Construct 3 - many questions (native exporterts)

c2 is pretty powerfull, only users are pretty weak.

p.s. if you never programmed in something "real" like c# c++ and you miss a lot of patterns and such knowledge i suggest you spend a month or two reasearching about that. then suddenly your logic will be up by 100x and c2 will suddenly become overpowered. (some things could get an improvement true, but you can practically do anything)

C2 is the best tool to make HTML5 games.

But export to other formats is not native that's why games work bad even if they are very well optimised.

For example, I created a physics puzzle game - it shows 60 fps on Crosswalk 7 but there is an alert from Google Play about OpenSSL.

That's why I used Crosswalk 10 - there is no alert from Google Play, but the game shows only 30-40 fps and there are freezes every 4-5 seconds.

Other example, NodeWebkit shows very bad performance if the user does not have Microsoft Visual C++ 2008 redistributable installed on his computer.

The game could run 60 fps on Google Chrome, but it will run 20-40 fps on Node-webkit if the user does not have Microsoft Visual C++ 2008 redistributable installed on his computer

What if it never gets better? How can you rely on third party if that can make C2 look bad if they do something wrong?

Well, it already is better - check yourself by running Canary. In time that will make beta, which will then make stable. Given all browser makers have been putting in a great deal of development resources in to browsers for the past few years and Chrome and Firefox in particular pushing very aggressive 6-week release cycles, I find it very difficult to take seriously the idea that anything will stay the same forever, particularly when Google say they are aware of the issue and are working to make v-sync scheduling flawless.

Modern software development without relying on third parties is practically impossible. For example some users suggest we use Haxe instead, which is just a different third party and we could equally be hosed by problems with Haxe. (I also doubt it has as much testing and development resources going in to it as Chrome does.) Imagine if we wrote an engine depending on XNA - Microsoft ended up dropping support for that. Or Silverlight - the same just happened (they officially announced recently Edge won't support Silverlight!) All platforms have their quirks and risks. Chrome's v-sync scheduling is a bug - it's been a problem for a while, but it's just a bug like any other - and as far as I can tell it's already fixed in Canary, where according to sbperftest on my dev machine the maximum jank is just 0.1ms off v-sync.

its nice to see all of these games engines being ported to linux as well.

First it was Unreal, then Godot, clickteam fusion 3 is planned and construct 3 as well! Now even Unity3d has a linux editor planned for a release - those guys got it to a point where they can show screenshots even!

Why the sudden interest after all these years? Is this possibly a result of Valve's push for steamos??

I agree with Ashley on the performance point. I just wish the html5 wrappers had a more consistent/stable performance.

1. Will it have native exporters to iOS/Android/Windows/Linux/OSx ?

2. Will it have an opportunity to export game to flash format (.swf)?

3. Will it have native physics engine?

4. Will it have native 3D support?

5. Will it support vector images?

To cut a long story short:

1-No

2-No

3-No

4-No

5 ???

Will C3 support vector images like .svg?

"native" is getting harder to define, anyway. Would WebAssembly count?

Well, "native exporter" means two things, imho:

1. It creates native app instead of creating container with portable brower and HTML5-game in it (NW.js and Intel XDK create such container, that's why apps are not as fast as they could be)

2. It is not a third party product, because only Scirra can make it work perfect.

I understand that it is very very hard to create native exporter - it's like creating another construct, but you can sell such exporters for extra money.

I have C2 personal license, but I am ready to buy native exporters for extra money with pleasure. And I am not the only one, who wants native exporters and who is ready to pay for that extra money.

paradine

It is equal to create a whole new construct, from engine to plugins.

If you could clone many Ashley for those native exporters, then it might be possible to have them.

If you could clone many Ashley for those native exporters, then it might be possible to have them.

challenge accepted .. Calling GRU

If you could clone many Ashley for those native exporters, then it might be possible to have them.

Very good idea, cloning Ashlay can really solve all our problems

paradine - as I keep saying, many games we see are GPU bottlenecked, so would not run faster in a native engine.

paradine - as I keep saying, many games we see are GPU bottlenecked, so would not run faster in a native engine.

Ashley you know that this is not true, even if the projects of some users are poorly optimized, the examples provided with C2 are not.

And even they work not well with the most of mobiles using XDK or Phonegap. (15 to 30 FPS default even if the game at runs the same speed you can still see the visual differences between 60 and 30 fps)

The good thing is that the new Canvas+ by ludei is working well on mobiles even if the exporting process has some more steps that you have to do manually.

I don't want to offend any solution for the exporting process containing mobile gaming, but it seems like Cocoon.io seems to be the better than Crosswalk at the time.

(But I don't want to open another disscussion about this because there are already over 100 of this kind...)

paradine - as I keep saying, many games we see are GPU bottlenecked, so would not run faster in a native engine.

Ashley - but many are CPU bottlenecked, they would run much faster in a native engine, I guess.

I always ask to see CPU-bottlenecked .capx projects. Usually either nobody sends me any, they send such appallingly inefficiently designed games it is obviously just bad design, or they are actually GPU-bottlenecked. Please do send any good examples of CPU-bottlenecked games, so I can investigate if our engine needs optimising. I rarely see such examples though, which is why I have concluded most games are not CPU-bottlenecked.

TheRealDannyyy - if a game runs at 15 or 30 FPS that could easily be due to GPU bottlenecking, or using software rendering due to poor quality drivers (in which case native engines would be susceptible to crashing, failing to work at all). Neither of those cases is much better by moving to a native engine. Anyway, if Canvas+ works better, that's not a native engine either, it just highlights that perhaps there is some issue with Cordova/XDK, and they can be improved, I don't see why this requires a native engine to solve.

Quite correct Ashley. Bad design in game makes it worse. One cannot produce game in short time. Some might yes you can but, I do not think so. I have been in game world since 1990. Needs lot's of work behind good game. One cannot blame Construct 2 if one has badly design game. If you have to many animations, it will bound to slow down. Blaming C2 is not the answer. For this reason I don't see why this requires a native engine to solve.