I need your help to figure out Airscape's performance issues

Discussion and feedback on Construct 2

Post » Tue Mar 03, 2015 11:15 pm

So.... the consensus is that webGL simply doesn't work (at any reasonable level, anyway) on Intel cards?

That's bananas. If this is really true, it's a *huge* deal for anyone making HTML5 games. It's a support nightmare no matter what, as, even if it's listed in huge red text in the minimum specifications, people simply expect 2D games to run well on low/mid-end machines.

Also, that's 20% of potential buyers gone just like that.

I do tend to believe it's a C2 problem, or at least a 2D problem however. I just tried out this benchmark, which renders 150,000 cubes. On my integrated chip it runs at 40fps easily.

Compare that to my C2 renderperftest which could only manage 3500 sprites at 30fps. Something has to be up. I just can't believe that 2D performance is that bad. Unfortunately I haven't been able to find any non-C2 2D webGL benchmarks - if anyone could point me to one, that would be awesome.
B
92
S
31
G
24
Posts: 3,191
Reputation: 32,679

Post » Tue Mar 03, 2015 11:16 pm

SgtConti wrote:
sqiddster wrote:@SgtConti

TiAm wrote:SgtConti's results are particularly bizarre...I wouldn't exactly call that a 'mid-range' system, unless we are throwing $5000 alienware's into the mix. No C2 game should be struggling on a computer like that.


Since it only is (nowadays) about a 700-800$ rig I would call it mid-range. As you can get a real High-End rig for about 1100$ (Or more) ^^
And well, without multicore support it still is easily possible for an unoptimised or really big game to struggle on such a rig.


Ah, that's a good point about multicore...doesn't your processor have 6 cores? If you are still subs-60fps at low rez, then maybe airscape is cpu bottlenecking on your system. Really doesn't seem like the demo should be doing that though...
Don't lose your work. Backup your game with Dropbox.
B
44
S
10
G
10
Posts: 1,106
Reputation: 9,187

Post » Tue Mar 03, 2015 11:22 pm

ImageImageImageImage
B
157
S
64
G
41
Posts: 2,590
Reputation: 34,640

Post » Tue Mar 03, 2015 11:43 pm

TiAm wrote:
Ah, that's a good point about multicore...doesn't your processor have 6 cores? If you are still subs-60fps at low rez, then maybe airscape is cpu bottlenecking on your system. Really doesn't seem like the demo should be doing that though...


Even though my processor has 6 cores it doesn't make any difference for *most* html5 applications as they only can utilise 1 core.
There are Web Workers that could add Multicore support or the new JXcore (I would love to see JXcore used for a C2 game).
And yes, that's exactly what I am trying to say that it may be a CPU problem. And I am quite sure that a 3.8GhZ core should be able to run it if optimised.
Based on my theory it would explain most stuff, as to why the 3D test run so much better (Because not much CPU power required there).
One thing it does not explain then though is the fact that Canvas2D is faster than WebGL.
"It's done when it's done"

Shadows of War
Buy on Steam ;)
B
24
S
10
G
7
Posts: 253
Reputation: 4,921

Post » Wed Mar 04, 2015 12:01 am

SgtConti wrote:
TiAm wrote:
Ah, that's a good point about multicore...doesn't your processor have 6 cores? If you are still subs-60fps at low rez, then maybe airscape is cpu bottlenecking on your system. Really doesn't seem like the demo should be doing that though...


Even though my processor has 6 cores it doesn't make any difference for *most* html5 applications as they only can utilise 1 core.
There are Web Workers that could add Multicore support or the new JXcore (I would love to see JXcore used for a C2 game).
And yes, that's exactly what I am trying to say that it may be a CPU problem. And I am quite sure that a 3.8GhZ core should be able to run it if optimised.
Based on my theory it would explain most stuff, as to why the 3D test run so much better (Because not much CPU power required there).
One thing it does not explain then though is the fact that Canvas2D is faster than WebGL.


