Kriegsland - Development Blog

Vaadin, re-building the Dream

2011-09-04T10:17:00.000-07:00

Kriegsland hasn't got much love in a long time. The game I mentioned earlier has likewise been left as is, people laughed at the legion of crabs concept. That, and my design was a little inflexible. I disregarded all OOD for the sake of performance but I ended up with one massive file which made adding new features a headache. I may come back to it later but for now look at this: http://vaadin.com/home

Vaadin is a brilliant new tech which provides a complete infrastructure for a Servlet based GWT Rich Internet Application. Now what that means is you can make a really awesome looking web based game for little to no real effort on the infrastructure side. That's the theory anyway. I've been working on some sample apps and it seems great so far, it shares all of the easy UI generation of GWT alongside a lot of built in visual themes and it manages things like the Session for you.

I'll report back with some specific examples but it really seems like technology is starting to catch up with my ideas.

Finally, Something Vaguely Interesting

2011-03-18T06:36:00.000-07:00

Okay, so here's something to report finally.

Out of all my projects one has got to at least a vaguely interesting stage. Ironically it is almost entirely contained in a single file.

It's a cool little Android game I'm working on. It's been really interesting so far, I haven't really gone beyond drawing guys to the screen and some basic enemy behavior but hell it's kinda fun.

So, I'll follow that for a while. At the moment I've just got basic touch based movement and some guys who spawn and run across the screen but after I've added a few new enemies it should get more interesting.

Here's a screenshot from the emulator. More to follow!

The Winding Technical Road

2011-02-14T12:46:00.000-08:00

I've been incredibly unfocused in this blog. It was supposed to be originally about a single project but since then has wandered in a large number of directions. Of course that's fine, that's how technology should be.

So, I think it might be more suitable to start again with that as the focus of my blog. I'd like to write meaningful things about technology, useful things that people can use. Stuff that you might google and end up saving you a day of figuring something undocumented out.

I'll be an architect soon, my job is promoting me. And I realize that a lot of things I have covered here is not up to scratch as it's mostly borderline propaganda and uneducated vague outlines of things. It's not much use to anybody. My new position should help me build up some useful knowledge however and I fully intend to share them with the world in an effective way.

As for this blog, well, Kriegsland was reborn in my mind recently as yet another entirely different game. And I will make it. Eventually. Right now I feel my time is best spent learning, not creating and I'm going to have to accept that. I have too many unfinished projects. I think in time I will finish them all but it isn't helping that I keep adding things to the list.

So! Hopefully you'll see a new and useful technology Blog starting up soon and watch this space, my projects can never die; THEY'RE UNDEAD.

The Quest for the holy GRAILS

2010-09-04T04:32:00.000-07:00

It's been five months since I was burned by JavaFX. Five months since sun taught me to hate new technology and the time have finally come to balance out this blog with something much more positive.

It started in the interview for the job I've just started. (Hello from Luxembourg by the way.) One of their technical head honchos mentioned GRAILS. It wasn't the first time I'd heard of it from a senior person in an impressive financial institution either. One of the greatest Irish technical success stories, Realex payments, use it too.

Well, I thought, two out of two places can't be wrong. I decided to check it out for myself. I promptly ordered a book on Groovy and a book on Grails and got stuck in.

Groovy! Not just an Evil Dead 2 quote any more: Oh man, Groovy, what a name. I love it, but the greatness of course goes much deeper than that. Groovy is for the most part a superset of Java even compiling to code that is essentially Java byte code. So what's the difference?

Ever think to yourself in Java, "Why do I have to type this stuff out every time?", Groovy has done that for you. Null protection is achieved by adding a question mark the your reference, Strings are now GStrings (Great name) which can contain externally defined properties. Want to define a populated list object?

def list = ["Hello", "I'm", "a", "list"]

I'm still learning but every time I read something in my Groovy book I laugh out loud like a super-nerd and say to myself "did these guys read my mind when they were writing this?"

GRAILS, it's about time: I've tried my share of web frameworks, definitely not all of them but one or two. There always is something that turns me off, mainly I suppose because I'm a newbie to the web world and everything I've tried so far assumed a lot of starting knowledge and it could be a week before you got anything to look at, if you were lucky. Not so with GRAILS.

No kidding, I had a simple database, a controller, a service layer and an automatically generated CRUD screen together in maybe thirty minutes? I could kick this off and play with it in my browser. (Oh, and there's a jetty webserver integrated with GRAILS, didn't even have to set one up.)

