This comes up a lot
. Seriously, a lot
. A LOT
I know Impact much better than I know Phaser. But I've been obsessively reading the Phaser docs
(they got much better recently!) recently and given myself a bit of time to convert the game I'm working on to Phaser.
My newish, slightly updated perspective on Phaser vs. Impact:
1. I really don't like Tiled
, the sort-of industry-standard tilemap editor you're going to have to use with Phaser. Since it's more generic it doesn't have the deep tie-ins that Weltmeister has with Impact. Then again, there's a team working on a game editor for Phaser
. I haven't played around with it much because, honestly and perversely, I like typing code myself.
2. The tilemap
integration in Phaser is so-so. I like how the collision/physics stuff is architected and separated; e.g. you explicitly call
on the things you want to collide – if you don't want them to collide, don't call the
OTOH, the collision system is too
simple for tilemaps. If it's not a completely rectangular tile then you can't collide with it without custom code in Arcade physics. Or you have to use Ninja physics which is kind of the red-headed stepchild and isn't fully fleshed out. I haven't played with P2 yet.
3. I think the amount of custom code you'll end up writing for a complete game is, on the balance, smaller for Phaser. Unless you're using Impact++ with Impact (which is all someone's
custom code, BTW), you're going to have to write a bit of UI framework for your menus, a particle emitter probably, some kind of tween functions, some kind of layered drawing so that this sprite always
draws behind everything else without sorting, maybe adjust the physics system so it does more than AABB with rotated sprites (Joncom's brilliant Box library comes in handy here), maybe something to tint sprites or do more complicated font handling (again, nod to Joncom)... etc.
Phaser's got a lot of that already in place. ¯\(ツ)
4. I find the HTML5GameDevs community, on balance, less helpful than this one. They're not unfriendly, but questions are more likely to go unanswered there or get unsatisfying short answers than here. For all that Phaser might be "beating" Impact, I don't see a lot of information out there about large-scale, completed Phaser games and how they were architected and maintained.
Meanwhile, ansimuz posts here a lot about Elliot Quest, for example.
5. OTOH: WebGL. Phaser has it with deep support, Impact doesn't. That might matter to you?