathing things, the last thing I want to do is to be maintaining piles of UML diagrams whenever there is a change in design/architecture.

The framework used is but a hammer anyway, its in the actual creation of a game where the real fun is.

As an exercise I tend to make juniors I'm training in Java development to create a command prompt hangman, polished as far as that is possible (neat textual output, using an actual word dictionary file, creating a neatly packaged application which can be executed with a single click, etc.). Its amazing to see their faces light up when people in the company start to seriously play it and provide positive feedback and even enhancement requests.

Can you confirm that I am understanding your intentions correctly.

- your passion really is in complex networking solutions and you are wildly interested in the workings of for example clustered game servers such as used for MMORPGs
- you don't actually find the creation of the client part of a game "fun", its a necessary evil to be able to play around with the "business logic" bits you do find interesting (such as AI). To compare: you like creating the back-end of a web application but you don't really like creating the front-end; you'd rather just use a framework to quickly crank out the user interface and not really bother with the markup and the styling.

Am I right?

I own many game programming books, all of them using C++ as the reference language but what is important about them is not the code, it is the theoretical knowledge kept in them. How do graphics work, how do you structure a game program, how do you solve common problems, etc. etc.

If I am perfectly honest, of that big pile of books I have, about 90% of it was completely useless for practical purposes in the end. Most books hammer for far too long on setting up a development environment and rehash all the "how to program for operating system X" without even touching game development at all, before going into a rehash of the API documentation of whatever graphics API is used. Sound and input tend to be rushed single chapters while topics such as AI are a paragraph mentioning that its hard and out of scope. Chapters on wiring an actual game together tend to be of the "Yeah I built this game for you, let me print parts of the code in the book and go through it line by line" - example game being broken and uncompilable by default of course.

But still that remaining 10% was theory that stuck with me. It wasn't worth the price I had to pay for such works though, but it was all I had with no real connection to the internet yet. Nowadays: nah, use the internet.

I took another route: to me engines don't exist except in the minds of Call of Duty and Battlefield players. There is code and there are tools, nice and simple

100% agreed. Its also the "is a tomato a vegetable or a fruit" kind of deal to be honest. To one group of people an engine is a set of tools to create games with while to a whole other group of people it is something that needs to be upgraded yearly to magically get better graphics.

Ha ha ha, that's a good one

A mission is fruitless, but an intention to keep the forum awesome is good enough. That is done in many ways, including having the occasional huge disagreement with flaring temperaments to keep the blood pumping and the ever so important programmer's ego in-tact. When it is too quiet one just has to drop a Maven or Eclipse bomb and off we go.

So basically you want to do it quick and with little effort. That goes into the territory of a game maker tool, not programming!

I once did a tile based game in "just" 2 days though, using C++ even and using no framework at all except for some code I used in a previous game; the graphics/sound stuff was done using very basic DirectDraw/DirectSound. I managed to do it that quick because, besides having experience already, it was a remake of an existing game for the old Commodore Amiga platform; I could rip the graphics, sounds and music from the original (and I had plenty of experience doing exactly that), I knew how it should work from the get-go so I didn't have to design anything, I just had to program it and make it work. It was also a puzzle/board game with no scrolling and no physics and such.

So if you limit very much what the game will do, MAYBE you can do something quick. But I wouldn't count on it, it takes plenty of experience and lots of focus to finish a game.

Apparently, it is already here:

http://blog.robovm.org/2013/05/javafx-openjfx-on-ios-using-robovm.html

Ah a nice thread derailment!

I will (again) say one thing about Maven poms: I don't write them, I copy/paste them. All my poms are nearly identical with tiny variations here and there (generally which dependencies are included and on what scope). I don't see how beyond your first attempts at using Maven you spend much time on setting stuff up, its usually a matter of 5 minutes for me and with those 5 minutes I also setup a proper Eclipse workspace with 90% of the build settings correctly in place. Its just too bad that the m2eclipse plugin sets some wonky defaults :/

There were some really excellent pom templates here on the site somewhere.

Switching to serious mode: I can't agree or disagree with you, I've lost exactly what you are referring to. If I compare my C code to Java code that performs the same task, the Java code wins at all times. Because its far easier to read, often more compact and written and debugged in far less time. That's my personal reality.

In other words: you suck at Java 

Yeah, screw that!

Besides Eclipse 4.2.2 is actually quite decent (once you set the look & feel to classic). No more performance problems for Gimby, UI updates smoothly, plugins just work - love it.

I've seen so many ways my (attempts at) games stuttered and bogged down over the years... the most critical was using integers to represent speed and positions where I now use floats. The second most critical was to let the loop timing ("when to draw the next frame") go out of sync due to not correcting for overshoots and such.

