Related
How would I go about modeling a turgid bag of fluid in Box 2D, i.e. what physics equations would be useful in modeling this? The bag of water could move when touched, but that is the only interaction. Any places equations or models would be much appreciated!
You could use a library for this. Google's LiquidFun is quite good http://google.github.io/liquidfun/
User99345's answer is the way I'd go to model a bag of liquid, but if you're wanting to use an unmodified Box2D library you can model it using instances of b2EdgeShape, b2RevoluteJoint, and b2CircleShape. Whether that's a good enough way to model it, you'll have to decide.
Additional basis/insight for this...
After seeing your question I put together a model for a bag of liquid as a Testbed demo in my dev branch of my fork of Box2D. The demo is called "Bag of Disks" because that's basically what the demo models and it's in the file BagOfDisks.hpp. The code uses a series of edge shapes connected by revolute joints to model a deformable container and fills it with circle shapes (called DiskShape in my fork) to model a liquid. If you build the library and Testbed of my fork you can see for yourself how the model looks.
As a model of a bag of liquid, I'm of the opinion that the code I whipped together has shortcomings like the following:
It must be computationally less efficient than using particle simulation like Google's LiquidFun does.
I don't believe joints as currently implemented in Box2D (or my fork of it), can always ensure containment. That's because I don't believe joints are able to 100% prevent alternative movement of the connected shapes.
The "bag" in the demo doesn't level evenly at it's top like I'd expect a normal bag of liquid to do.
The "bag" doesn't model a buoyancy force that I'd expect the top of a bag to do.
I imagine things can be done from a user level to improve the model's behavior but I suspect less can be done from a user level to improve the model's speed. By user level, I mean as a user of the Box2D library.
While my fork has made many changes to the Box2D library (especially in naming), I don't believe what I've done in the demo itself can't be reproduced pretty closely using the original Box2D library and syntax as I'd said in my first paragraph using instances of b2EdgeShape, b2RevoluteJoint, and b2CircleShape.
As to what physics equations would be useful in modeling this beyond the equations that Box2D already uses, I'm sorry to say that I have no idea at the moment. I am interested in that however and am looking into it as well. Physics equations for this are available of course but the closest related work that I'm aware of from a Box2D user level is what iforce2d put together in his Buoyancy web page.
Hope this answer contributes helpfully to what's already been said.
I am working on a game scene with multiple objects that need multiple materials. I extensively searched online, but I could not find any satisfactory solution.
My scene will have like a river flowing by and the material there will require a separate shader anyway (it will combine many specular and normal maps into what would look like a river) then there is a terrain that would mix two (grass and sand textures) requiring another shader. There is also a player with hands and amour and all.
EDIT: Essentially I wish to find out the most efficient way of making the most flexible multiple material/shader implementation.
Briefly, there is a lot of complex objects around requiring varied shaders. They are not many in number, but there is a lot of complexity.
So using glUseProgram() a lot of times dosn't seem like the brightest idea. Also Much of the shader code could be made univeral like point light calculation. Making a generic shader and using if's and state uniforms could possibly work, still requiring different shaders for the river and likewise diverging materials.
I basically don't understand the organization and implementation of such a generic system. I have used engines like Unreal or possibly Blender which use Node based materials and allowing the customization of every single material without much lag. How would such a system translate into base GPU code?
If you really face timing problems because of too many glUseProgram() calls, you might want to have a look at shader subroutines and use less but bigger programs. Before that, sort your data to change states only when needed (sort per shader then per material for example). I guess this is always a good practice anyway.
Honestly, I do not think your timing problems come from the use of too many programs. You might for example want to use frustum culling (to avoid sending geometry to the GPU that will be culled) and early z-culling (to avoid complex lighting computations for fragments that will be overriden). You can also use level of detail for complex geometries that are far away, thus do not need as much details.
I'm coming to the end of my first year of CS and I thought a great way to consolidate all the things I've learnt this year would be a personal game project.
