Timeframe for EXE exporter?

Not to sound like an arse, but what I read in many of these comments basically sounds like: "This 20 pound product better let me make my epic AAA product to be sold on Steam, dammit! Also, yesterday!".

I think a little perspective is in order. Not to talk down on C2, it's awesome and a lot is already possible (the contest was a great demonstration of this), but it's still in early stages and the tech and support in browsers is a bit early, so to kind-of lock yourself in with an exe exporter would seem like an additional waste of resources, since it'd need to be updated and supported all the time.

Not to sound like an arse, but what I read in many of these comments basically sounds like: "This 20 pound product better let me make my epic AAA product to be sold on Steam, dammit! Also, yesterday!".

Just to inform you: My "product" is not AAA and it's free. It's just a big JRPG, that will weight quite a bit. I expect it to hit the 1GB mark, but not sure.

I have never talked down on C2, there's a reason I want to work with it, and that reason sure as hell is NOT that it sucks, even if it's only good for web games. But since it doesn't have an EXE exporter, then I cannot. Simple, no mystery.

The main reason anyone says that it's not good for big projects, is because it can have it's drawback if you lack experience with it.

You are more or less going to run into some problem or another, usually with the way something works (or doesn't work) and have to find creative ways around the problem. It is nearly always possible, and more often than not there are people here to help.

I guess, be prepared to do a little learning with regards to "what not to do", and you can really make anything.

Ideally start small, then see what can be accomplished. Maybe using CC as a learning platform may not be such a bad idea, since C2 works more-or-less the same way... it won't be "wasted training".

I suppose it wouldn't hurt to give it a try. It's just I am undecided if I should use CC or wait for something that may never come.

Well one thing for sure, if it comes, it will come in some time (a year, maybe two, nothing sure as stated before).

You could as well "make your hand" on some lighter project (sort of prototype/working project to show for once done) with CC.

What you will learn from it will be useful anyway in C2.

I also don't know what is your prior experience in game making but fact is that ambitious projects (like 1Go of datas by a single person ?) takes time, and 99% of them never see the light of day in the end.

Go progressivly with lighter versions in CC, raising the amount of mechanics/features and datas in CC until you hit said quirks and are forced to workaround them.

Sure the projects won't be compatible, meaning that if/when C2 has exe exporter, you will have to remake most of it.

But in the mean time you will have gained valuable experience and already founded a user base that will know and support your game.

I think it is the most reasonable choice in the end.

Another thing I think is that by the time you will have made 1Go of datas and put them in a working project, computers and HTML5 in general should be able to handle it, high bandwith internet connexions shall even be more available through the entire world.

It is your call that HTML5/C2 can't handle it, but I'm confident it already could.

The major problem would be the download time, problem that should be fixed if/when a "download datas for layout" (orsomething of the like) feature is added to the current HTML5 exporter.

Will a web browser be able to play a huge open-world RPG, over a gigabyte in file size, aimed at computers with a minimum of 512mb of VRAM on their GPU

Wow. That's a crazy big game for a tool like Construct. It's so big, an EXE exporter probably won't save you. You probably will have to use traditional programming tools to pull that off as an EXE, it's definitely far, far beyond what Classic can do.

We aim for C2 to be able to make big games, but wow, I thought 100mb was about as big as anyone would ever get, ever.

For example, if HTML5 performance gets good enough that an EXE exporter can only get a 10% performance edge, then an EXE wrapper around the HTML5 runtime should be perfectly adequate (possibly with a plugin to access Steam API features).

Is making an EXE wrapper much more feasible than a full exporter? And if so, are you more willing to work that in more quickly?

Again, I'm just trying to decide where to go from here. There aren't many places outside of Steam that are suited for products that aren't simple and casual, so if it doesn't look like I can get my project onto Steam, I'll put it aside and work on something else until I can. No pressure or anything, and I don't wanna sound like I'm attacking you, but... this stuff is important for us devs to know, y'know? <img src="smileys/smiley5.gif" border="0" align="middle" />

I was read this topic carefully, and imho many users can not understand

restrictions of any game constructor.