It took a long time to get to the stage where this was possible in any language but even better this is all built on Groovy so every feature it contains is available on every level of your application. Configure your program on the fly, open your grails console and write code that calls subsystems you want to try out.

Too good to be true? Well, I can see why people are using it. Not just your startup up the road with more ambition than sense. GRAILS takes everything Spring (the EJB killer) has and makes it even better. Enabling you to use Groovy out of the box puts it significantly beyond EJB3.0 for me.

My current project will be GRAILS. (The prototype was in EJB3.0). I'm looking forward to getting to know it better.

Java FX is a steaming pile of unusable crap

2010-04-03T12:31:00.000-07:00

Yes, harsh words but after having a quick go at it I am left with an incredibly bad taste in my mouth. I was trying to build a nice simple web app, just a few text boxes, nothing fancy, well, this is how my Java FX journey went.

First stop was the website looking for examples. http://javafx.com/

This is the only website I could find that was any sort of resource for the language so I went through any examples I could find, any tutorials etc. Well, the resources are truly pathetic. The examples are poorly laid out and poorly written and the website itself is thoroughly amateurish. Forget about comparing it to something Adobe has created, it doesn't stack up to student projects that I've seen put together in an evening using FRAMES.

But I could look past the sub-amateurish look if the resources were any good, but I couldn't make any progress. Now, Sun are bad for this in general but the thing about Java is that there are hundreds of resources available for learners, JavaFX has sweet FA which made this website so important.

Well, I did what it said anyway, I installed the eclipse plugin and did what it told me to do. Now, the language itself despite being called "JAVA" FX looks absolutely nothing like it. It reminded me of VB actually. Yes, VB... VB 6.0 in fact...

So, despite having an eclipse plugin there is no visual component so you have to run it every time to see what the hell you are making. This involves eclipse jarring everything up or whatever well, by this stage I was incredibly annoyed. I also wasn't sure if debugging would even work and all "tutorials" I could find about blending JavaFX with vanilla Java were freaking awful. One in particular was titled "How to use Java objects in a JavaFX application" but actually only showed how to do the opposite.

The majority of the tutorials were actually about how to make a ball bounce around the screen and things like that. How many people want JavaFX for that purpose? Maybe students? Is that the only group of people JavaFX has been created for? Students looking to make incredibly bad looking graphical projects?

Well, that was the end of the line. Sun, you suck at this stuff give up.

Oh, and, changing the subject slightly. Oracle, the big fat stupid tech support guy has bought Sun the incredibly Nerdy guy who can code like hell but expresses himself as well as a cabbage. This is not a good team up guys.

Sun's certification website for example has been consumed by Oracles one which is a massive mess of rubbish that takes forever to navigate. Look.

Tech support people would be the first to admit it, they deserve all the crap because they spend all day looking at status bars while sitting on comfy chairs. Developers don't deserve this, we have enough to worry about for god's sake.

Oracle, go to hell and take Java FX with you. Good night.

Building everything back up

2010-03-27T05:27:00.000-07:00

I did learn a lot with PHP and reading my new JEE web services book I happily realized that I'm already comfortable with many of the concepts. POST, GET, etc.

In terms of work on the game at the moment I'm doing the painful part first and getting the HTML together. I'll be honest, I hate that stuff. HTML + CSS is often completely backwards and things get thrown around the browser window often with little rhyme or reason. Of course, that's even before you take into consideration all the different browser "quirks".

So I just want to get a template done and dusted and never have to look at it again, at least not in this iteration. It's not nearly as cool as playing with EJB3.0 or JBOSS but sometimes this stuff is just necessary.

Finding a Base for the Boss

2010-03-04T13:48:00.001-08:00

I've been planning for some time now to set up a proper server to host JBOSS, it was going to be guts of current machine tied up to the internet using no-ip.com to make up for the fact that I don't have a static-ip. The question I'm asking myself now however, is that still the way to go?

Hosting has become pretty cheap. I'm already with Hostmonster which basically provide everything except Java support, for hardly anything a month I could host a PHP site with unlimited databases and what have you. The thing about Java you see is that need a lot of memory, making hosting more expensive. Also in a shared environment you can forget about being able to configure everything, and if somebody is hogging all the memory, well, you're screwed.

