A question for experienced developers in C2/C3...

Post » Wed Aug 09, 2017 7:21 pm

Games similar to Child of Light and Rayman Legends are definitely possible, those games rely heavily on artwork, you just need to be a bit smart about it. In terms of graphical fidelity I'm doing a game with similar results. You have to rely on third party applications and plugins to get up to the same level, but you will get there eventually.

I use Construct 2 with spriter as animation tool and magicam for camera panning and transitions, then I use HTML5-effects such as basicBlur, somebody's effects "flag" and "skend" for foliage (together with built in sine-values for angle and size of the objects). I think the performance of Construct2 is great. You won't get to same performance levels as ubi-art but to be perfectly honest ubi-art is not really a 2D engine. It uses a lot of 3D technology which makes the game even more optimized.

If you target PC and consoles you have a lot of headroom. if you target mobile you need to be smart about layers and overdraws (you can have a lot of layers, but try to keep overdraws to a minimum) download rayman jungle run and you will probably get what I mean with a lot of parallax-layers, but never at the same screen, and hardly ever overdrawing.

I'm making this game, perhaps it is a good example that will showcase what you can do with construct.
Currently working on "Unsung" - an open world platform adventure for PC and Consoles. follow me on Instagram :D
B
18
S
2
G
2
Posts: 50
Reputation: 1,230

Post » Thu Aug 10, 2017 9:49 am

Message: maranpis can only post plain text URLS until they have 500 rep. 2 URLS modified. Why?
g3nki wrote:Games similar to Child of Light and Rayman Legends are definitely possible, those games rely heavily on artwork, you just need to be a bit smart about it. In terms of graphical fidelity I'm doing a game with similar results. You have to rely on third party applications and plugins to get up to the same level, but you will get there eventually.

I use Construct 2 with https://brashmonkey.com/]spriter[/url] as animation tool and magicam for camera panning and transitions, then I use HTML5-effects such as basicBlur, [url=https://www.scirra.com/forum/effects-by-somebody-neat-things-big-gifs_t123401 somebody's effects "flag" and "skend" for foliage (together with built in sine-values for angle and size of the objects). I think the performance of Construct2 is great. You won't get to same performance levels as ubi-art but to be perfectly honest ubi-art is not really a 2D engine. It uses a lot of 3D technology which makes the game even more optimized.

If you target PC and consoles you have a lot of headroom. if you target mobile you need to be smart about layers and overdraws (you can have a lot of layers, but try to keep overdraws to a minimum) download rayman jungle run and you will probably get what I mean with a lot of parallax-layers, but never at the same screen, and hardly ever overdrawing.

I'm making http://pixelshade.se this game, perhaps it is a good example that will showcase what you can do with construct.


Beautiful art work! it's good to know that this kind of beautiful games can be done in C2.

keep up the good work!
B
7
S
1
Posts: 14
Reputation: 401

Post » Thu Aug 10, 2017 1:36 pm

@g3nki:
That's a fine-looking game you got there so far ;)

Colludium wrote:I would be very interested if you could put together a GMS/C2 bunnymark comparison!

Okay, here we go:

GMS2 - no render - 9500 objects - 55 fps
GMS2 - render - 9300 objects - 55 fps
C2 nwjs - no render - 12650 objects - 55 fps
C2 nwjs - render - 10900 objects - 55 fps

This is on an i5 2.67ghz processor, 8 gig ram, intel iris 6100, and windows 8.1. For the no renders I just turned off sprite visibility on both builds.

It seems like C2 outperforms GMS2 by quite a bit on this benchmark. Which makes me wonder why my own game performs significantly better in gms than construct. The game logic is structured about the same, and the gpu load should be more or less identical. Hmm...

Possible culprits: The platform behaviour bogs things down in my game. Or maybe the code refactoring helps performance out in the gms version (things were getting a bit unwieldy in the c2 version).

Anyways, pure speculation! On this benchmark at least, C2 comes out on top.