That was my point. Your single thread performance isn't as good as a modern intel processor because you have more cores. I totally agree that most any C2 game that is properly optimized should run fine on a system like yours...but, as of yet, airscape's code is a black box, so it's impossible to know if it's being dragged down by un-optimzed code.
Don't lose your work. Backup your game with Dropbox.
B
44
S
10
G
10
Posts: 1,106
Reputation: 9,187

Post » Wed Mar 04, 2015 12:37 am



I was able to keep 30+ with 50K bunnies - I can't even begin to imagine doing a similar project in C2.

Can anyone explain why Pixi is so fast?
B
57
S
19
G
9
Posts: 639
Reputation: 9,533

Post » Wed Mar 04, 2015 1:09 am

@Elliott

Bunnymark was brought up before. Actually, my modified devilmark doesn't do too bad (at 30fps: 27k for devil, vs ~65k for bunny), but mainly it's because bunnymark is spawning very simple objects that are similar to our particles, and don't have the overhead of a proper sprite object. Devilmark actually maxes my cpu not my gpu, which is likely the overhead of the event engine and the sprite object, compounded by the ludicrous number of objects in either case.
Don't lose your work. Backup your game with Dropbox.
B
44
S
10
G
10
Posts: 1,106
Reputation: 9,187

Post » Wed Mar 04, 2015 2:04 am

SgtConti wrote:@Silverforce Correct me if I am wrong, but the you can't just compare a a range of products like 6300M with a single product like the Mobility 5430 in such a case.
It would be like if I would compare the GTX 900 series with the GTX 770.
So it would be usefull to know which of the 6300M models have been used.
Since a 6370M is better than a 6330M which you both would compare (and set equal) to the 5430 in your case.
I'm sorry but this had to be :D


The 6300 and 5400 series of mobility radeon (entry, lowest performance) all have 80 shaders, the difference is clock speed but the performance scaling on those older parts aren't linear. At best, you have 20% performance gap. Even the fastest 6300 series is slower than newer iGPU from Intel (when it works).

On desktop, entry dGPU typically have 500+ shaders, so they are many orders of magnitude faster, and indeed is much faster than Intel iGPU.
B
70
S
24
G
19
Posts: 1,757
Reputation: 17,609

Post » Wed Mar 04, 2015 2:11 am

Elliott wrote:


I was able to keep 30+ with 50K bunnies - I can't even begin to imagine doing a similar project in C2.

Can anyone explain why Pixi is so fast?


Perhaps its GPU accelerated, ie. GPU compute. It's definitely multi-threaded, which isn't something C2 is (besides pathfind).

PixiJS.jpg


It's not good to be in 2015 with a game engine that is single threaded. Definitely it limits creativity since the hardware available aren't being fully utilized. Quad-cores are quite common these days.
You do not have the required permissions to view the files attached to this post.
B
70
S
24
G
19
Posts: 1,757
Reputation: 17,609

Post » Wed Mar 04, 2015 2:41 am

sqiddster wrote:I do tend to believe it's a C2 problem, or at least a 2D problem however. I just tried out this benchmark, which renders 150,000 cubes. On my integrated chip it runs at 40fps easily.


3D engines can take advantage of GPU acceleration much better. It was around the time of the hardware transform & lighting being incorporated into the GPU, before that, 3D games were also limited by the CPU to setup the scene.

2D tend to be CPU limited, since the fill-rate of modern GPUs are insanely high, its usually not the bottleneck.

I'm wary of this limitation & C2 logic being single threaded. Already in my early dev of SN2, I'm seeing 50-60% CPU usage (the GPU is barely doing any work at all!), all on 1 thread on my quad-core CPU.

After SN2, I'll move to a 3D game engine for future games.
B
70
S
24
G
19
Posts: 1,757
Reputation: 17,609

PreviousNext

Return to Construct 2 General

Who is online

Users browsing this forum: brushfe, UberLou and 5 guests