But things have changed a bit now, which brings me onto my little guide. Finding a base for your JBOSS. I'll talk you through some of the features to look out for in a JBOSS hosting provider.

VPN or Dedicated: A VPN is essentially a virtual server and also the reason having your own hosted JBOSS server is actually something me and you, the plebs, can even consider. Dedicated solutions (i.e. having your own machine in somebody else's server farm) has always been prohibitively expensive for the likes of us. With the wonders of virtualisation we can have the benefits of having our own machine at only a fraction of the cost.

Do I get JBOSS: A bit of a no-brainer maybe but every provider seems to have different deals. For example you could be signing up for Tomcat hosting and only get a shared instance of Tomcat and no ability to use JBOSS at all. Make sure your deal explicitly states that you are getting your own instance of JBOSS.

Do I get a Shell: Moving files around, reading logs, checking on your database, the list goes on. Don't underestimate the power of having shell access to your JBOSS machine, it is definitely something you will regret not having later.

Memory: I'm writing this because I've noticed some hosts offering "JBOSS" hosting with only 128mb of memory. If you try to run JBOSS with this much memory it will laugh in your face before either spontaneously combusting or spending the next four hours crawling on the floor. You need at least 750 megs if you want to be safe, though you could probably get by with 500.

OS: Everything I looked at so far offered a selection of linuxes to choose from. If you have a preference you might be out of luck, Fedora 12, CentOS and Ubuntu seem to be the most popular. Since all you probably care about is JBOSS I might recommend going with CentOS as it's based off the same source Enterprise grade Red Hat linux. (Red Hat own JBOSS since a few years ago, if it'll work with anything it will be their OS.)

So hopefully that helped you out a bit. As for who I will be going with my self, the best I've found so far is this outfit. JSP Zone. They seem to have by far the best package for the money they are asking for. If you find somewhere better please do leave a comment, preferably before I take out my credit card.

The Plan as it Stands

2010-02-15T05:41:00.000-08:00

The plan as it stands is this. I'll create a generic "Krieg Engine", this will essentially take the form of a java based sub system which will eventually be thrown into a nice reusable jar. The idea occurred to me that I could use the same combat engine in my Zombie game, this would provide lots of great testing and also speed up the development of that game a lot. So more thought will definitely be going into that. If I can create a nice game engine based on slick and a nice combat engine from Kriegsland it would be two major modules that would make future game development a lot faster and a lot more fun.

The krieg engine itself will be a highly detailed combat engine. The idea is essentially something like Heroes of Might and Magic 3 but instead of just applying an aggregated damage amount individual units will be matched off and will each have unique weapons. Detailed output will be generated from every encounter allowing it to be formatted into a nice message for a web based user or even to be visualised in a game like Creeping dead. Well, that's the long term plan anyway.

I've been through so many failed projects now, I really feel like I am learning a thing or two. It really is not straight forward to achieve even the smallest things in a project. I'm certainly not a bedroom coder, at least not yet, though I now grasp the immense dedication and self control that such people have. What really drives me is the constant stream of ideas I have, it would seem such a shame to let them all go without doing something about it.

Well, once more if you feel like you could contribute to the project please contact me. Otherwise thanks for reading and click my links!

A Technical Roundabout

2010-02-10T11:52:00.001-08:00

Massive inaction from me for the past while. I started dedicating my time to a project with a friend, it was to be a JBOSS based application that would handle a massive government archive. It kind of hit the rocks however when my friend caved. But never mind I learned a lot.

Now I'm back onto Kriegsland, I think. Ideally I'd like to move everything onto JBOSS/JSP or, even better, JBOSS/FLEX but JBOSS hosting seems prohibitively expensive. EJB3.0 could be worth it however, everything is just so much easier in a managed container scenario. While the initial learning curve is high I think there is definite payoff to be had.

Anyway, I'm going to re-evaluate the whole thing now. JBOSS would handle the heavy logic required for a game like Kriegsland beautifully and the development tools would be far superior than PHP. So, I believe I will look for a decent JBOSS host and move away from PHP. It was a fun diversion but I'm SUN CERTIFIED now and I have a head full of enterprise knowledge.

Java Mad

2009-12-07T15:56:00.000-08:00

I've really been in gear lately, productivity on the coding side of things has been way up. Unfortunately for Kriegsland however, not much of this productivity has come its way.

