The only obvious CJS advantage for Android is that it is 2.3+ whereas Intel XDK/Crosswalk is 4+ so CJS can reach more devices.

It's probably better now with lazyload/memory management with their updated CJS plugin, but I haven't tried. Before, it was horrible for large C2 games because it loaded everything on startup, taking a very long time and a lot of memory.

I have no issue with XDK, in fact, its great, my games run very well and I use a lot of particles, webgl shaders and multi dynamic faction AI:

https://play.google.com/store/apps/deta ... lite&hl=en

I would have to admit with C2/XDK, its definitely the case now that one HTML5 code-base even for a complex PC game, can run very well on Android too. The mobile version of my game is identical to the PC/MAC version in texture quality, number of units, particles and effects. As an added bonus, it also works on x86 mobile devices!

**On iOS, definitely with the new and improved CJS with JIT support will be the way to go.

Gillis and anyone else, any experience with releasing PC games into Windows 8 store, whether its worthwhile compared to going with Desura or Steam?

The engine definitely does destroy all non-global objects when switching layout, so I would say the most likely cause is you accidentally made some objects global.

Still, you can already do a 'destroy all objects' type event by putting a 'destroy' action under a 'pick all instances' event.

I noticed it sometimes doesn't destroy particles already in flight, ie. one-shot spray and rarely, sprites with a short fade out. Not sure why but i never set global for any object.

Thanks! You are a rock star! That worked.

Whats you opinion .. have arrow control (D pad) for movement, or just use touch on the screen?

I went with touch controls for movement & combat.

I really hate digital d-pad or joysticks in my space games for mobiles. It's just too clumsy and not taking advantage of the inherent advantages of the platform: Touch, poke, tap, slide, multi-touch etc.

Try it here and see how it feels for you: https://play.google.com/store/apps/deta ... lite&hl=en

Very cute, looks like it will be fun for kids.

You can use the Is Touching condition, and invert it (right click, select Invert), so it becomes Is NOT Touching (Fire Button).

[attachment=0:1ouid3uc][/attachment:1ouid3uc]

In combinations with Global Variables, you can filter the touch action to trigger a specific effect.

Nicely done.

Have you thought about using a fullscreen capture from C2 in Chrome Preview for screenshots? It'll be a lot clearer as well as larger.

Going into this path as a gamedev, I felt that a good game will sell itself (but it may take some time to gain traction).

I still feel this way.

Here's an interesting current breakdown from small time gamedevs:

http://www.reddit.com/r/gamedev/comment ... fessional/

http://www.reddit.com/r/gamedev/comment ... developer/

Other successful indie devs I have spoken with have always come to the same conclusion: do not pay for reviews, do not buy fake downloads/ratings, don't waste money advertising on browser websites if its for a mobile game (because click-through rate is abysmal).

The dos: do contact game press and youtube reviews with a alpha/beta build, some will be interested and play it, get their feedback and improve upon it. They are all gamers and if they see their feedback is implemented, they will respect you. If you have $, do spend a small amount on in-game advertising that leads to direct downloads, its a much more effective form of advertising for mobile games. If its not a new or unique game, make damn sure its GREAT to stand a chance.

There's probably a lot more, I am still learning as I go. But there are a lot of devblogs from successful small indie groups or solo devs.

Just to update this since the thread is quite old.

When I posted my response, I was polishing my spacesim Star Nomad. I struggled to get a good combat AI with fluid dog-fight/swarm movement for a long time, I even posted on the forums and a few responded with inputs but it wasn't effective so I gave up and made it simpler.

Recently I started work on a sequel, and it took me 5 minutes of problem solving/logic and a few lines of events and I have a perfect swarm AI, it looks exactly like the Starcraft Protoss Carrier launching drones to attack and peel off then re-attack and was what I had wanted.

As you use C2 more, you learn a lot of clever ways to solve problems and you will get better at it.

UID is work-able.

newt Thanks.

6000 Sounds about spot on. What pushed me over the edge was offering my game free on Desura for one day and getting a reddit chain going. I had 20,000 visitors on desura that day! Carefully link to the Desura Game and the Greenlight Page in large lettering if you plan on doing this. If I would have let that campaign run for 48 hours I think I could have gotten all the votes I would have needed in two days. Good luck!! Really love your game.

I also heard that once you are successful with Greenlight and publish your game on Steam, future games can be released directly onto Steam without going through Greenlight again... a few other indie devs have confirmed it but others have said it's "not guaranteed" and "its up to valve".

Valve is very non-transparent so we have no way of knowing. But if true, it's a huge incentive to basically give out your first Greenlight game for free or next to nothing (in bundles) just to pass, then focus on making $ with the next games.

Also, thanks for voting yes on Star Nomad.

ie. A condition or filter that identifies the current target of a specific turret.

The one easy way I can think of, is to spawn a sprite line on target acquired and do a condition of line overlap/colliding with the target to toggle a target variable (isTarget = 0/1). But that would get extremely messy with more turrets.

Hmmm....