I would like to implement a 2D based rts, I'm thinking along the lines of starcraft I, warcraft II or even command and conquer. I will have about 3 months without interruptions to implement the game.
So to anyone experienced with java game programming, I have a few questions:
Is it realistic to design a 2D rts engine from scratch in 3 months?
If so what are some good books/resources to get started?
Would it be better to modify some existing project? I would think the experience of having to work with a lot of someone else's code would be good since our exposure to such topics in an undergrad cs degree seems very rare, if non-existent.
Are there any decent open source 2d rts projects that anyone could recommend? I've looked through a few but most seem to be written in c/c++
My humble thanks
Edit: Thanks for the quick responses, I think that perhaps it was a bad idea to post this in a rush since I think I misrepresented what I want to do.
When I say "along the lines of warcraft II etc" I mean more like that style of rts using sprites. I don't intend to implement a game nearly that complex, more like just a basic prototype.
My goal would be some thing more like a flat textured map with some basic obstacles like trees, a single unit producing structure like a barracks. I'd like to have the units to have health bars, be able to move and attack and die (and possible morph into another unit).
Far off goals would be to implement some basic pathing using a modified version of the dijkstra shortest path algorithm, ranged units with missle attack, etc.
I don't plan to implement any opponents or ai or networking or anything like that.
I'm thinking along the lines of starcraft I, warcraft II or even command and conquer
Make sure you purge your mind of matching the full scope of any of those. They took large teams of developers multiple years to make, with multi-million dollar budgets, so you can't even hope to approach those. They're called AAA for a reason. That being said, there's no reason you can't very minimally ape their design, or make a tiny game in their genre, assuming you have previous experience making small games.
A sub-genre of RTS that might be doable in that amount of time is a Tower Defense game. Plants vs Zombies is a good example. The reason I suggest this sub-genre is that you can avoid implementing any sort of AI or path-finding, which are notoriously difficult to get working, and I think technically impossible to implement "perfectly", especially with a limited CPU budget.
Make sure to reign in your scope. Favor a "complete" game over new features, because you can then call it "done" at any time. Get your game playable ASAP, and don't sweat the polish or details until you have to. Add one enemy type and one type of player unit (with only one ability, if you were thinking of implementing multiple abilities per unit). Make a title screen, menus (even if the menu is just "click screen to play"), game over screen, level complete or stat screens, cross-level player statistics, etc. Once you have all that ironed out, spend equal time adding new features and polishing the gameplay/graphics/bugs.
Once you have a playable, "complete" game ready (no matter how small in scope), find a real artist to do graphics for you. A shiny game always draws an audience, no matter how simple the gameplay.
It is very unrealistic to think you could implement a 2D RTS engine anywhere even close to the complexity in those kind of games. You could maybe get something very rough if you were experienced, but with only one year I think it's doubtful.
I can't help but feel like it would be much better for you if you used an existing engine or framework and built off of it. Like you said, working with other code would probably be a good learning experience as well. It would allow you to experiment without getting bogged down in having to do everything.
Keep it simple or you will simply drown in complexity before getting around to have anything playable. Since you have not tried it before, you will have a lot of nuts to crack and you don't know how long they will take.
Also remember that report writing and documentation takes time too.
The idea is good, and I think you can pull off a whole game if you find good building blocks. I would suggest discussing this with your teacher to hear what is acceptable for you to use. Would it e.g. be ok to do a game on an open source engine if you add some non-trivial functionality?
Update: Seems to be several engines available from Java at http://www.devmaster.net/engines/list.php?fid=6&sid=1
People often forget, that creating games is MUCH MORE than just coding the technique thing. Its about content creation, game design, sound and music, the "fun factor". If you make heavy use of existent APIs or engines, it will be possible, but writing it from scratch with no experience in 3 month is like asking yourself if you can code 100,000 LOC in this time which means 1111 LOC per day. This might be possible, but not if you have to desing and think, and just having the code makes no game.
