tgeorgemihai's Forum Posts

    Agree with Ashley there. Too many people complaining about mobile performance when they clearly don't know the limitations of mobile development. Just because you can make something in C2 and can export for mobile doesn't mean that it will play well on mobile. I've learnt the hard way. Investigated what's going on under the hood, what features to use and what not to use.

    X - Don't use webGl effects - They are neat but usually heavy & not useful for mobile.

    X - Don't use Force own texture on layers.

    X - Don't use constantly updating Text object if you can do the same with Spritefont.

    X - Don't use particles, unless it's very simple and low amount of particles.

    X - Don't expect heavy amount of physics to run well.

    X - Don't update things every tick unless you have to. Use 'Every X Seconds'

    X - Don't use big images with large areas of overlap.

    * - Always limit your Top level events. The less events the the engine has to go through each tick the better.

    * - Group event's and only activate groups when needed, close when not needed.

    * - Use smart and cheap ways of picking objects & Structure your conditions accordingly.

    * - Disable collisions for objects that don't need it. (Personally I think this should be disabled by default)

    * - Keep an eye on memory, cpu, fps, draw calls, ammount of sprites, etc

    * - Study and try different approaches of doing the same thing in a smarter/cheaper ways.

    If your game is not performing well, it's probably not because C2, it's probably because you don't know what the hell you're doing, and have no clue what the mobile limitations are.

    Or simply is HTML5's or wrapper's fault. Come on, today we can run 1080p/4k content and play HD games without any issue on our phones, but can't run a f*cking 2D game, we need to optimize the s*it out of it ?

    I've compiled to Android a few GM:S demos and examples that I've found on the internet. They were pretty complex (intended for desktop) and guess what ? They ran great on my Galaxy S4. And were compiled with the basic option, there is also an option that automatically optimizes stuff even better.

    Here a clear example of how the "export HTML5 only" affects in the long run: http://www.perfectly-nintendo.com/the-n ... e-to-wii-u

    So yeah, If he made that game in any other engine that had at least one native exporter (C++ or C#) he wouldn't had to remake the game.

    Anyway, I've made my choice. While CF 2.5 is a little faster to develop, I've had it with 3rd party exporters, so I will go for GM:S (all exporters included and better documentation). It's a shame, because I really liked developing with C2. Would be great as a visual scripting interface for Unity (like Playmaker, but only better).

  • Not sure about Steam version, but there are 33% off sales from time to time on Scirra Store.

    spy84 Is no use, this was requested a lot of times before but Ashley is set only on HTML5. While I see the advantages of HTML5, in practice there are a lot of drawbacks.

    I also have GM:S Pro + Android exporter (12$ on Humble Bundle) and Clickteam Fusion 2.5 (27€ Steam Sale). And now that I will have some time on my hands I will try both and see witch one I keep (I can still refund CF 2.5 in 14 days).

    I've made around 1800$ profit with Construct 2 and I've learned a lot of useful stuff, but the "no native" is starting to become a problem. Adding 60 MB ~100 MB to a simple platform game is not good and the performance on mobiles is hit or miss. Also, we stay at the mercy of node.js or other wrappers.

    I will still use C2 for prototyping my games (is easier, faster and better than pencil and paper) and wait to see C3, but in the meantime I will try another engine.

    To be honest, I think that Clickteam Fusion 2.5 (27€) + HTML5/Android/iOS exporter (45-75€) is a better deal than Construct 2 (100€). Even trough is not easy as C2, is still easy enough.

    Of course it has it's fair share of bugs and issues and hidden costs (like the need of Developer Upgrade in order to use Ads), but at least you get native support with the possibility of upgrading to different platforms (no more "close to native speeds" and the "joy" of using wrappers). There is also a 3rd party exporter for consoles named chowdren.

    If you want to make small/medium games quick, C2 is the best choice, but for larger games I would go for Unity, GM:S or CF 2.5.

    Keep in mind that C2 was developed when HTML5 just appeared, so it's core is not up-to-date. For example arm.js is used only for physics and not for whole C2. Who knows, C3 might be a nice surprise with latest HTML5/Javascript technologies implemented (a huge boost compared to C2).

  • What resolution should I use as a base for my projects ? I intend to publish to mobiles and desktop. I know about aspect ratio and bigger resolution sprites (rendered at runtime), but some elements have fixed size (tilemap, tiled background, 9-patch).