I've just got my SCJP qualification and as a result have been Java crazy for the past few weeks, building a Slick2D based game which my brother should be doing the art for and as I'm speaking I'm downloading the needed components for a go at an EJB3.0 driven JBOSS based application. God knows how that will turn out but I'm bloody excited about it right now.

JBOSS is where I started in a way, my fourth year project was a SEAM implementation and my 6 Month internship had me working with it without me having a clue what was going on. I really want to return there now and finally understand it all implicitly.

So, sadly Kriegsland is on the back burner, and my reflection based implementation will have to wait for the moment. I did have an idea for a Slick2D based client tying back into my MySQL database with PHP providing an access anywhere front end, much like Pox Nora has but that'll have to wait until my Slick2D engine's completion which won't be for a while. The game I am working on is zombie based and will have a strategic map much like Total War, I've got clicking and dragging working and everything.

So I might post here about Slick2D in the near future, it really is a fantastic little library and it's been a joy to use so far. Its website. Until then, hold tight Kriegsland fans, this project can never die! It's undead!

Obstructive Abstraction (Layers)

2009-11-11T15:35:00.000-08:00

I'm nearing the next level of Kriegsland, the work on the underlying engine. It's now I'm considering everything and anything that will simplify and stream-line the process. Something that I've looked at before and I now find myself poring over again is the PDO based ORM and abstraction tool Doctrine.

Doctrine is very similar to Hibernate in that it creates a load of objects that allows object orientated access to your database tables and also creates a layer of abstraction over your database.

So, pros and cons.

Pros:

1) More efficient use of resources (probably). This is something that will probably really only have returns in much larger projects in that the actual overhead of running Doctrine will probably be higher than any saving that dynamic loading of objects etc. could provide.

2) It's the "proper" way of doing it. Fully implemented it would be solid, robust, easy to expand and build on top of. Additionally, if I wanted to switch to a different database in 5 years time it would be drastically simplified.

Cons:

1) Time. It's a new framework to learn and though it would streamline elements of my work I'm sure it would add a great deal of complexity to getting anything working at all.

2) Is it really necessary? This is a basic enough project, my first of this type. Is being tied to a single database such a bad thing? I have very solid SQL and am a wizard with outer joins and such so why throw that all away just to go with Doctrine.

So, after thinking about it I will postpone my inclusion of Doctrine in my project. As long as I keep things properly layered and implement some kind of DAO library it shouldn't be overly problematic (famous last words) to slot it in at a later date if I so choose anyway.

Sucker for Stats

2009-11-07T06:19:00.001-08:00

Well, subversion has provided a cool little diagram of my continuing efforts. I now have a screen where you can buy units that are tied to your army type and got around to figuring out outer joins in MySQL so you can view units which don't currently have items equipped.

My current approach is just to get the functionality in there and focus on making it more robust/prettier at a later stage. So far it's been going pretty well and I'm knocking items off my list at a pretty consistent rate.

Now I have it so a player can register themselves, buy a unit (though of course there's no limit on this yet but it's fine for testing purposes), then move their army around a strategic map. All these need expanding and prettifying but the underlying data is all there and it's time to look into the real meat of the game. The combat, the war, the krieg engine.

This is going to take the form of an entirely object orientated php program that can be executed both on the command line (For later cron-script integration) and from the kriegsland website front end.

Awwwwwww yeah.

Design Process - Poor Bloody Infantry

2009-10-25T16:35:00.001-07:00

Taking a break from the technical side I'd like to write a bit into some of the underlying concepts of Kriegsland. First I'll work through the four army types available, the first of which is infantry.

These are the aspects I hope to capture in the game engine.

Numbers - This will be the biggest advantage Infantry units have. They will all number ten to a squad so compared to a similarly equipped special operations unit they will have twice the firepower and take twice as much in return to take down.

The downside to this will be their vulnerability to area effect weapons but sometimes you just can't beat having more guns on your side.

Field Adaptability - I'd like to see the infantry type armies being adaptable to make up for their lack of tactical mobility. They will be hunted by vehicles and special ops armed specifically for their destruction so they had better be equipped to deal with them to some degree.

Firepower - Infantry can be equipped with heavier weapons, killing their mobility but allowing them to take on enemies at severe odds and win. I'd like to see mobile anti-tank guns and heavy machine guns make an appearance, though I am sure ammo will be scarce.