Perhaps it would make sense to look at some existing efforts to get a feel for the scope of what you are looking at. These should give you some ideas or even code to build on:
http://www.duncanjauncey.com/btinternet/old/javagame/game.html
http://en.wikipedia.org/wiki/Lightweight_Java_Game_Library
http://www.ardor3d.com/
http://en.wikipedia.org/wiki/JMonkeyEngine
It would be a lot for me to bite off (from scratch) in the time given that is for sure. That is about all I can say.
EDIT: I thought maybe JOGRE was not what you are looking for. Then I thought about it and it seems like it would have all the right kinds of plumbing for what you are trying to do.
EDIT AGAIN: After my answer, one of the related questions links on the side seemed relevant: Java Game Programming: JOGL vs LWJGL?
Well if it gives you any hope at all, my team and I are currently working on an RTS game called "The Genesis Project". We call ourselves MotherBoard Games, or MBG for short. If you would like, I am always looking for more coders. You can email me at mpmn5891#gmail.com, I can give you some advice and tips form my 6 year experience, 2 of which have been spent making this game (to give you a scope)
I am an undergraduate student. I was exposed to basic programming couple of years back in school. Till now I have an understanding of Core Java, Core Python and basic C and C++.
Every time I start off with some GUI programming so as I can start off with a project of mine, I get boggled by the sheer amount which is to be done, API to be learnt, MVC architecture and everything programmers talk about, event handling etc etc.
Studied awt and swings for a while. Tried my hands on Qt and Gtk, could not find much of documentation. Tried to make sense of pygame. I end up at the same place, knowing the core language.
Tkinter on my zenwalk Linux is broken so could never start it athough I own a book on python with Tkinter explained.
But I end up at the same place, with just the basic understanding of the language.
Want to start over, seriously now. I would like to choose python. How should I go about studying GUI programming?
I need some Internet resources and direction so that I don't end up at the same place!
Since it sounds like you want Python GUI programming, may I suggest PyGTK?
That's probably a pretty good place to start for someone who knows Python and would like to start small on some basic GUI apps. GTK can be complex at times, but with PyGTK there's plenty of open-source example apps you can study, from simple to complex.
Edit: This tutorial from LinuxJournal seems pretty helpful.
Edit 2: Here's the tutorial from PyGTK's site, and another tutorial I randomly found from Google (seems like that whole blog is pretty useful for what you want to do, actually). Finally, the snippet at the bottom of this page might be helpful, courtesy of Ubuntu's forums.
If you are leaning more to games...
I suggest you install Pygame and Python, and go through their tutorials. The pick a simple game or graphics project and program it!
For Python GUIs I like wxPython (www.wxpython.org). It is pretty easy to get started with simple controls and layouts. It is also cross platform. Plenty of tutorials out there. Just search for wxPython tutorial.
I know how you feel--I learned a whole lot of computer programming during my CS degree but very little about GUIs. I ended up teaching myself Cocoa/Objective-C for a project. Cocoa is wonderful for GUI stuff but often a royal pain with a steep learning curve. If you don't have any experience with C programming, don't bother.
First step: familiarize yourself with the MVC (Model/View/Controller) design convention, because nearly every GUI framework will reference it. Google it--there are lots of resources about it. My quick, simple definition is:
The model level defines the data or the logical model for the application. For a web app, that would be the database. For a game, it could be stored data and game logic/rules.
The view level is what the user sees and interacts with (the GUI).
The controller level is the logic that connects the two. For example, the controller knows that when you click the "start game" button in the view level, it does some stuff with the model (say, setting up the board and the players.)
Step two: Figure out what you want. Are you interested in desktop applications specifically? Games? Web apps?
If mostly what you want to do is to be able to develop something that people would actually use, another option is to learn a web development framework. The frameworks make stuff easy for you. I love Django, personally, and if you know a little Python and a little HTML and a little about MVC, you can pick it up quickly. (Just don't be confused, because what Django calls a view is actually a controller.)
If what you want to do is games or graphics/animation stuff, check out pygame. I used it for a class project--basically taught it to myself in a couple of weeks--and it worked great.
I'd say stay as far away as you can from Java Swing/awt/etc.
