jayderyu

The problem with clay.io is that is rather buggy at the moment (especially under accelerated canvases), it also requires a (constant) internet connection which might result in a reject by Apple, it's also another closed source library that could close tomorrow for all we know.

Cross platform I would go with openFeint at the moment.

OpenFeint is now out of buisness. I won't disagree about the bugginess. However Austin is always keen on getting "required' features and bug fixes rolled out quickly. Though I agree that the hack to get CJS working is annoying :| Clay.io does not require constant internet connection. I have a game that works off line fine. However the Clay.io services I do agree seem to be online only for now, but talks have been done to getting Clay.io to be off line as well. But Again I do not disagree that Game Centre is more robust at this time. It's just Clay.io at the moment is the only alternative to cross platform for JS gaming features :).

en.wikipedia.org/wiki/OpenFeint

Shutdown as Dec 4 2012

Jauntar

Do you think it could be something to do with the first frame rendering. Because it seems weird that an object has to move first before it's rendered.

jayderyu Game Closure looks interesting, but it doesn't support Game Center and having access to xCode gives you the freedom to add some very nice features like cloud saves that can really boost the sales of your game.

I'm now testing True Type Fonts in Ejecta, I'll keep you posted. Please update this thread with all your experiments because this product is the way to go for iOS currently.

Well I do agree that there is no Game Center support, but I'm a pretty adamant supporter that such aspect should be 100% cross platform. Though i like the robustness I would rather use Clay.io(http://clay.io/) or Hydra(https://hydra.agoragames.com/) if it ever comes with HTML5/JS support. Clay.io supports data saving to the Clay servers. So it also supports saving games to a cloud network :)

I however do agree do to extension capability it seems like more opportunity than CJS. Also as support seems to be lost in some kind of confusion it might be more reliable as it's expected as only community support.

I'm impressed. I was interested in Ejecta last month. I was planning if no option showed up for the Ouya to port Ejecta to Android/Java with the Canvas using OpenGL ES and just layer faking the Canvas calls to OGL.

However Game Closure showed up and I would rather build on top of that. As for Game Closure. You don't need a Mac. You can get working on Windows with some hacks. Also the GC team is working on preparing it for Windows. So it's just a matter of time for viability for those who don't want the hacking headache.

To the OP. I appreciate some of the work you have there. Currently I injected some cr_createRuntime with a cr_createGCRuntime(<canvas>). A customized. Your looks easier to just set the car global_ctx = .... I'll probably look into that instead of the way I did it :)

I can't specifically say for SpriteFont, but I'm aware that a number of community plugins use jQuery. It could be that we will just have to do plugin work around until the fateful day jQuery is phased out..... a long time coming.

I have C2 workign with Game Closure on the web simulator no problem. I have C2 working starting to work with GC on Android. But I received an error that looks like GC has a problem I need to get over. So until I get around the GC error I'm not sure that C2 will for sure work.

When it work I will update in the GC thread. Also I will use the bouncing gravity ball test that I posted on the web with comparisons to CJS.

However I was able to remove jquery. It wasn't needed. Thanks to Ash letting me know that it can be removed :)

Well when I was tinkering with Game Closure I ended up just not loading jQuery. The program still worked. So you can try removing jquery and see if that does anything.

Well... I think your going to have to write your own. Over at Ludei forums, Ludie is under the belief that it is being maintained by Ashley, and here the impression that the plugin is being maintained by Ludei.

However there might be a silver lining. With the new GameCenter and IAP for Apple Ash has hinted that there might be some work done on the plugin. Though I still think writing your own might be the way to go.

The answer is doable. It's also my main plan for my online web rpg I plan to do. After using C2 for a few months I can say your plan can be done. I say go for it.

hmmm. kept on tinkering. on the off chance of just not loading jquery into the app. I ended up stop getting any more C2/jquery errors :) :\ ????

So i'm testing Game Closure here and there as time permits. I've managed to get pretty far in relation to getting a BASH commands for OSX running in windows. Most of it's good to go except I've hit a hurdle

First. In the GC web simulator the game works fine. However it's the android export where I have hit the problem. I'm using the HTML5 export.

When running on Android, the game continues to believe that it's running with as DOM. The canvas is being passed to C2runtime by hacked in code. No DOM reference should be made to jquery as it's the DOM reference is breaking the game. What can I do in the C2runtime for it to adjust this. There is a lot that controls the c2runtime and it's a little over my head.

Alternatively the CocoonJS c2runtime is more ideal, but I'm not sure if the CJS calls are the only ones in there. Can I extract the CJS c2runtime and use that instead?

edit: clarifying, hopefully :D

I tried this

this.isDomFree = true; //this.isDirectCanvas || this.isCocoonJs;

but it's still using jquery and getting hung up there

Looks good.

On a technical note performance isn't dead because you using immovable objects. Immovable objects aren't in the physics simulator. They only exist in the collision checks. Also as information Bullet doesn't use brute force collision checking. It uses binary or quad tree collision detection. which means that the collision check can cut out half the objects in one pass. Then another half in a second pass. so in about 4 checks it's cut out all but just a couple of objects to check.

en.wikipedia.org/wiki/Binary_space_partitioning

so on the technical note. Your physics should only be doing collision checks on binary space partition detection.

With all this your performance shouldn't be really too bad.

Sogen, I guess by "free version" you mean Construct Classic (and not the 100 event limited trial) in the hope to turn free CC users into paid C2 ones, right? Not sure whether this could work, though it may make a little more sense than killing the C2 Free Edition which is fundamental for spreading the word about the tool...

I'm pretty sure he means C2 Trial. In another thread he states that the only model he will consider is a "profit from day one" in which there is no IAP or demo. An upfront cost only model. I would love to hear a more robust explanation to understand his business views, but his responses are terse and explain nothing but a simple thought.

I'm all up for a great discussion on marketing, maybe there is something in his years of research that backs up his belief of upfront cost only. maybe he has analysed the market and we might be in a swing trend to go back to the upfront model of software consumerism.

but nothing of such comes forward :(

However, roberto you have a point. Sciraa might want to consider pushing the CC group into the C2 and eventually stop supporting CC. Maybe a price discount if they have any games released on CC or CC games converted and running on the Scirra arcade. $10 for a simple game for price reduced or some such thing.

I would suggest using the path finding plugin to do this. It can plot the where to go based on your touch xy.