Infantry will be the superior defenders of Kriegsland, perfect for holding an area with heavier weapons. Their natural enemy will be artillery armies, able to take them out from range with high yield explosives. Special operations and Armor armies will have to be careful, assaulting infantry is always dangerous and even when on the offensive suitably equipped infantry can end ones journey in the Kriegsland all too prematurely.

Two unit concepts I have come up with for this unit type are as follows.

Scavengers

Scavengers are freshly hired soldiers with little to no battle experience. As such they are extremely poorly equipped, often entering battle with improvised weapons. However, their desperation to acquire new weapons and armour proves a huge asset as they will risk life and limb to get their hands on a decent piece of technology. If the technology is too good it may not be possible to part them from it however.

Scavengers have very low stats but whenever they are used in battle there is a chance that they will return with a piece of low grade technology. There is also a slight chance they will acquire a high grade technology but this will be tied to them permanently and they will no longer acquire any more technology.

Equites

Not all weapons were successfully destroyed by human society. Some family heirlooms were hidden away by rich families. Additionally, the new rich of human society have the means to acquire some equipment and training before entering the kriegsland, often by dubious means.

Equites have decent starting stats and come equipped with a basic grade technology. There is a chance that they will receive a “package from mummy.” Which will contain a random low grade technology.

This is the kind of thing I'd love an art team for. If you'd be interested of having a go at drawing these guys I'd love to see it! Otherwise it's time for me to brush up on my Photoshop.

Next time, Special Ops.

The work so Far

2009-10-23T12:07:00.000-07:00

Alright, so if you are at all interested the latest version of kriegsland is now online. Check it out!

It's messy, but you can do the following things so far.

1) Create a new user account
2) Log into that account
3) Move your newly created army around the strategic map

Everything is set up in a nice expandable way in the database and it feels nice and robust so far. Next will be setting up a way for armies to face off when they're in the same area against the indigenous population, or more likely a method of building your army.

Oh, by the way, the strategic map is a jpeg generated on the fly based on a locations co-ordinates that are fetched from the database. Neat eh?

One point, if there are any PHP heads reading this who have any advice or fancy tricks please leave a comment. Particularily if you know a way where I can remove that aliasing from the combined jpeg. (The cloudy effect looks pretty cool but if I'm to have it I'd prefer if it were by choice.)

PHP - From a Java Perspective

2009-10-19T09:32:00.001-07:00

Now, I'm just starting PHP. I acknowledge that but it might be interesting for those who have never touched on the language before to hear my early impressions of it.

First of all, coming from Java it was very easy to adapt to the PHP syntax. It's extremely similar for the most part and I've gotten on well with the compiler so far, it's been intuitive all the way through. One thing that I'm not sure about at this stage is the IDE however. I'm currently using Dreamweaver to write my code, which feels a bit strange. I was using PHPEclipse before, which offered much more features but I'm finding Dreamweaver friendlier for now. It's also superior for when you are laying out tables and such in tandem with your code. (In fairness, these are probably two pass times that should be kept separate but this is what is working for me right now.)

Database stuff has also been very straight forward, dealing with the resultant arrays and collections, fine. I can't say anything within the language was cumbersome. I have also worked with the graphics manipulation library (GD) and it's powerful and easy to use. No complaints here either.

One part that I'm yet to fully throw myself into however is OO in PHP. So far I have an object handling my session logic but other than that I haven't gone into it. It's something that I'll definitely be using in the back end calculations. (Going into that stuff procedurally would take years off my life.) Anywho, while the manner in which you must access object variables seems a bit funky looking nothing seems too far out there.

It might be time to get my PHP book back off my friend... But all in all PHP is a damn pleasant language to work with, so far.

Subversion, the peoples choice of source control

2009-09-22T06:32:00.000-07:00

I've finally gotten around to sticking my work onto an svn repository, something I'd encourage anybody working on any project to do.

A big part of any project of any size is source control and anybody in the know will tell you that subversion is the only way to go. Coupled with the tortoise shell it's a comprehensive source control utility suitable for amateurs and professionals alike.

In the last few years the amount of online SVN hosts has exploded. With these people you can usually get cheap hosting with plenty of space for your project and without the bother of setting up everything yourself. By now there is a long list of providers to choose from. At the moment I'm giving Code Spaces a try and am liking it so far. But it looks like most of the providers have similar deals so you might as well shop around and see what suits you best.