I've heard good things about wxPython--I almost ended up using it instead of Cocoa, because the wx stuff is available in several programming languages and it's all cross platform.
Good luck! Stay strong! I know it's really intimidating, because I've been in your shoes. You can do it with some work, practice, and motivation.
Many have recommended wxPython, and I second their enthusiasm - it is a great framework; it also includes a serious demo (with code and live applications) which will be extremely valuable for learning.
Now, BEWARE!
It is very simple to confuse the end with the means. Programming GUIs can be extremely attractive but not very productive. In my early days I spent days and days trying to get a simple plotting application (reinventing the wheel); a simple GUI for solving quadratic equations; a simple GUI for calling database queries by clicking on certain locations on a map, etc. During all this time I never actually dug into algorithms or more general and productive computer science and computer engineering topics. In retrospect, I should have. Granted, I did learn a lot and I don't totally regret it, but my advice stands: worry about your algorithm first and about your interface second. This may not apply to every field (I am an engineer for NASA). Nowadays I work with number crunching applications with no GUIs whatsoever; I don't think they need them!
Anyway, I just wanted to share my two cents with GUI programming - have fun but don't overdo it.
What do you mean by "Graphics"? Do you mean game graphics, or do you simply mean user interface code (forms, webpages, that sort of thing)? In the case of game graphics, there's a limit to how simple things can be made, but http://www.gamedev.net, for example, has tons of introductory articles on 2d and 3d engines. For something more along the application line, you might simply download Visual Studio or Eclipse and spend some time looking at the code that is autogenerated by their WYSIWYG editors.
For GUI work in general:
Less is more
GUI work (even in productive frameworks) is about as fun and productive as painting the Eiffel Tower with a toothbrush. Go for a minimal design.
Avoid State Like The Plague
Do you put state in your GUI, or in the model? If you put it in the GUI, you are going to mess yourself up with redundant and inconsistent code paths. If you put it in the model, you risk an overly complex system that gets out of sync when your GUI fails to update from the model. Both suck.
wxPython
If you want to learn wxPython, here are a few traps I noticed:
The tutorial
Use this tutorial - http://wiki.wxpython.org/AnotherTutorial
It's the best one I found.
But remember to toggle line numbers, for easy pasting.
Events
Events are a bit like exceptions, and they are used to make things interactive.
In a vanilla python program, you write something like:
def doit(i):
print 'Doing i = ',i
for i in range(10):
doit()
print 'Results = ',result
In a GUI, you do something like:
def doit(event):
print 'An event',event,'just happened!'
event.Skip()
import wx
app = wx.App()
frame = wx.Frame(None, -1, 'The title goes here')
frame.Bind(wx.EVT_KEY_DOWN, doit)
frame.Show()
app.MainLoop()
Every time the user presses a key down, an event will be raised. Since frame is bound to the event (frame.Bind(wx.EVT_KEY_DOWN, doit)), the function doit will be called with the event as an argument.
Printing to stderr isn't too hot in a gui, but doit could also call up a dialog, or do anything you want it to.
Also, you can generate your own events using timers.
Apps, Frames, Windows, Panels, and Sizers
Everything has a parent. If an event is raised, and the child doesn't skip it (using event.Skip()), then the parent will also have to handle the event. This is analogous to exceptions raising up to higher-level functions.
A wx.App is like the Main function.
wx.Window isn't really used. Stuff inherits from it, and it has all the methods for sizing and layout, but you don't need to know that.
wx.Frame is a floating frame, like the main window in Firefox. You will have main one frame in a basic application. If you want to edit multiple files then you might have more. A wx.Frame won't usually have parents.
wx.Panel is part of a parent window. You can have several panels inside a frame. A panel can have a wx.Frame as a parent, or it might be the child of another panel.
wx.Sizers are used to automatically layout panels inside frames (or other panels).
