For anyone considering switching from MochiBot to Google Analytics, I wrote up my way of doing it. I'm really happy with the Dashboard I put together. Hope it helps somebody!

http://www.allworkallplay.org/blog/tracking-player-metrics-and-views-in-flash-games-with-google-analytics/

Yes, and Mochibot is shutting down along with it . Does anyone know any good alternatives for Mochibot? I currently also use Google Analytics, but it doesn't have a 'Top Host' report.

Try creating a new dashboard in GA, and then adding a table with dimension 'hostname' and metric 'views'. I think that's what you're after?

[Game released!!!]

27/03/2013 - Game released!!!

We finished the game, and it was released today! Feels like forever ago that I started it. Please give it a go if you have a spare minute and rate.

You can play on Kongregate: http://www.kongregate.com/games/funstorm/siege-knight
Or NewGrounds: http://www.newgrounds.com/portal/view/613651

Reminds me of flattr.com, but more restrictive. Flattr got a lot of attention when it was announced, and I occasionally see a button for it here and there but I don't think it has really taken off in a large way.

In case you reconsider, I'm super happy with asmallorange. I started off on their smallest plan at $35 per year (domain not included), and then moved up to their $5/month which I'm still on. http://asmallorange.com/hosting/shared/

What I like about them is excellent support from technical people who actually know what they're doing instead of just following a script. And apart from that everything just seems really well set up.. never had any issues in 2 years.

Yea, those questions all make a lot of sense... basically I only included the minimum code to illustrate my problem, and I guess it doesn't show why I'm doing it that way in the first place.

It's a multiplayer game, and messages like 'enemy X dealt Y damage to player Z with fireball', etc all arrive in the same place and I have to figure out how to show that which has already happened on the server in the client visually. Sometimes that means playing an animation and dealing damage on frame X, or shooting projectiles, or any one of many things. So that's why fireProjectile (actually called handleAttackMsg in my real code) tells the projectile what happened. Normally I do it your way, but in this instance I like the anonymous function approach because I don't want to make the projectiles smarter. I like having control over what happens in the place where the message arrives.

Anyways, now that it's working I'm going to let it sit this way for a while and see how much I still like it a month from now.

Thanks for the detailed reply! Always good to find out more about this stuff actually works instead of just guessing

I tried option #3 first, but I couldn't get it to work because TargetArray[ i ].targetId has the same problem as before... due to the for loop "i < TargetArray.length; i++" by the time it is used, i has become TargetArray.length and therefore gives a null reference error when trying to access the array entry.

I'm going to try #1 now even though I agree it's ugly. But I really really want to avoid #2 because I simplified my example a little and so I would actually have to pass ~10 pieces of data from the world to the projectile and store them, for the sole purpose of passing them back again a little later which I think is even more annoying.

[edit] #1 using a function to create the callback worked. I'm going to leave it like this for a little while and maybe revisit and try to improve the code structure at a later date. Thanks again.

Hey Chevy, I think you're right.

I want each projectile to maintain a reference to it's own target. But it appears the reference for all projectiles is updated in the loop.

Any idea on how I can achieve what I want?

I have this code, that is supposed to fire projectiles at multiple targets and on impact do things like deal damage, spawn particle effects, etc.

The problem is that the effect spawns for each enemy in the list BUT the damage is only dealt to the last target in the list.

Code:

public function fireProjectiles(Caster:Actor, TargetArray:Array, ProjectileProperties:Object, Damage:uint):void
{
    for (var i:uint = 0; i < TargetArray.length; i++)
    {
        var target:Actor = actorDictionary[TargetArray[i].targetId];
        trace(1, target.name);

        spawnProjectile(
            ProjectileProperties,
            Caster,
            target,
            function():void // on hit callback function
            {
                target.hurt(Damage); // this only happens to the last target in the target array
                trace(2, target.name);

                if (ProjectileProperties.impactEffect != null)
                {
                    spawnEffect(ProjectileProperties.impactEffect, target); // this happens for every target in the array
                    trace(3, target.name);
                }
            }
        );
    }
}

Resulting trace output:
- 1, Goblin Mercenary
- 1, Goblin Archer
- 2, Goblin Archer
- 3, Goblin Archer
- 2, Goblin Archer
- 3, Goblin Archer

Desired result:
- 1, Goblin Mercenary
- 1, Goblin Archer
- 2, Goblin Mercenary
- 3, Goblin Mercenary
- 2, Goblin Archer
- 3, Goblin Archer

So it's really weird... why is the effect spawning above each correct target even though the name is of the other target??

