Is Construct 2 the right choice for this project?

Me and a few friends are currently developing an Action RPG akin to Legend of Mana on the Playstation. We're currently using Construct Classic to build it but we'd really like to be able to export the game to the Mac also once it's done.

After seeing Ashley's post of Construct 2's HTML5 games being wrapped in Awesomium as an .exe (and possibly as a .dmg file too?) I can see how HTML5 may actually be viable as a multiplatform solution for Construct even if it's not exactly exporting as a native-run program.

The question I have is, would Construct 2 really be the right choice for our project? It's going to be a really big game with lots of hand-drawn animation per sprite on 720p backgrounds and we don't intend (or even see it as a good idea) on it being ever run in a browser. Is HTML5 at its current state strong enough to handle an RPG of that scale? I really don't want to switch IDE's. We're currently looking at GameMaker which I still find clunky and from working with Construct Classic for around two years now I just find it hard to let go of our lovely and robust event system.

I can't seem to link any screenshots for some reason so if you need a more visual idea of how the game would be like, here's a link to our devblog.

http://cryamore.tumblr.com/

Thanks for your time, guys and I'm hoping for good news. <img src="smileys/smiley5.gif" border="0" align="middle">

Hey! I've just checked the blog of your project, that's really great!

And well, on my mind I'm not sure that Construct2 was the better solution for your game, furthermore if the aim of the game wasn't the internet or mobile support.

An rpg will require a ton of sprites with several animations each, but its just not viable to load a lot of assets at one time.

Hopefully we will get the ability to load an external sprite with animations in the future though.

Most browsers nowadays support hardware acceleration (including Chrome and EXE wrappers based on it). That means it's perfectly capable of drawing in HD, since the rendering is offloaded to the GPU in exactly the same way native engines would.

Most desktop systems nowadays also have enough RAM and VRAM to fit entire games in memory at once, although we're planning some load-by-layout features anyway (mainly to help mobile, but it will also help very large desktop games).

I'd recommend making a prototype - whip together a demo of what you're after and see how it runs. Make sure you have an up-to-date graphics card driver though, so you're not getting software rendering. (In Chrome, you can check its status in chrome://gpu)

Thanks for the quick replies, guys! We're still busy making a crapload of assets for the game anyway so there's time enough to hopefully see C2 expand further and hopefully be able to support this project as I really am not up to bothering with GameMaker since a lot of the features that we can easily do in Construct's event system requires GML there and I really don't feel I'm cut out to be a programmer.

It'd be nice to be able to represent Construct with our project since this has got to be hands-down the best WYSIWYG IDE. If C2 ever gets to about the exact same features as CC is currently at or surpasses it even and can totally handle large desktop games through HTML5, one probably wouldn't have to look anywhere else and it'd definitely make quite an impact on things.

I was formerly under the impression that HTML5 was just going to be limited to mobile games and it'd be stuck competing with nothing but Angry Birds, Cut The Rope and the like. From what I've been seeing though, perhaps HTML5 really will overthrow Flash in almost everything it covered. If Super Meat Boy, Castle Crashers and Scott Pilgrim vs. The World: The Game were all done in Flash (I may be wrong on the last two), then with enough time HTML5 may just be able to do that as well.

My thanks again, Ashley, newt and wilfryed!

Congratulations !! The sprites look great!

What platforms are you targeting ? I too hate web games. Not the games exactly but the environment. For mobile html5 is definitely not ready, though CocoonJS makes things much better. Working with Construct 2 is super awesome. The only problem is exporting to html5. My dream is to have a Construct exporting natively to all platforms, but i understand it's a huge undertaking. Even then others have achieved it (with restrictions of course). So it's possible. Maybe HTML5 turns out as the ultimate solution to multiplatform , though i don't see it happening anytime soon sadly. If it happens we're set with Construct . If not...

Anyway your game looks great. Do as Ashley said, experiment with C2 to see how things turn out.

Thanks, Kiyoshi!

We're mainly targeting PC and Mac (if HTML5 was/is capable enough Linux could be in too) though we've made sure the control scheme can easily translate to consoles in case we decide to port it over there too.

We share the same dreams of C2 exporting natively but who knows how things will play out with HTML5 and these wrappers? Maybe it's all we really need in the end and HTML5 really will be the answer to a big chunk of our game developing desires.

I'm really hoping Construct 2 grows way further, it's my ideal WYSIWYG game-making program really. I'll probably try to work on the HTML5 prototype tomorrow, right now we're (sadly) trying to come to terms with GameMaker's GML which is just confusing to me really, it's good that my friend who I'm working with has some experience with C++ but even he's puzzled by some of its workings.

If ever I get a nice enough HTML5 prototype up, I'll see if I can throw it in for testing in the "Your Creations" forum.

Yeah i hope html5 catches with native power. But it appears that'll take some time yet. Multiplatform dev is a nightmare , specially in C++. I myself am trying with C# and Monogame right now and even there there's still some headache involved. A common runtime with common programming language for all platforms is the ultimate dream. But for that to become true the concepts of native and web must merge, not substitute one with the other, like many think. The idea of NaCL was really awesome in that sense. We then wouldn't code in native level anymore but a level above, common to all platforms. Just dreaming :D