    854x480 -- 409.920px; FWVGA; 44% of 720p;  20% of 1080p
960x540 -- 518.400px; qHD, 56% of 720p; 25% of 1080p
1280x720 -- 921.600px; HD; 44,4% of 1080p
1920x1080 -- 2.073.600px; FHD; high-end[/code:8goyadmp]
Will there be any noticeable benefit to use qHD instead of FWVGA, or the difference would be negligible ?

  • This is more of a marketing practice (all stores use it)

    You can still find a lot of free stuff on forum or on the internet.

  • You can export .exe games and upload then to steam using NW.js (node-webkit).

    You can also use Steam services in your games with Greenworks plugin.

    In both cases, expect to find some bugs/glitches

  • This looks very promising

    Ashley What do you think about this workaround ? Can you implement it in the official plugins ? Would it have negative effects (bugs/glitches) ?

  • Bookmarked

    Very useful until an official solution is implemented.

  • R0J0hound

    Hello, I need to implement a split screen in my game (2, 3 or 4 split-screens). Is there any way to copy the layout only in a section of one Paster object, or I need to use 2~4 separate Paster objects ?

    Also, my primary target are mobiles (browsers/WebGL). Which should I use for a better optimization (better performance) Paster or Canvas ?

  • Try Construct 3

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

    Try Now Construct 3 users don't see these ads

  • however tgeorgemihai your are citing some pretty pricey alternatives to C2.. and calling Cocoon IO expensive. it's $120 a year for a very reasonable plan.. higher priority.. 10 simultaneous projects.. yes the logo thing is pricy but like i mentioned before.. if you wanted to debrand unity you're look at $4500 to do it for ios/android..

    You've got that wrong. I've enumerated alternatives compared to Cocoon.IO's logo removal of 500$. Like I've said: "I can understand the monthly fee (they have servers and people to pay)".

    And you tell me that I presented some pricey alternatives (< 800$) and after that you tell about Unity Pro (4500$)

    Just so you know, for 4500$ (9 games splash removal) you get a little more in Unity than just the logo removal.

    Just to be clear, this wasn't to bash C2, it does it's job great as a HTML5 game engine

  • While Cocoon.IO is a nice service, I'm sorry to say it, but I see very little reason to start using it now. Here is why I think this way:

    Usage:

    The only "real" use for iOS is on older devices like iPhone 3GS, iPad 1, iPod Touch 4 (the ones that didn't get over iOS 8) and the fact that you can build the .ipa file without a Mac.

    On Android it is useful on <5.0 (before WebView+ is updated separately). That is where the Canvas+ is great (performance and disk space). On Android 5, Canvas+ and WebView+ gives me about the same performance (and WebView+ supports DOM elements and other HTML stuff).

    No support for other platforms.

    Pricing:

    I can understand the monthly fee (they have servers and people to pay) but 500 f***ing $ to remove a logo

    Here is small comparison:

    You can get GM:S Master Collection for 800$ and export native on pretty much any platform with no monthly fee, anytime you want. During summer sales, you can get it for 500$ ... Heck, I've got the Pro version + Android export + sources from some games for 12$ (Humble Bundle)... But yes, it involves code and very old interface and room/level editor

    You can get CF 2.5 (Android and iOS export) for 350$ (or 650$ if you want to include ads/IAP). There was the December sale when you could get it at around 1/3 from the regular price... A decent alternative to C2 If you could get it discounted.

    If your game is not that big, you can literally pay someone to port it.

    So unless your games is pretty big and hard to port or you are already dependent on Cocoon services, I see no point especially since all the platforms start to have a better support for HTML5 without Cocoon.IO .

    Cocoon.IO really have to step up their game for indies, something like to include your game logo inside the Cocoon logo (something like a collage) or % of the ads revenue instead the monthly fee.

  • Sorry to say it, but this is a stupid idea. It would be a waste of time and resources.

    Like other people said, you can get a cheap Windows tablet, use Team Viewer or have C2 on a stick and continue your work on any PC.

    Also I seriously doubt you can do anything productive on a <5" screen.

  • TGeorgeMihai, maybe I misunderstood but you seem to have the mistaken idea that nw.dll is included exports from Construct 2 intended for mobile.

    No, I was referring to desktop/NodeWebkit export.