Pode I see your point, but JIT is an implementation where code gets translated to bytecode at runtime (same as what JVM does) which is sort-of compiling it.

I would like to see the benchmark you are talking about, and while they are more than possible for certain tasks, we are talking about gaming and I still see .exe files in my steam directory with small exceptions for flash and JAVA.

The problem as I said and as is pointing — is the compatibility issues. I'm a web developer and still in 2012 I have to waste time for cross browser compatibility because there is too much fragmentation and I'm not speaking of internet explorer, but also about the differences between firefox and Chrome.

HTML5 is not yet a standard and everybody is basically doing what they want. Without going too far in c2 the audio is still an issue across browsers (especially mobile).

And that is the problem right now, fragmentation, not all browsers supports webGL and when they do each one of them is doing it differently as well as not every browser supports webcams/gamepads and so on.

HTML5 will be the future for application development because everything is moving toward the web for games I'm not completely sold. Flash is becoming geared toward gaming as they lost the race for audio/video and as much as I hate him, flash has the advantage of being the same everywhere, regardless of system/browser, there is also unity3d that has the webplugin that powers game that have graphics light years ahead of html5.

In the end are the developers that steers the market and I've yet to see someone able to monetize html5 games.

It's still an open match, only time will tell who will come out as winner, that is the reason why other "game makers" allow exporting to several different platforms, because they now that they may be betting on the wrong horse..

Pode are you seriously saying that an interpreted language can be faster than bytecode?!?

Do you have sources to support your claim? I'm always open to learn something new..

Kyatric In fact I said "in theory" because in hte future it may be, but in practice have you seen my benchmark? html5 is not compiled, and it will probably never be and intepreted vs compiled will never be the same.

There will be room for improvement but we will never see bf3 on html5, at least not for a long time. This is to say that html5 will never be a viable solution for desktop gaming as games are moving towards incredibly complex (and optimized) engines that are not in the same league as anything that is made or ever will be made in html5. Example? Bastion has been made in nacl not in html5.

Pode Everything can be made in theory, but with what performances? There are effects that are hardware accelareted at driver level and webGL will probably never get them as driver manufacturers will have to code for too many cases. It's the same thing as directX vs openGL. DirectX is incredibly optimized right out of the box while with openGL there's a lot more work to do (in exchange of cross platform compatibility).

HTML5 with WebGL support is nearly as fast as native EXE games. Poor performance in HTML5 games is caused almost exclusively by no WebGL support, which is usually due to having old drivers which the browser blocks for security reasons.

Ashley

I'm sorry but this is not true at all, in theory maybe, in practice it's not at least currently.

Without going too far away, I'm converting actRact on cc and even on dx9 the difference in graphics between c2 is huge.

Also compiled application gets native effects and other neat things like distortion meshes that are almost impossible in html5.

And even without those I actually took the time to create a benchmark:

Compare the fps of this Native EXE (HGE)

To this exact same example in c2

I don't want to sound like a douche but one thing is believing in html5 (and I believe in it too) another is believing the impossible.. we have to be realists, right now performance is not even in the same league..

Awesome!! i tried the example it was slow little and also the image not alpha transparent well, when i close the game it show a problem with C2wrap.exe or something :).

Exactly the same here. win7x64

I really liked the squeeze anymation! I think that the layout is to small compared to the size of the sprites and that makes it a little confusing, you should consider making the sprites smaller (or the layout bigger).

Other than that good job!

I would like to join the ranma bandwagon and say that appmobi sucks, and I've tested it extensively, and phonegap is the best choice right now. It's basically useless apart from playing audio more easily. I really think that scirra should concentrate more on phonegap that each day gets more and more amazing plugins.

That being said amazing job ranma (BTW how did you make the audio play?),

— It's not easy to make a game run as well as furries, html5 is not a good bet right now for mobile development, especially with iPhone. Furries is not an intensive game computation wise, and ranma is very skillfull, I don't think I could accomplish what he did so mobile dev with html5 is very hit or miss. I researched and games in html5 are on the wild in the iOS appstore, but all are really really simple games, so far furries is the most complex one I've seen that actually works (the appmobi examples are amazing in theory but simply don't work and when they do they are unplayable for the lag). Bottom line think really well about what you want to accomplish, and buy construct2 in any case because it's an amazing tool, even if you'll use it just for prototyping.

I agree, you take too much damage from touching the walls.. it's a bit unbalanced.. In the "vortex" level the issue is that you can't see enough of the level to plan and it ends up being a little frustrating, the exit thing is not clear as it has two arrows that seems to say: "carry on in this direction" you should make the first level end in a "cul de sac" so the player knows where he's supposed to go.

The game is wonderful, the graphic stunning and it looks really polished, but you have to work on the mechanics because as it is it's too hard and for the wrong reasons.

My 2 cents

Wow Ranma, I think you are the first to put a c2 game on android? Amazing how did you manage to get such steady fps?

Does anyone has any ideas about including this: developers.google.com/mobile-ads-sdk/docs/android/fundamentals in a construct2 application?

That being said right now construct2 is no more than a prototyping tool.

Thats currently how i see it, but its a great prototyping tool.

The best there is!

What makes me rage is that I can't use for my games, c2 is perfect as it is right now, you can do every kind of game with it, it's the exporter the only issue. I would throw money at scirra to be able to publish game with c2 instead of Corona, but for that I guess we'll have to wait for better support.. I guess in a year mobile phone will have full html5 support, and that will be a happy day!

Havok I agree with you, but I changed my perspective a little in the last few weeks. HTML5 IS the future of gaming bar none and application wise (not gaming) is already becoming the de-facto language for cross platform deployment. That being said right now construct2 is no more than a prototyping tool. The chrome web store is joke, and I'm not only speaking about my game performance (poor, but one can say that my game sucks) but also other developers experience, take a look at these links:

techcrunch.com/2011/07/11/html5-game-maker-looks-to-transcend-sleepy-chrome-web-store-sales-on-ios

ttp://techcrunch.com/2011/01/04/sales-have-slowed-to-a-trickle-on-googles-chrome-web-store/

ttp://www.itworld.com/internet/132535/googles-chrome-web-store-slow-start

I know some of you may say that these articles are old, but speaking with other dev the situation not only didn't change but worsened. So basically right now construct2 is tied to facebook/kongregate to make money. Personally I had problem with facebook integration and I don't care for kongregate (for now html5 is too limited).

Also there are issues across browsers, especially with audio and performance is still a little limited. That being said as far as HTML5 construct2 is the best engine, there is no doubt about it, but I mantain that Scirra should start working on other exporters.

I would love to develop games on construct2 but I'm not in college anymore and I need to put food on my plate and HTML5 currently can't do this.

Money right now is in the mobile market and in steam/desura, not in facebook apps (unless you are Zynga), but even if you were, programming microtransactions in c2 would be pretty hard.

I know that Ashley hates me as I'm very vocal about this, but I don't really understand these choices, cc and c2 could be the best tools for indie development just if:

cc cross platform with openGL instead of directX or at least an implementation of xna + mono.

c2 cross platform for mobile.

With a software as easy and as powerful as construct I would easily pay 500$ for cross platform compiling..