Hey, look, many favourite and profitable games was made with fewer resources. Look at "Angry Birds" or "Worms!". You definitely can do same things by using Construct-2. For example it's possible even in GM (like this yoyogames.com/news/27). Construct-2 is definitely better even now, in beta stage.

You can create a masterpiece even with good 2D graph.

But why so many people try to make new MMORPG or another big projects without any small but look professional games in portfolio?

I remember 199x years, with Atari and ZX-Spectrum, when we had only 48k RAM and people do great games on ASM or C.

Why are you talking about Steam? Steam publishing required professionally made games. Even games on Steam made in Torque or another engine (not constructor!) was polished by "hands" with editing engine source code and etc.

Construct-2 is great thing to TEST your ideas. It perfect way to make your portfolio for future presentations (your projects or yourself as game designer). Without small but successful project *no one* will be talk with you about publishing on Steam or in any another serious place.

With Construct you can test your gameplay ideas. You can make simple presentation of future project for your dev team or investors. You can learn/develop some skills needed to build bigger and more complex games.

but if you want to make 1Gb RPG...correct me, if I wrong, but I am sure you need to lear C++ or C# and start using something from XNA framework to UDK SDK.

p.s. Our industry (game dev) need more good ideas, not programmers. We have *many* talented programmers, but we have only 1 or 2 really attractive gameplay ideas per year. This is why people sometimes play in 10+ years old games like Diablo-1 or "Little Big Adventure". This is why sites like gog.com are still profitable.

Come on, people still download and play on ZX-Spectrum emulators with 256 colors games!

Any user can made a new MOD on CryENGINE-3. But make a talented text MUD is MUCH harder

I'd just like to point out that there's no such thing as a gig game.

Anything that big would have an exe that's some mere megabytes, and the rest is resources.

I could see why they stick with HTML5 exporter however I have some questions, is it possible to play HTML5 offline? I have a laptop and I bring it to places where there is no wifi or free wifi in some cases, so it will be impossible to play HTML5 games online in such cases.

So what I wonder is, will it be possible to play Construct-made HTML5 games offline?

Adobe flash can make flash files to be played offline by exporting them as .swf files and run them without internet connection, so this is the only thing that is bothering me with HTML5.

if it is possible to play HTML5 offline then I am happy with that, I could see future games to be played in a browser. (anybody heard of Quake Live? it is one of the experimental games that run in a browser and so far it's going good)

So what I wonder is, will it be possible to play Construct-made HTML5 games offline?

This feature has been available for two releases now (since r63).

I haven't been following Construct releases for a long time, thanks for confirming that.

This shouldn't be read as a rant against html5 or me screaming for an exe exporter. I'm actually very pleased with c2, and am willing to fall back to cc if need be...

But of all the features that may or may not be implemented in the near or distant future, I'd have to say an exe exporter is far and away the most appealing to me.

Even for free games, even for casual games. Yes, browsers are convenient and good for many types of games - but I cannot get around the fact that I find them simply annoying. They get in the way of the game. Maybe other people like playing games in tiny windows while they browse facebook, but not me. The first thing I do when I find a good browser based game, is look for a way to download it and play it in fullscreen.

(this is usually doable in flash, but not quite as easy in html5 yet)

Don't get me wrong, I like casual games - and sometimes it can be preferable to play them in a window - but even then I'd prefer an exe to a firefox tab. As a player I don't like losing mouse focus just because I've hovered over the edge of a window, and as a developer, I don't like leaving so many variables up to the whims of the web browser. Essentially, the only browser I would be happy presenting my game in is one that is effectively invisible. No tabs, no pop ups, no non-game related menus, no address bar, no browser branding. Also, having to jump backward in terms of shaders and effects is quite frustrating.

