Phaser 2 vs pixi vs Phaser 3 (amount of sprites per screen)

[Riddik]

Hey!

Did anybody compare Phaser and pixi for performance? We're going to make isometric game. bottom line is about objects you could render per screen. What do you think? What to choose? Phaser 2, PIXI, or maybe Phaser 3?  Need the canvas renderer of course, not webgl.

 

Update: I've the same question in the phaser slack channel. It seems that the PIXI would be better for isometric game then Phaser 2.

[8Observer8]

On 2/1/2018 at 7:59 PM, Riddik said:

Need the canvas renderer of course, not webgl.

Why? I think WebGL is better for performance than Canvas because WebGL works on GPU, not CPU.

[Riddik]

4 hours ago, 8Observer8 said:

Why? I think WebGL is better for performance than Canvas because WebGL works on GPU, not CPU.

Because WebGL still has poor support on mobile devices.

[8Observer8]

All mobile browsers support WebGL 1.0 excepting Opera Mini: https://caniuse.com/#feat=webgl

[8Observer8]

But why do you use WebGL for your game?

Quote

Phaser v2.6.2 | Pixi.js | WebGL | WebAudio

 

[Riddik]

7 hours ago, 8Observer8 said:

But why do you use WebGL for your game?

Because the game isn't sold yet.
Usually the publishers/sponsors want use only canvas.

[mattstyles]

21 hours ago, Riddik said:

Because WebGL still has poor support on mobile devices.

This isn't true.

It'll be older browsers that they are worried about, not Opera Mini, even so though, WebGL has excellent support, the percentage of users without support (depending on where your target market are) is tiny (caniuse shows 0.83% global for <KitKat Android, Opera Mini users won't be stuck with it—thats 2% anyway, its tiny—, iOS has supported and supported well for a long time).

Publishers/sponsors requiring it is another matter though. 0.83% is pretty compelling. It's totally dependent on the actual target market, largely split by geography, so the numbers might vary.

On 2/1/2018 at 4:59 PM, Riddik said:

Update: I've the same question in the phaser slack channel. It seems that the PIXI would be better for isometric game then Phaser 2.

What was the reason?

Phaser 2 uses Pixi, albeit a slightly older version (I think, may have changed), so I'd expect them to be almost comparable in performance. Phaser doesn't add much (anything?) to the rendering layer, so perf should be mostly the same (the older version, if true, could accrue more of a difference, maybe that accounts for it).

It's not a like-for-like comparison though, Phaser is a framework, Pixi is a rendering lib.

Phaser 3 (I think) uses a different rendering mechanism, again, I'd expect perf to be comparable or better though otherwise it would be a very odd choice to ditch Pixi as a rendering engine.

[themoonrat]

I created a tool that allows direct engine to engine and version to version performance comparison: https://github.com/themoonrat/webgl-benchmark