Here's the GMS2 builds:
https://www.dropbox.com/s/30yeiludx4ast ... d.zip?dl=0
https://www.dropbox.com/s/dy85n6qv7urkt ... w.zip?dl=0

And the project files if anyone wants to load them up in GMS2 themselves:
https://www.dropbox.com/s/aclujpdr5c27g ... h.rar?dl=0
B
38
S
16
G
6
Posts: 537
Reputation: 7,582

Post » Thu Aug 10, 2017 7:48 pm

I think you'll find that HTML5 just generally feels more "janky" than most native platforms, so although "in theory" / simple tests like this C2 might out-perform, it's just not a consistently smooth feeling even if the FPS number is the same.

ErekT wrote:The platform behaviour bogs things down in my game



+1 to this, I've seen that most people who don't run into major performance issues either use their own platforming behavior or are not making a 2D platformer
"Construct 4 lets YOU make advanced games! (but not play them)" Construct Classic - Examples Kit
B
113
S
39
G
17
Posts: 2,184
Reputation: 19,217

Post » Thu Aug 10, 2017 11:10 pm

Jayjay wrote:+1 to this, I've seen that most people who don't run into major performance issues either use their own platforming behavior or are not making a 2D platformer


Perhaps I'm jinxing myself, but I've never had a problem with it. I'm currently making a 1080p platformer and being real sloppy about graphics optimisation tbh, and it seems to lick along at 60fps like it ain't nothin' 8-)
Image



Coming soon on Steam!
B
16
S
9
G
3
Posts: 202
Reputation: 3,462

Post » Fri Aug 11, 2017 8:25 am

I try to avoid using all kinds of effects these days. My current project doesn't use a single effect and it runs really well on even low end PCs with full HD assets. However as soon as I add a few effects it starts lagging a lot on some of them. Even on some good computers as well.

My other game for example ran good on all my test computers but when I released it on Steam I realized that this wasn't the case for tons of users. So I removed the effects in my game and now I haven't had a performance issue report on steam in months.

So I avoid effects like the plague now :lol: It's not worth the risk imho. I use other techniques to achieve the look I want instead.
B
52
S
22
G
12
Posts: 727
Reputation: 11,720

Post » Fri Aug 11, 2017 9:13 am

Anonnymitet wrote:I try to avoid using all kinds of effects these days. My current project doesn't use a single effect and it runs really well on even low end PCs with full HD assets. However as soon as I add a few effects it starts lagging a lot on some of them. Even on some good computers as well.

My other game for example ran good on all my test computers but when I released it on Steam I realized that this wasn't the case for tons of users. So I removed the effects in my game and now I haven't had a performance issue report on steam in months.

So I avoid effects like the plague now :lol: It's not worth the risk imho. I use other techniques to achieve the look I want instead.


Hello Anonnymitet i saw your game and it looks very fun to play :D what kind of effects made the lagging?
B
7
S
1
Posts: 14
Reputation: 401

Post » Fri Aug 11, 2017 9:39 am

It was just regular effects like adjust hsl, and brightness and stuff like that. But not many and I didn't overuse them at all. Integrated graphic cards are a nightmare and doesn't seem to handle effects well and I had no idea so many people used it until I released my game :P I have exactly the same problem on unity with image effects so it isn't just c2/c3.

But I really like working in construct and for me it doesn't really matter. As I said, I can easily make it look the way I want without using effects.
B
52
S
22
G
12
Posts: 727
Reputation: 11,720

Post » Fri Aug 11, 2017 10:31 am

@ErekT Thank you for posting your results.
B
129
S
33
G
17
Posts: 1,552
Reputation: 20,515

Post » Fri Aug 11, 2017 11:26 am

ErekT wrote:It seems like C2 outperforms GMS2 by quite a bit on this benchmark. Which makes me wonder why my own game performs significantly better in gms than construct. The game logic is structured about the same, and the gpu load should be more or less identical. Hmm...

People often accuse Construct/HTML5/browsers of having poor performance, but when it comes to actual measurements, it actually performs excellently.

I think it's quite possible that if you reimplemented your whole game exactly the same way in a different engine, it would still be slow. If you use a huge amount of intensive logic for example, that is still going to have to be run no matter the technology you use, unless you optimise it. I'd guess since events and script are so different, you actually used a lot less logic in GMS, so it's not really a fair test.
Scirra Founder
B
387
S
230
G
88
Posts: 24,251
Reputation: 192,454

PreviousNext

Return to General Discussion

Who is online

Users browsing this forum: No registered users and 1 guest