Can't you just use an interface? You'll have to implement hurt twice, but whatever.

Also, from having done just that - don't focus so much on having clean code. If you special case for the player here and there, its ok. Slightly less sanitary code that is there > delightfully future proof code that doesn't exist.

I hear ya, the first version I was just making things up as I go along since I wasn't sure on the design but it turned into a total clusterfuck. I simplified things to give a straightforward example, in actuality I have a lot more classes (eg destructible, npcVendor, etc) and a lot more methods (spawn, die, etc) so I was duplicating lots of methods in lots of classes. It was too messy... total nightmare that needed some serious refactoring.

I've changed it over to the 'BasePlayer as a property' approach now. Much more straightforward and less code than before.

Thanks again everyone.

Yes, I'm using plaeryIo. It's a multiplayer server that you add your game specific code to. BasePlayer is a compiled dll that lets you do things like send/receive messages to/from connected players.

Thanks for the suggestions, I'm still kinda new at this so it's really helpful.

1) "Entity -> Actor -> BasePlayer -> Player" - unfortunately I don't think I can do this because it doesn't look like I can change what BasePlayer extends

2) Making BasePlayer a property or component of player class - interesting I think it could work.

3) A suggestion I got on the playerio forums: make Actor a property of the IEntity interface by using getter/setter methods (since you can't have properties in interfaces).

Now I'm unsure which I should go with... 2 or 3? Both sound like they would work in theory and I can't really think of any important pro's or con's that make either a sure choice.

Thoughts?

I'm using C#.

Here's my code structure:

> interface Entity

> class Player : BasePlayer, Entity

> class Enemy: Entity

This allows me to do things like (pseudocode):

Code:

attack(player or enemy)

...

function attack (Entity target)
{
     // hurt the target
}

My goal: now I want to add a hurt() function to Player & Enemy so that in the above example I can replace the comment with target.hurt().

The catch: because I am using playerIo for multiplayer, Player has to extend BasePlayer. And Enemy cannot extend BasePlayer. So Player & enemy cannot extend the same class. And C# does not have multiple inheritance.

My normal approach would be to create a class Actor with function hurt and have both Player and Enemy extend it. But because of the catch, that won't work here.

Solutions? There's a couple I can think of but they are all crappy.

1) Add function hurt() to the Entity interface and then copy paste the actual hurt function into both Player & Enemy. Sucky code duplication.

2) Create a class Actor that has function hurt(). Give Player & Enemy actor as a property so then I can do player.actor.hurt() or enemy.actor.hurt(). But then I end up with a loads of if statements like "if (target is Player) then (target as Player).actor.hurt() else if (target is Enemy) then (target as Enemy).actor.hurt

3) Any suggestions?

Whenever I have looked for an artist (most recently in December 2012) I am pretty much guaranteed a message from this guy. Sometimes even multiple applications for the exact same posting if I post it on multiple websites. Seems to be one of those apply to everything without even reading it type of guys. Luckily I googled his name and came across the threads that have been linked.

Just a little thing, but you can right-click files in your flash develop project file list and select 'generate embed code'. Saves you 0.2 seconds of typing 

Apart from that I also make a class with my embedded files like everyone else is suggesting. I break it up as Art.as, Fonts.as & Audio.as. Once you reach like 100+ embedded assets it can get a little tedious, but it's really not that bad.

A different way of doing it is to create an fla file in flash, import your assets, give them an actionscript export name, export the file as swc, and include it in flashdevelop. So you would be managing your asset list using the flash library. But unfortunately this approach is not supported by flixel... if it sounds like something you want to do search the flixel forums, there are some workarounds to get it working.

