One of the most common things to do when developing iOS or Android application is to interface with external plugins (i.e. facebook, gree,urban airship) that are not provide natively through Unity. The old method used to be sticking a scripts in the “Assets\Editor” directory called either “PostprocessBuildPlayer” on iOS or adding a menu item for windows accessible via the editor. This method of madness is documented on Unity Build Player Pipeline page. However with the advent of 3.5.2, you can now write C# scripts to automate the process with having to write python/perl scripts. Documentation is lacking at this time but you can see an example here. Hopefully Unity will start documenting these features soon.
This blog is more of a reference (for me) to keep track of my efforts to research creating plug-ins for iOS and Android and streamline the build process out from Unity. Right now besides the two links describe above, I am also looking into efforts to add files, directories to an XCode project. My initial research starts with:
- Mod PBXProj – Which is a python script which allows you to modify the xcode project. While is still useful for research purposed with the advent being able to control the build using C# scripts this becomes less relevant. Still looking at the script allows an insight into how and what to manipulate to modify an xcode project to allow references to be added.
- dcariola/XCodeEditor-for-Unity – This seems a lot more relevant but haven’t found out how usable is it. If you grab the source at the time this blog was posted, you will also need to grab a the MiniJSON.cs from here and not the darktable one that pops up when you do a MiniJSON search via google. Basically what this should do is similar to the Mod PBXProj allow you to add files and directories to the XCode project. Importantly, you can also import a ‘projmods’ JSON files which will allow you to specify sequence of commands to add specific files. I haven’t yet been able to track down what format this file takes and will post more when i find out.
My current path is to start with the XCodeEditor for Unity to see if it does anything useful yet and perhaps contribute to it. If anybody can point me to the format for the projmods JSON file and the format of the .pbxproj file that would really help.
I have been using Unity for a while and just my my sake I wanted to compile a list of Do’s and Don’ts that will save you a lot of time (or if you are like me stop from going Doh! and hitting your forehead a few hours later).
1. Importing from Blender to Unity. Make sure you set the import to calculate your normals again. At least in Blender 2.56a the exports product models with extremely dark Textures. So unless you want environments in your game to be a homage to 2001 Space Odyssey.
3.Never use gravity a local variable name. In fact make sure if the editor highlight any variables that you aren’t using as a public or private variable. Bad things happen.
4. Always print out the gameobject you are acting against if you can’t find the components on it. I was trying to figure out why my projectiles wasn’t able to SendMessages or access components on a gameobject. The Tag was right. Made a fool of myself on Unity answers.
5. You CAN attach multiple scripts to a player. You can attach it to colliders on the bones that are hidden ONLY if you drag it to the scene and inspect it there. It won’t show up on the prefab.
6. Use the 80/20 rule of networking (well my 80/20 rule anyways)..If you think or findout that 80 percent of the time the variable needs to be updated, use state updates to do it versus RPC’s. I haven’t done a benchmark on Unity3D RPC’s but in my past experience with newtork coding RPC are slow, not timely and fat on network bandwidth. Easy to use again, but not to often.
7. Put something in your script name to help you find where it is. When inspecting a game object you can find scripts attached but have to hunt for the scripts. Putting all the scripts in 1 place and for me starting with a Prefix is a must. For example call player code ‘PlayerInputController’ and put it under a player folder in scripts. Good for you and good for any programmers that follow
8. Understand the code your write (or copy). There are alot of helpful scripts but the need to complete a project will drive you to cut and paste code. Soon you will have jumble of unoptimized code and if you don’t understand it a nightmare just waiting to happen.
9. The inspector can be useful to debug your game. If you select it during gameplay, you can check out public variables. It’s a quick and dirty way to do debugging rather then setting breakpoints which I haven’t done yet because it’s a pain to switch to the MonoDevelop (which is buggy, the scroll bars don’t respond to my mouse wheel and the folder ‘lock up’ so I can open it to click on scripts).
10. Beer and code don’t mix unless you have to re-write some particularly painful code. ;P
This post is a follow up to my first one which gives you some concepts on working with networking. In this post, I will cover getting an a client/server connection going which is the first thing you have to do in building a multiplayer game. Your first point of reference is to make sure you are familiar with the Unity3D Network class which provide most the functionality you need to manage an initial network session.
I would like to cover some Unity3D specific topics on how networking is handled. Unity uses a Network View Component that is attached to every game object who’s state you want to synchronize with. It also allows you to send messages directly to those objects because each one is issued a network ID. The following picture should make things clearer:
Let’s also cover the ‘responsibilities’ in networking between the Client and Server.
