Ran into the exact same issue when using the iframe for Kongregate. My solution was similar, but to also account for the user clicking over into chat during game play (i.e. after on start of layout) I instead went with using the browser focus action on the mouse any click event. That way when they click back into the game, it gives focus back to the game and the keys again work.

Something else to consider when designing apps that sit in an iFrame that the player might click outside of.

It depends on your idea of an MMO. If you're thinking WoW, EQ/EQ2, etc...then no. If you're thinking about the browser MMO genre like you see on Facebook then you could do something like that.

I'm working on a browser based MMO myself at the moment...it's a rewrite of my original browser professional wrestling MMO UWE. Originally it was a standard html + js front end and used Ajax to contact a php/MySQL back end. This version will use C2 to create a HTML5 front end which uses AJAX to communicate with the back end.

tl;dr - Depending on your idea of a MMO, yes is is possible.

Are you sure one would even need to save the state of a game?

Couldn't one simply use global objects (arrays, variables, etc)?

When you switch to a layout, it resets the layout to it's default position...so any spawned sprites are gone, and if you've got a player actor they'll be back at whatever position you started them at. You're going to need to remember where everything is at and then set the scene back up again when you come back to that layout.

I would tuck your sound check as an else event to event number 1126 and drop the variable all together.

Unless I'm reading it wrong, the only time the sound plays in that block is if 1126 is false.

Your psudocode there should work. If you alter a global in an event, it will be that way for later events. Work flow is linear.

Can we see a capx or screen shot?

Might want to make that >= 12 on the check incase of some weirdness that happens to cause adding more than 1 to the variable. It's a safe coding practice.

Check Sprite.CustomMovement.dx...

If it's positive, it's moving right, if it's negative it's moving left.

Inverse of isvisble added to your conditions works great when using nonvisible layers and mouse on object clicked. Yes, it's one extra check, but it's a lot less work than trying to save the state of your game and switching to another layout.

No one is arguing that HTML5 is or will ever be exactly as fast as native, just like no one would argue that Java is as fast as native machine language code on a PC...

However, advancement in the interpreter, the hardware and the developers for Java has allowed for the platform to perform so well that the consumer can't tell the difference between native compiled binary code and interpreted code. 60 fps in one is the same as 60 fps in the other. You claim HTML5 is a failure. Lots of people claimed Java was a failure as well. (there are plenty today who still hold to that...it's their opinion and they are entitled to it.) I disagree. I see it not only as the next Java, but as a superior platform to Java in almost every respect. And as hardware advances, vendors adopt and adhere to the standard better, and as the standard itself evolves, I believe it will be one of the best platforms to develop on as we move forward for the next decade or more.

Is HTML5 there yet on your mobile platform of choice? Apparently not. That's a shame. I can understand your frustration and desire to be at the tip top in that market.

But that isn't the only market served by the C2 IDE. I did not purchase my licence to develop mobile apps. I like the ability to export in a way that would let me move some of what I write to mobile, but that's not my primary use of C2. I live in a Nevada casino town, and I'd be happy to lay odds that I'm not in the minority there.

This leads us into a conundrum: You want the IDE to operate in the best way for you. I want the IDE to operate in the best way for me.

If Scirra decides to stop working on enhancing their game development tool that currently can be used across multiple platforms in it's current state and instead focus on making their IDE optimized for a one specific native language that means that only a percentage of people are getting benefit for the man hours they are putting into their product.

The fact of the matter is that C2 has a "*" next to the mobile export on the website. It's not a native compiler. I bought the product with that in mind. I want them to develop the IDE to make it even better for making games. That's what I bought. It exports to HTML5, the platform that it's advertised to export to. That's what I bought.

That is also what YOU bought. I'm sorry that you bought the product thinking it could do something that it's not billed as directly doing. Does it meet your performance goals right now? According to you, it doesn't. Does it export and run, through CocoonJs, on the platform you wish for it to run, regardless of the performance? Yes. It does.

You have gotten what you paid for. Scirra could have every right to not even try to make ANY enhancements from there, but they choose to devote a proportionality equal amount of time on your desired platform while trying to work on EVERYONE ELSE'S desired platforms as well, fix bugs, and add new game making non-platform specific features to the IDE.

Personally, I think they would best be suited on just working on IDE features to make the games in HTML5 the best they possibly can and leave any other exporting to 3rd parties. That would be spending their time to the benefit to the most people who have purchased their product, but that's my personal opinion and it is the path that would suit me the best, so I'll admit to being biased.

Actually, I'd love to see a survey of license holders as to their primary development platform. If the majority use C2 to develop for iOS, then by all means, put the time into making the iOS export as powerful as possible. If the over whelming majority do so, then I'm all for dropping all other work on the IDE (except bug fixes) to make the HTML5 export work as flawless as possible with iOS.

I do not support, however, 1st party native export. I didn't buy an iOS development tool. I bought a HTML5 development tool.

Well, if the coffee maker advertises soda making capabilities, why shouldn't I comment?

A valid point. In this case, the coffee maker in question says you can make soda by using a 3rd party carbonator and adding it to the syrup you can create in the coffee maker (which I have done, btw...)

If you are making the green bar smaller inside of an event where you picked that sprite instance it applies to just that container. If you are making the green bar smaller in an event that doesn't pick one of those sprite instances (or a particular instance of that green bar) it's going to apply it to all or them (or just the first one it finds, depending on the circumstance).

Yes, that works as well.

I had figured the problem is that you might have more than one spawner (i.e. you have a spawner off layout that you placed at design time, then you spawned another one at run time.) and it was evaluating based on the first spawner, which probably wasn't moving.