But I'm not able to create my own browser... I mean, I'm using Construct to make the game - what does that tell you about my programming skills?

...but all that said. If an exe exporter never comes for C2. I'll use Construct Classic for any serious projects... just waiting for someone to add less atrocious aspect ratio handling.

Keep up the great work, no matter what my personal preferences are, the construct environment is hands down the best beginner game development software I've tried... hence the early adopter license.

Can HTML5 support controllers/gamepads on windows/linux/mac?

Can HTML5 support controllers/gamepads on windows/linux/mac?

I'm guessing the best response to that will be about as reassuring as it was when I discovered the gamepad support in The Binding of Isaac consists of "google JoyToKey".

Thanks for the input everyone, it's good to hear these opinions. Here's my comments:

Is making an EXE wrapper much more feasible than a full exporter?

Yes, it is actually, and could be relatively straightforward to do. However, I'm worried about causing confusion if we do later release a "native" EXE exporter - I anticipate a lot of users confusing it with the HTML5 "wrapper" EXE, or not being clear on the difference (it may seem obvious to experienced users, but remember Construct 2 is aimed at largely non-technical users). So I would prefer it to be an "unofficial" or third party option, not integrated with the editor. Some people posted some links a while back already - perhaps it'd be worth double checking how well those work?

is it possible to play HTML5 offline?

Someone's already confirmed you can, but I just want to link you to this tutorial on offline support which has more info.

The first thing I do when I find a good browser based game, is look for a way to download it and play it in fullscreen.

Try making a game using one of the 'fullscreen in browser' options. Then hit F11. In the top 3 major browsers, you're now playing in hardware-accelerated fullscreen. Fullscreen gaming is perfectly possible in HTML5. The main barrier is people aren't used to it - like you say you assume you need an EXE to play fullscreen, but it's perfectly possible to do it in a browser, you probably have just never hit F11 to play a fullscreen browser game before.

There's even a Fullscreen API for HTML5 in the works.

Can HTML5 support controllers/gamepads on windows/linux/mac?

Mozilla are working on a Joystick API for HTML5, which they'll probably push to have standardised. So hopefully real joystick and gamepad input (not joy2key) will be possible in HTML5 games.

Overall I think there's a big culture shift here - who expects to visit a web page and start playing a hardware-accelerated fullscreen game with shaders controlled by a gamepad which also works offline? However, it seems this will be possible in the near future. Then, remember that to download and install an EXE, the user must go through several intimidating security warnings. Added to that there's a very small but genuine risk of viruses, malware and adware whenever downloading EXEs from the web. Some non-technical users are advised simply never to download EXEs from the web as a general rule. Compare that to an ordinary web page.

HTML5 is a work in progress, but I think in the near future it will be doing a lot more than you expect!

Disclaimer: I'm still not saying we won't make an EXE exporter. HTML5 is unlikely to ever out-perform a native EXE, so there's still a solid case to make an EXE exporter for performance reasons. My main argument is that I don't think fullscreen or offline support require an EXE exporter, because HTML5 can do it already, people just don't realise.

I'm guessing the best response to that will be about as reassuring as it was when I discovered the gamepad support in The Binding of Isaac consists of "google JoyToKey".

I had a nice big L O L when I read that.

: D

As Ashley noted what many of us (I think) feel; namely, that an EXE will always be, perhaps, a "gamer's preference." I certainly feel that way, in any event -but then again, HTML5 is in its infancy. But it's definitely nice to have everything as a neat and tidy little packaged EXE, wouldn't you say?

However, I feel that there are more important functions and features to work on and implement into C2 (like, ha ha, functions), before EXE support. I wouldn't care too much for an EXE supporter for C2 if C2 itself didn't have many of the integral and intuitive features of Construct Classic!

In other words I can play the waiting game whilst I continue to work, train myself, and experiment with Construct Classic 1.2. Ever since the shiny new brother C2 arrived on the scene, people seem to forget about the old gal.