- The Server’s primary responsibility is managing the network session and propagating/validating state changes to connecting clients. It is also responsible for control of non-player game objects(i.e. enemies) by creating them when needed, running the AI routines which tell them to move. There are 2 modes that the server can operate under, authoritative and non. Authoritative servers have more work because they have to validate state changes from the client rather than merely passing it along to other clients. This is usually a slower way to operate as each move, action needs to be checked. If found to be invalid then the client that sent this original state changes need to be ‘brought in line’ with the way the server is thinking. A non-authoritative server on the other hand is more of a delivery person who simply relies information over to connected clients
- The Client is responsible for the view of the world a player sees and for receiving input and forwarding it over to the server. In a non-authoritative mode, the client can also process game logic (i.e. decrease health, move the player, move other objects) and these are sent to the server to be relayed to the other clients.
The first step in getting a network session up and running is initialize a server, which should be done when a level in the game loads up. This is done using the method on the Network class called “IntializeServer“. The first parameter is the number of connections to take. The second the listening port for the server, which is used by the clients to determine which application to connect on the server (as a server may be running multiple network applications). The last parameter if Network Address Translation should be used or not. All these except for the number of connections should be eventually made user configurable when you are writing your game.
Once the server is running, clients can connect using the Network.Connect method. The most important parameters for now is the ip address of the server and the remote port(which was the listening port specified earlier when the server was initialized).
That’s pretty much all there is for setting up a session. Be sure to check the Network class for more information on how to handle disconnects and network failures.
I have set about learning about networking on Unity3D in order to implement multiplayer support for Zombie Gecko’s Comrade In Arms. I have a fair amount of experience with networking, having first cut my teeth writing real-time transactional system allowing Unix hosts to update banking details to a IBM Mainframe using XNA, then working on early Internet credit card authorizations system and finally developing the server/networking code for Fung Wan Online MMOG. Well enough about me ;P
Unity3D uses RakNet for standard networking and who ever is implementing things will have a lot of plumbing to take care of inorder to get things working. Unity uses a “Networking View” component which must be attached to all objects which in turn Network State Synchronization and Remote Procedure Calls(RPC’s) to occur. You can set specific spatial components to be observed by a network view but usually you will implement a script that will pre/post process network information sent. This was something I missed when I first started out tinkering with networking on Unity
Let’s go back a bit and talk about the 2 different methods of communicating in Unity:
- State synchronization - this is basically a method of setting the network view to observe a script component that processes changes in states that can be sent to the server which in turn are transmitted to all listening clients.
- Remote Procedure Calls – This is a common enough networking concept that allows you send invoke functions on the server or all clients with parameters being passed.
You can do some additional reading on networking by looking up the following resource:
- Unity documentation
- iPhone Multiplayer demo – still works and you can substitute a web view for an iphone
- Unity Networking Example – Still works on Unity 3.3
- M2H networking tutorials – It cost money now, but there was a simple network tutorial that is floating around the network.
In a follow up posting, I will go through making connections,extrapolation,interpolation and hooking RigidBody objects.
Although an component system is the basis for most modern game engines and considered as the better(if not the best) way to put together a game. One of the biggest drawbacks is that when you create an architecture that is modular and loosely coupled, you introduce performance overheads.
On PC and other higher end devices, the tradeoff is smaller in exchange for elegancy of design and implementation. On the iPhone, especially older ones you have to to be care about how to implement things. The most common discussion on most game dev sites on the iphone is to watch your draw calls. On older 3G devices, 20-30 is a bandied around number…well that obviously and how many polys you are using plus if your code is running native or script…where is all this leading too….Ah yes the component system I was building on top of Cocos2D.
Here are some of potential performance overheads I ran into while implement it:
- One of design limitation you have to work on cocos2d uses a SpriteBatchNode to allow ‘sprites’ using the same texture to be drawn in a single draw. In a component system, everything is based around composition which means there is a lightweight ‘entity’ class with ‘components’ attached to it. So 1 entity = 1 draw call at the very minimum. With the older 3G devices, you are restricted to 25-30 draw calls (as a rule of thumb that is float on the Internet).
- Sharing of properties between components. While I have allowed the position property to be stored on the entity itself, other functionality specific properties aren’t. In order to do so you will have to either incur a dictionary look up (i.e. component foo = entity.lookupComponentByName(Bar)). Or you could directly link it up be assigning it to a pointer. Doing that however breaks a design principle of trying to make components loosely coupled for modularity
- Inter component communications. Besides directly looking up a component, a better way to talk to between components is to use ‘events’. For my component system, I built a addEventListerner,dispatchEventListener system similar to what Flash has. Like all events though everything is unidirectional and has a limited data payload of a single object.
In conclusion, I think for most of the newer games going with component system makes sense because once you get started you will be amazed at how everything clicks into place. If you are writing some more performance intensive, you will find components losing some of its appeal.