Code:
def doit1(event):
print 'event 1 happened'
def doit2(event):
print 'event 2 happened'
import wx
app = wx.App()
frame = wx.Frame(None, -1, 'The title goes here')
panel_1 = wx.Panel(frame,-1,style=wx.SIMPLE_BORDER)
panel_2 = wx.Panel(frame,-1,style=wx.SIMPLE_BORDER)
panel_1.Bind(wx.EVT_KEY_DOWN, doit1)
panel_2.Bind(wx.EVT_KEY_DOWN, doit2)
panel_1.SetBackgroundColour(wx.BLACK)
panel_2.SetBackgroundColour(wx.RED)
box = wx.BoxSizer(wx.HORIZONTAL)
box.Add(panel_1,1,wx.EXPAND)
box.Add(panel_2,1,wx.EXPAND)
frame.SetSizer(box)
frame.Show()
app.MainLoop()
I've been really bad, and not used OOP practices. Just remember that even if you hate OO in most contexts, GUI programming is the place where OOP really shines.
The MCV
I don't get MCV. I don' think you need an MCV. I think a MW (model-widget) framework is fine.
For example - 2 frames that edit the same piece of text:
class Model(object):
def __init__(self):
self.value = 'Enter a value'
self.listeners = []
def Add_listener(self,listener):
self.listeners.append(listener)
def Set(self,new_value):
self.value = new_value
for listener in self.listeners:
listener.Update(self.value)
import wx
app = wx.App()
class CVFrame(wx.Frame):
def __init__(self, parent, id, title, model):
wx.Frame.__init__(self, parent, id, title, size = (100,100))
self.button = wx.Button(self, -1, 'Set model value')
self.textctrl = wx.TextCtrl(self, -1,model.value)
self.button.Bind(wx.EVT_BUTTON,self.OnSet)
self.model = model
model.Add_listener(self)
sizer = wx.BoxSizer(wx.VERTICAL)
sizer.Add(self.button,0,wx.EXPAND)
sizer.Add(self.textctrl,1,wx.EXPAND)
self.SetSize((300,100))
self.SetSizer(sizer)
self.Center()
self.Show()
def OnSet(self,event):
self.model.Set(self.textctrl.GetValue())
def Update(self,value):
self.textctrl.SetValue(value)
model = Model()
frame1 = CVFrame(None, -1, 'Frame 1',model)
frame2 = CVFrame(None, -1, 'Frame 2',model)
app.MainLoop()
wxPython has a listener-subscriber framework, which is a better version of the model I just sketched out (it uses weak refs, so deleted listeners don't hang around, and so on), but that should help you get the idea.
If you have already gone through pygame, tk, Qt, and GTK, then really the only thing left that I can think of is pyglet, which I admit I have not tried, but I have read uniformly good things about it.
Still, more than anything it sounds as though you have trouble sticking with a framework long enough to really grok it. May I recommend starting with a small project, such as Pong or Breakout, and only learning as much as you need to make it? Once you have finished one thing, you will have a feel for the library, and continuing past there is a lot easier.
whatever language you choose you will have to deal with the many details involving GUI programing. this is due to the nature of the window based environment usually used for GUI.
what can help you move forward quickly in developing GUI based application is less the language and more the IDE you use. a good IDE can do some part of the less interesting stuff for you letting you focus on the big picture.
with C# in VS 2008 its all about choosing elements and methods from lists boxes. its very easy to get started and have a working project.
you can then try to customize your application to gain better understanding of whats going on behind the scenes
One of the greatest Python GUI you can study from is the source of IDLE. It always comes with Python.
For Java, you could also look into SWT.
While I have never used AWT or Swing, I have read that SWT is the easiest of the three to learn.
Here is a decent comparison between the three.
We're a team of a programmer and a designer and we want to make a medium-sized java game which will be played as an applet in the web browser. Me (the programmer) has 3 years of general development experience, but I haven't done any game programming before.
We're assuming that:
We'll decide on a plot, storyline of the game, etc.
We'll create a list of assets (images) that we need, i.e player images, monster images, towns, buildings, trees, objects, etc. (We're not adding any music/sound efffects for now)
The designer will get started on creating those images while I finish reading some of the game programming books i've bought. The designer will create the first town/level of the game, then pass on those images to me, I will begin coding that first level and he would start on the next level, and after 4-5 levels we'll release v.1 of the game.