Your rights are subject to the terms of the site you are uploading to. Kongregate (not sure about NewGrounds... I don't know their terms in detail) specifically forbid the copying of user content.

But we all know it happens all the time.

As a flash developer myself, I find this situation unfortunate. I wish the sites like Kong would change their terms to match reality and make it clear during the upload process that your content may be copied.

Sponsors pay for flash games because they want views. Copying games off Kong/NG -> more hosts -> more viral spread -> more views -> more value. If games didn't get copied off Kong/NG, they would get less views and be worth a little less to sponsors. So I want my games to get copied as often as possible.

[The Game:]

I desperately need a hard working artist for an online RPG I'm working on.

The Game: is about teaming up with others to conquer randomly generated dungeons. It's made in Flash. Currently you can walk around with other players, enter a basic randomly generated dungeon and use a couple abilities on enemies that fight back. It looks like this...


Deadline: I'm making this for the Epic Playerio Flash Game Contest, so the game needs to be complete by mid February.

About You: I realize I'm on a ridiculously tight schedule... the most important thing is that I need someone who can start right away and do a minimum of 10 hours / week, even throughout most of the holidays (with a day off here & there for family of course). Also please don't be a jerk.

Art Style: I am open to pixel or vector. I am open to fantasy or steam punk or your suggestion for theme. Due to the tight deadline I prefer a style that is fast to produce.. I think Realm of the Mad God is a good reference for something that looks great but also doesn't take forever to make.

Assets Required: I have kept scope to a minimum. I need...
- A basic town with a couple shops and job board for quests.
- A dungeon set that works with the procedural generation and can be recolored for increased difficulty / variation.
- 3 animated player classes. Possibly with a little customization?
- Some animated monsters that can be recolored for increased difficulty / variation.
- Attack effects.
- UI / Menus

So basically this means: 2 weeks for a town + dungeon, 2 weeks for players, 2 weeks for monsters, 2 weeks effects, 1 week UI, 1 week polish.

Compensation: An even share of revenue AND to sweeten the deal, I am also prepared to negotiate an advance on the revenue split to be paid at development milestones so you have a guaranteed minimum income. But I cannot afford much... it's coming out of my personal savings so thinks hundreds, not thousands.

About Me: I finish what I start. I work hard. I work fast. I communicate frequently. I'm a nice guy, but not a pushover. Here's some of my previous games...
- Siege Knight (just finished and being put up for sale)
- Treasure Sweeper
-


-



Interested? Questions? Feedback? Get in touch or leave a comment!

In Flash, when you right click, it opens up a little context menu that you can't disable. If for no other reasons I consider flash a bad choice for rts'es because right click is so essential to the genre. In future versions of flash player you'll be able to use right-click properly, but who knows what %age of players will have upgraded to that version by the time your game is out.

And another +1 for Dropbox. I also like to use this little task list software which I set up on my webhost http://www.mytinytodo.net/demo/ It's super duper ultra simple but I like it better than more complicated alternatives because I find the more complicated, the less people want to use it. As the programmer I also use Mercurial/TortoiseHG for version control but just for myself. It's a bit of a pain sometimes, but those rare times when I screw something up being able to revert and compare previous file versions makes it all worthwhile.

Basecamp - wow $20 per month with no free option is really pricey! I was expecting a free version limited to 2-3 people at least. Has it always been that way?

[Click for a step-by-step of how it's done.]

I didn't manage to finish my game for Ludum Dare 24, but I did end up with this little random island generator. It was my first time trying anything like this and I was under a lot of time pressure, so I was basically making it up as I went but I'm still quite happy with how it turned out. I had a hard time finding good and easy to understand articles on the subject so I wanted to share my approach in case anybody else finds themselves in the same situation one day.

Click for a step-by-step of how it's done.


If I had more time, the first thing I would improve is to add a step between 7 & 8 to curve the sections to create a more natural shape. More features such as rivers, lakes, etc and terrain types like desert, snow, etc would also be cool.

Let me know if you have any questions and I'll do my best to answer. Or if you have any suggestions on how it could be improved that would be awesome too!