I just started porting Honor and Money to the Mac yesterday. Having not used Torque for almost a year, there have been tons of WTF moments trying to get the it working. Initially I wanted to do a straight port, but I figure I should also fix a lot problems while adding a few more features:
- Potential for Multi-player support. Right now the codebase makes many single player assumptions. I need to re-thinking and rework some of the scripts and engine code to make it ‘work’ again with networking support
- Move save system to something better the script files, probably sqlite.
- Cleanup the script. While I will never be free of it, much of the scripts have alot of junk in it. Same with the assets
- Migrate from 1.4 to 1.5.2 Torque Game Engine. I’m not even thinking of getting it on iOS at all, well unless people ask for it and it makes sense business wise. It’s a old engine, but the game itself has great art and graphics so shouldn’t need to do much but find a high quality splash screen logo;P
Working with Honor and Money has given me flashbacks on how Garage Games killed their own goose though but releasing multiple engine versions on the same platform. There was seldom an upgrade path(or it had a lot of fine print) and the convenient excuse of ‘You have source’ doesn’t go well while plowing through the code. A Game Engine is a cumbersome beasty even if you have written yourself already.I tried using Torque Game Engine Advanced, the Advance Engine they promised to develop for the longest time then they dropped it and started selling Torque3D which was their ‘new engine’. That was the last straw for me. I am sure they comforted themselves by saying that other people shipped stuff before on it which is OK if you can get support rather than step trace debugging through code to try to understand it.
Enough with the ranting…Here is the moment of Zen…1 level loaded…working but horrible hacked camera….
OpenID is being touted as the way forward in allowing the chaos of user identity management to brought to a manageable level. OpenID has been available for a while, but has finally started to build traction primarily because the “big boys” of the Internet have decided to get behind it. Unfortunately, what is being said and the reality are two different things.
Google, Yahoo, Microsoft have all setup their own “unique” way for other websites to allow access to their user data with each requiring their own login step (and appropriate logos) to sign in. My initial naive exception was that I could key in a username and password and I would instantly be logged into the website that supported “OpenID”. Sadly that isn’t so, and the mess is left to you to implement a login system. I won’t go into the theory behind OpenId as that is already well covered by many sites including Google, OpenID itself and by various articles. I will however include a diagram which illustrates the steps behind an login process as it is useful to keep in mind when implementing it.
For the past week or so, I have been investigating various Remoting solutions that I can implement for the Flex Games I have currently developing. The basic features that I am looking for were:
- Translation of complex data types (being able to map classes on the client and matched on the server)
- PHP supported
- Compressed data format
- Transparent RPC mapping
I initially started with AMFPHP but switched to ZendAMF. AMFPHP seems to be falling behind in support with the last release in 2008 which is decades in Internet time. Zend AMF is also backed by Adobe which is a good sign. On the side note, iPhone integration looks pretty good to with a Cocoa AMF. I would have preferred C++ but I guess it will do in pinch for now. I will be testing that next week or so.
Zend seems a bit fat, because you have to also include the Zend Framework as well. That’s something I’m going to have to figure out to see if it can add value to services I am writing.
Game developers are fairly practical in their implementations. The most practical of Design Patterns that every game developer should understand intimately are:
- Singleton – Singletons are needed but often abused as a design pattern. A singleton is basically a class with one instance. Most prior singletons were not thread-safe but with multi-processor architectures of hardware, this should be a consideration. Singletons are common used for creating loggers, database/networking access objects.
- Factories – Factories are usually coupled with singletons. Factories are often found in resource/assset managers allow managed assets to be created. The role of a factory class is to “create objects” based on a template with specific functionality or generic
- Model-View-Controller – The Model-View-Controller is a way to separate (but not necessarily simply) functionality of complex systems. A Game Engine could use the MVC design pattern to sort out the various components into logical blocks of code. MVC is used extensively in Flex development to break up monolithic applications into manageable portions. The Model contains all data used by the application. The View is what the user sees and the Controller is Application core, responding to events from the View and coordinating resources from the Model.
Right now there aren’t any good open-source development environments for Flex. Everybody seems to be using Flex Builder, which you have to pay for. For me, I like things free especially since my primary web development platform is Ubuntu 9.10. I did find AXDT which, which I feel shows the greatest promise in evolving as a free alternative to Flex Builder. Plus it helps that the primary target seems to be Ubuntu.
For more information, you can go to: www.axdt.org. I have got it working with the latest Flex 3.5. I am having some trouble with Flex 4 though but that’s still in Beta. Primarily I wanted something to allow me to edit mxml and Actionscript files not caring much about layouts. I have also been able to get it working with AMFPHP. So far it has done most of what I want.