Once you've got a host sorted it is then just a case of downloading a program called tortoisesvn. This actually integrates itself with your windows shell so committing and updating your files from the repository is always only a right click away.

I'll show you how to do it just in case you haven't used it before. First of all, you should have a repository address. In the case of Code Space it will be https://svn.codespace.com/[your company]/[Your project].

Then, after installing tortoisesvn just right click wherever you want your new workspace to be and click SVN checkout.

Then it's just as simple as pasting in the repository address and clicking okay. You'll be prompted for a user name and password but as long as you click "remember my details" this is the only time you'll ever need to enter them.

Then paste everything you want backed up into the new folder and right click again and click commit, select all and click okay. You'll be prompted to enter a message explaining your commit and then that will be it. Everything will be backed up somewhere remote. Your kid brother can try as hard as he likes to delete every trace of your work but it'll be backed up nice and safe where he can't get at it.

SVN is also great in that it saves every version of you work, so if you regret removing something later down the line or want to track down the code change that introduced a particular bug you can hunt it down efficiently.

Kriegsland is safe now.

PHP - SimpleXML forever

2009-09-20T09:09:00.000-07:00

Part of the Kriegsland game is to be a modular customizable rule system. Now, I have a lot ideas for this game and the prospect of hard-coding the entire rule system seemed a bit short sighted. I don't want to end up with pages and pages of spaghetti code down the road.

So instead I've thought about the extent of the rule systems power. Based on stats it will generate a random variable which will determine the success or failure of a possible action based on its probability which will be predetermined by the specific weighting of the various variables.

So... Yeah, if you got through that well done. It's a bit to get ones head around but I think I've almost cracked it. I'm going to write the engine once and then have an easily editable xml file to fine tune until I'm happy with the rule system.

Which brings me to programmatic XML handling. There are basically two types, event based (Eats XML piece by piece to avoid a large memory footprint) and DOM based (Eats the entire XML file and creates a nice file to play with).

I've looked around for a nice DOM solution in Java and C-Sharp but was actually having trouble finding something that fit the bill. Eventually on my return to PHP I've found SimpleXML. Sample code here.

// displays contents of nodes
if(!$xml=simplexml_load_file('rules.xml')){
trigger_error('Error reading XML file',E_USER_ERROR);
}
echo 'Displaying rule names of XML file...
';
foreach($xml as $rule){
echo 'Name: '.$rule->name.'
';
}

?>

Simple as that! The elements of the XML file are all exposed as arrays. If the rule element given as an example above had multiple elements of type name you could create another loop like this.

echo 'Displaying rule names of XML file...
';
foreach($xml as $rule){

foreach($rule->name as $name) {
echo 'Name: '.$name.'
';
}

}

Great, simple and short. I'm going with this. The rule system will have a single instance kicked off by a cron-job so the memory footprint isn't an issue. My Plan is now to create a unified Rules object that can handle the parsing of the XML and then act as a DAO for the game engine.

Good fun anyway.

Overkill?

2009-09-16T07:17:00.000-07:00

I decided to start a separate blog to chronicle whatever work I get done on my Kriegsland website, more as an incentive for myself to continue than anything else.

My work has since brought me into contact with an interesting cross section of available technologies and I think one's quest to set up a new game web-site is probably an interesting enough topic to read about.

I'm still pretty much at the proof of concept stage and I have a very very basic php website together. It's possible to log on and it stores your session, it also fetches all of the update news from a database. Let's have a look at what it's built from.

PHP: This is the incredibly popular language for building modern dynamic websites with. It's competitors would be ASP.NET (expensive) and perhaps Java Servlets (A bit unwieldy, also requires an application server). PHP in my opinion wins out easily as it's free and simple to use.

MySQL: I suppose this almost classes as a no-brainer, MySQL is a modern fully featured and free transactional database. Very often used in commercial applications and well tried and tested. One place where it falls down is support but if you mind what you are doing you should never need any for a project of this size.

I also toyed with the concept of developing an off-line "simulator app" for the fine tuning of the combat engine. I looked into C# for this and while it has streamlined many of the tasks required things got messy when I was looking into a way to transfer XML into objects. There didn't seem to be anything that automated the process available and that, in addition to the pointless duplication of effort between C# and PHP meant that I decided not to go down that road.

And that's pretty much where I am now. I'll post next time something interesting comes up, it could be a while but that's the nature of the beast.