Announcements, feedbacks and other discussions about GDevelop.
User avatar
By Darkhog
#65030 Perhaps it would be prudent to keep releasing GD versions with old editor while developing new one as a lower priority project until it's on par with GD4 editor? Basically, the current situation seems like development had stalled or even went backwards (as the GD5 editor is hardly usable, especially if you want to make a native game - couldn't find where to select extensions and stuff).

My point is, keep maintaining GD with new engine features and old editor until the new editor is up to snuff. Yes, I know that with 4ian doing the most for the development, dev speed of new editor would greatly slow down, but I'd take that if it means I'll have something usable and with new runtime features to work with. Replacing the old editor is important, but it is not THAT important to pause development of the entire application over it, frankly I think 4ian is missing the forest for the trees.
User avatar
By ddabrahim
#65031
keep releasing GD versions with old editor while developing new one as a lower priority project

I need to disagree. GD5 must be the priority because the old IDE is old, completely useless on Linux and Mac and the native code base is also old and outdated, many people complaining for problems especially with new graphics cards.

GD5 editor is hardly usable - couldn't find where to select extensions and stuff).

Disagree again, it come a long way. Almost all core features supported and even more. All events of the extensions can be used without need to enable the extension and to add a behavior to an object just right click the object in the list, select edit and select behaviors on the top of the window. If something is missing probably because it is not implemented yet but I argue that it would be hardly usable.

especially if you want to make a native game

That is because native is currently not supported in GD5, it is HTML5 only. I don't know what is the plan regarding native engine but I hope that SFML wont be coming back. Not a nice engine.

My point is, keep maintaining GD with new engine features and old editor until the new editor is up to snuff.

Since you left GD was pretty much dead, it was maintained only with bits and pieces I'm glad 4ian is motivated again to develop it even if it means only a new IDE and new IDE features and even if it become an HTML5 only tool. Yes I would love to see new engine features both native and HTML5 but need to accept the fact, 4ian is still the sole developer and he need to priorities. He decided the new IDE and HTML5 engine become his priority for many reasons. The main ones, easy to maintain, easy to extend, easy to contribute for others, can support all platforms without effort.

I recommend to accept GD4 is the past, we already discussed this and most people on the forum didn't care as long GD5 brings improvement. It too late to argue that and even if we would, it does not change the fact 4ian is on it own with this and would like to move away from the native code base :evil:
User avatar
By Darkhog
#65033 Not working on Linux? That's funny because when I first started using GD, it was on a Linux and it worked just fine. The only bad thing I remember was lack of crosscompiling meaning to make a Windows and Linux game you'd need to have both a Windows and Linux machine (virtual or not).

As for SFML... it's the best object-oriented C++ engine available. Even if it sucked, there's nothing better unless you are ready to cook your own framework.

As for making it HTML5-only... that's not a good idea. Construct 2 went this way... and people forgot about it subsequently because performance of HTML5... is even worse than early versions of Java (I mean JRE 1-4). Much depends on the browser used, of course, but even then, the performance of JS across the board is pretty even and even slowest browsers don't lag too far behind the leaders.

Going back to Construct, people aren't likely to recommend it to you, even if you want "no coding required" tool. No, they're far more likely to recommend stuff like Clickteam Fusion or Game Maker which can export native binaries and not just a webpage wrapped up in an exe and therefore are faster.

Again, 4ian is missing the forest for the trees, i.e. he focuses on making something using "cool" technologies with lots of buzzwords surrounding them instead of making something that is easy to use, feature-rich and fast.

Addendum:
Since you left GD was pretty much dead, it was maintained only with bits and pieces

And see what you get without my guidance? Shouldn't have left at all, maybe I could've prevented it from happening!
User avatar
By Lizard-13
#65036
And see what you get without my guidance? Shouldn't have left at all, maybe I could've prevented it from happening!

:lol:
Seriously, I'm pretty sure the new IDE will be great, just give it some time. Also it's much easier to develop it, and the compiling time is almost null, you can edit the IDE on real time, very good to improve it fastly (wait 10 minutes to get a compilation error because a ";" is just annoying) :)
User avatar
By arc1tec4
#65037 I hope it will be good....imo I agree that html5 only sucks. I use the old editor it starts a lot faster, has more features, runs better for me.

As far as the Linux support it works great on Linux I actually run both of them in Linux, (windows and Linux versions). It runs the windows version in Wine better than windows does. Thats been my experience for the last 7 years, if I don't want to lose work and time stay away from winblows.

If GD4 becomes useless for some reason, although I can't see why yet. I'll just move on to something else. Keep checking in on GDevelop, for all the visual scripters I like it the most.

I believe that 4ian wil make what he wants, and it is what it is.
User avatar
By ddabrahim
#65038
And see what you get without my guidance?

Agree, it is your fault , 4ian lost the path without you :twisted:

Not working on Linux? That's funny

No it is not. If you get a glorious update tomorrow that fix a million bugs may just cause GD crash then you come here and jumping what a trash GD is, though it is down to Linux and the million different packages and their million different versions it uses. :(

As for SFML... it's the best object-oriented C++ engine available

Maybe, but it doesn't change the fact it keep crashing on many configurations and for the exact same reason even the devs them self give up on porting to other platforms. Even if it the best, I don't want it.

As for making it HTML5-only... that's not a good idea.

I didn't nor 4ian say that it is going to be HTML5 only, but this is what 4ian focusing on. If he simply haven't got time to maintain multiple code bases there is nothing we can do about it other than try to understand a decision has to be made and accept it. As HTML5 is faster to develop and require the least amount of effort to run it on multiple platforms, this is what he decided to go with and personally I'm happy for it.

people forgot about it subsequently because performance of HTML5

Oh no, there is nothing wrong with the performance. What the problem is, lack of support. Obviously for gaming the V8 engine is the fastest but not the only one unfortunately and many people judge performance based on the trash JS engine their system and web browser is using. HTML5 is not the problem here but the support which is growing and improving fast.

Going back to Construct, people aren't likely to recommend it to you

For making HTML5 games, Construct is pretty much the most complete solution out there. If you want to develop something else, obviously there are better solutions. No one saying HTML5 is the best for everything but only that it is the most simple way to get something everywhere especially for a small team of 3 like Scirra not to mention teams of 1 person like 4ian...