Doesn't that leave the poor spread of OpenGL 3.X on 'regular'/nameless PC's/Windows though? I remember a discussion which stated that in order to get a decent enough sized target audience you'd need to stick to OpenGL 2.x at the most. Is that no longer true (or am I making stuff up)?

Excellent question, often goes side by side with "you ain't gonna need it".

The performance in Java depends greatly on how well the compiler and runtime can optimize your code and application. A recent Minecraft fix greatly improved performance because a certain piece of code that was basically executed for every drawn block was head-butting with the optimizer. If the optimizer can do its job well, you might even see more sparks flying than a similar native application.

The question is: how do you make sure you do not get in the way of the optimizer? That requires intimate knowledge I'm afraid, which hooks back into my favorite saying: programming is and will always be hard. You just don't get it for free.

Its not a necro when the thread is as young as this, and I missed it and now I got to enjoy that epic read so thanks for that.

I've done some PHP myself in the past, only back then I was simply too inexperienced to put my finger on why it sucked. All I knew was that I was so darned unproductive with it and it was hard to create something interesting that was not slow. That's not to say that I was more productive when I switched to working with servlets & JSPs on Tomcat 4 at the time, but at least I had the sane Java JDK and virtual machine to back me up. Java has been my best friend since, as long as you stay away from applets of course.

I also did some Ruby on Rails which I am marginally impressed with, but it is so flexible it scares me. Any platform that allows you to add methods to a specific object INSTANCE is mad in my opinion. Great for runtime instrumentation in frameworks, but it is available for anyone to do it.

Every tech related forum should have it IMO. 180 days seems like a good bet for this particular forum since topics don't tend to zombify so quickly. For a Stackoverflow or an OTN I'd have opted to set the deadline to 2 months if it were up to me.


For the WIP its actually a good thing, keep the progress going or have your thread necro'd  The other two are good points however, you'd want people to be able to leave a congratulatory message or ask a question no matter how old your game gets.

Hence my comment to shift focus after restarting once I'm not saying to not do it, just to not keep going for too long.

I wouldn't even dream about creating a moddable game until you actually have a few games under your belt. And by then you'll probably have a very good idea on how to do that.

There is a huge problem with generated worlds too: they're often very dull and riddled with impossibilities, no matter how good you make the generator. People have used the word "soulless" and that describes it very well. Minecraft is both an example of and an exception to that rule; most worlds you generate are pretty useless, you actually need to go to minecraftforum.net and look at the feeds thread to get ones that are by chance fun to use.

Except the difficulty of creating a game loop with a steady heartbeat, which LWJGL helps you do quite efficiently especially in more recent versions.

Of course a good reason to use LWJGL over Java2D (which I assume you're talking about, although it could also be JavaFX) even for simple 2D games is reliable performance across computers. It all may run just great on your development computer, now try it on your dad's laptop.

I think you should revisit the definition of purist and make your life easier. And also: hi

Google? I input JglfwApplication and get the javadoc:

http://libgdx.badlogicgames.com/nightlies/docs/api/com/badlogic/gdx/backends/jglfw/JglfwApplication.html

Which simply states it wraps around GLWF. So back to a google for "glwf":

http://www.glfw.org/

About 10 seconds of effort.


You should consider going to do something that is not related to computers in any way before you hurt yourself.

So basically physics based platform games are not really your thing I like Trine well enough, for a couple of days and then it got really boring.

You should definitely check out Giana Sisters: Twisted Dreams, Fez and Rayman Origins. Braid is nice if you like puzzle games. One that I would recommend keeping an eye out for is the remake of Flashback that is pending; that is going to be platform awesomeness of epic proportions.

http://www.youtube.com/watch?v=HxfNmui1S1Q

The fun comes from creating -and finishing- the game, not from reinventing the wheel and doing what legions of other people have done, which is focus on the tech. After having been stuck in that "engine building" bottomless pit myself for years I implore you; the first time you restart from scratch to make things better, switch to working on a game in stead of an engine. I waited until I was ready to restart for the fifth time, 4 years of my life were 'lost' without any actual game to show for it.

Really nice article focusing on the primary pain points without digressing into off-topic ramblings. I also like the concluding message (at least my interpretation of it): the tool does not do the job, you do.

It does look insanely cool though. The fact that he managed to make it work like this on his development machine is still an achievement, even if you'll never make it perform well in general.

I believe some Battlefield game (the futuristic one) had something like that. At least I remember reading about it, I don't play the games myself.