Mobile games performance, let's be honest

0 favourites
From the Asset Store
Firebase: Analytics, Dynamic Links, Remote Config, Performance, Crashlytics on Android, iOS & Web Browser
  • Arima yeah i do it for all similiar colisions condition i check CPU usage and i saw no better:

  • Try Construct 3

    Develop games in your browser. Powerful, performant & highly capable.

    Try Now Construct 3 users don't see these ads
  • The top way requires 5X more collision checks for the same outcome.

  • yes i understand this and i implement this, result is on upper screenshot.

    Any other ideas?

  • I dunno because its worse from your screenshot, now its bonkers 80% of CPU for your events but down to 12% for your engine (before your engine was 37%)... so i really have no idea how you put your events together.

  • delgado

    Might be able to help. Sent you a PM

  • Ashley, I read your last question on what I meant, but I didn't have any time to reply until now. But I read you previous post here and that's actually what I meant about what I wrote.

    I'm not saying that C2 is inherently slower, I'm saying that because the ES are based around inserting code in to the game loop on per layout sheet; and how new developers aren't used to OOP design. It's easy to fall into the example above of inefficient design.

    now GMS can do same thing, but GMS really pushes that code and behaviours results are triggered based on conditions. This leaves that GMS builds and reorganize for efficiency. While dev's can still do poor design it's less natural to do so. As another analogy it's C or C++ in comparison to Java's OOP only design.

    So it's not that I believe that C2 gameloop design is inherently slower; working with C2 just requires more diligence and know how. If it seemed that I said C2 was inherently slower then I failed communicate what I was saying.

    The only element that could/would/does make C2 slower is the JS. So after debating supporting my iTouch4g(and deciding not too), I checked out the common hardware and found that Tegra2 and the A4 chips(and equivalent qualities of both) are becoming obsolete. The Tegra3/4+ and A5+ are the most common chips out there; and for the most part they run JS games pretty darn well.

    So as a general guide for everyone. Learn efficient code design, test on your weakest device often(if not your primary test device) and if your using an old device; get something newer(I also suffer the same budget constraints).

  • delgado - the easiest thing to do is reproduce the problem in a new .capx that you are willing to share. Then everyone can help instead of 6 pages of speculation.

  • Ashley Of course, I try to do everything all over again if necessary But i wants you to know what I have observed.

    I am doing this game for many months and every few relases of C2 i use CocoonJS to do apk. Hoping the game will runs better than before. I'm not entirely convinced that the fault code, as in previous versions of C2 my game runs good, 25 FPS.

    Thank you all for your interest in my topic.

  • It's your code, because the events are sucking up 80% of the CPU cycles.

    That is highly abnormal.

  • Obviously something must cause the high cpu cycles and today it looks like its this events. On the other hand, this theory is not entirely true, and with this I disagree because as I said already before the game could have had the normal performance in the past

    Something's going on between relases of construct 2 a side effect of something(?)

  • delgado

    What Ashley is saying is that you need to post a capx.

    If you are uncomfortable putting up your whole capx -- understandable -- then disable events until you find out what is causing your cpu drain, and make a simple capx without your assets or any unneeded code, and upload it.

    Otherwise, there's nothing we can do.

    Cheers, T

  • TiAm i send Ashley capx to mail. I think i do what you say. delete some code and check

Jump to:
Active Users
There are 1 visitors browsing this topic (0 users and 1 guests)