Question 1: Is this the correct methodology to use for this project?
Question 2: What format should the designer create those images in. Should they be .bmp, .jpeg, or .gif files? And, would he put all those images in one file, or put each monster/object/building in its own file? Note; We are sticking to 2D for now and not doing 3D.
Question 3: I've seen some game artware where there would be a file for a monster, and in that file there'd be about 3-4 images of a monster from different directions, all put in one file, i think because they're part of an animation. Here's an illustraton:
[Monster looking to right] ... [Monster looking in the front] ... [Monster looking to right[
And all of them are in one file. Is this how he'll have to supply me with those animations?
What i'm trying to find out is, what is the format he'll have to supply me the designed images in, for me to be able to access/manipulate them easily in the Java code.
All answers appreciated :)
I have some comments for each question.
Question 1: You say that you will begin coding level 1, 2, .. one by one. I recommend you to create a reusable framework instead or see it in the big picture instead. For the information you provide I think you are going to make some kind of RPG game. There are lots of things that can be shared between levels such as the Shop, the dialog system, for example. So focus for extensibility.
Why wait for designers to pass on the image? You can begin your coding by starting with pseudo graphics file you created yourself. You can then work with designer in parallel this way. And you can replace your pseudo graphics file with ones provided by designer later.
Question 2: JPG is not suitable for pixel-art style image, that appears a lot in most 2D game. And the GIF support only 256 color. The best choice to me seems to be PNG.
The designer should always keep the original artworks in editable format as well. It's likely that you want to change the graphics in the future.
Question 3: It depends. The format mentioned, where character's animations are kept in single file, is called Sprite. If you kept your resource in this sprite format than you will have some works reading each of the sub-image by specifying coordinates. However, sprite helps you keep things organized. All the 2D graphics related to "Zombie" character is kept in one place. It is therefore easy to maintain.
About the image format: don't let the designer deliver anything as jpg, because you'll lose quality.
Let him send it as png instead, and convert it to your preferred format as needed.
Also, remember to have him send the source files (photoshop/illustrator/3dsmax/whatever) in case you'll ever need tiny changes that you can make yourself without hiring the graphics dude. Who knows if he'll still be available in the future anyway.
I want to suggest to you that, before you make any decisions about your workflows, you and your colleague go have a look at JavaFX and see if maybe that's the toolkit that best meets your needs.
http://java.sun.com/javafx/
The [Monster looking to right] ... [Monster looking in the front] ... [Monster looking to left] style of animation demarcation has been around for as long as I've been peeking into game data, so I would suggest going with that path.
I was about to make the same remark as Wouter: use PNG, modern format which is highly compressed (as opposed to BMP), lossless (as opposed to Jpeg) and full color and with several level of transparency (as opposed to Gif).
Why people put several sprites in the same image? Actually, for Java, I am not sure, if the images are part of a jar... I know it is interesting in CSS, for example, because it reduces the number of images to download, so the number of hits on the server, which is a well known Web optimization. For games on hard disk, reducing the number of small files can be interesting too.
The designer can appreciate this too. At least in times where sprites used a color palette: you had only one image, using the same palette: easier to edit, and slightly reduce the overall size (in times were memory was costly!).
I can't answer on the methodology, I never did a game in team... If it fits your needs, it is probably the right methodology...
duncan points to JavaFX, I will point to pulpcore which seems to be a promising library. Of course, there are plenty others, like JGame and such.
Bunch of pros here: http://www.javagaming.org/
This is not answering any of the questions. But for game develop/Simulation Engines learning if u need a reference:
http://www.cs.chalmers.se/idc/ituniv/kurser/08/simul/
It's a link for the class lectures of Simulation Engines at Chalmers Univ in Gotembourg. The teacher as a game company and gave quite good lectures. Check the slides we had in the classes, maybe they'll help you a bit.