In this post sleeper_service mentioned that GD4 was very slow when working on a larger project.
Is GD5 able to run with WebGL? Could that speed up the scene editor.

4ian, at the link in Wendigo’s post, sleeper_service said he tried opening and previewing very large projects and he had to wait for five minutes because of background compilation. Is the HTML5/JavaScript platform unaffected by that problem? As far as I’m aware, JavaScript is never compiled, so is it immune to the problem mentioned by sleeper_service? That would be great! I became scared when I read his post because my game might become very large after more years of work.

I’m anticipating GDevelop 5.0 stable!

You’re absolutely right, don’t worry about compilation time for HTML5

Eh, just saying if you have a bunch of images and sounds to be loaded even the HTML5 preview might take some time for the first time at least. Just saying because some people expect things to start immediately as this is “only a preview”.

What some people don’t realize, the only way you can get immediate results in case of some engines is a WYSIWYG editor as the complete game essentially running in the editor and you are editing the game in real time. The editor of GD is NOT WYSIWYG so better expect some time between pressing preview button and the game start running even in HTML5 and especially in case of native.

Beta 11 is available : compilgames.net

With JS event support (very simple for now, no syntax highlighting - I made it as quickly as possible), copy/cut/paste in project manager, CPU usage reduction in the scene editor.

HTML5/JS is not affected by this problem. The only thing that can happen is when the game becomes large to have to wait for a few more seconds before the preview is launched. This is mostly due to having to copy project images (this could be improved).
If native game support is added later, I’ll try to improve to avoid having background compilation taking too much time and eating all the CPU.

In the current state, HTML5/JS is more adapted to large games than the native platform.

Yes. Scene editor and the game engine are based on Pixi.js, which is running with WebGL and is the fastest existing 2d renderer with webGL.

GD5 fully work without internet connection once it’s installed

Same, download the full version to have this.

Another question : do you think you would be interested by an automatic service that would be building an android app (.apk) directly from GD5?
I’m considering having a service like this to ease the building of mobile games for new users. Would have a small fee/subscription surely, as there is some infrastructure to set up/maintain but should be no more than a few euros / months.
If you’re not interested, do you think that’s something that would interest other users ? (Knowing that you are still able to export as now, it’s just a service to have a build-in-one-click. Could have the same for desktop export later : the default manual one and an online “one-click” service)

Thanks for your feedbacks

Personally, I’m not interested in developing a mobile game right now so I would not likely pay a subscription fee for only that.
But if it would be highly optimized to get the possible best device compatibility and performance and it would include a “Player” that we can install on Android and iOS devices and preview our games remotely on the device without need to export/copy/install the game it would be definitely something worth considering if you are developing a mobile game in GDevelop.

But as soon as you introduce a monthly fee you can’t avoid being compared to Construct 3 and to be honest, it is offer a lot for relatively low price. In my opinion you need to go at least 50% below the price of Construct 3 or even more if you plan to offer nothing but a build service to be used with GDevelop.

What Construct 3 also offer that you need to keep in mind when you set the price:
-remote preview

• you can store all your assets and projects in the cloud don’t need to use 3rd party cloud storage
  -you can download the project including all the assets to save it offline or share it outside Construct
  -can build/export directly from the web to any platform including PC and mobile no need install and launch an offline version of the tool and use 3rd party services
  -free signalling server for multiplayer games for users of Construct3 even for commercial purposes

And on the top of this the actual engine also offer that GD does not:
-networking/multiplayer
-dynamic lights
-support for shaders
-sprite blending modes

To mention few things C3 offer for it monthly price

I think it is a good idea but not sure how low you need to go with the pricing to make people pay at least that much it worth maintaining.
To make me subscribe, a build service would be not enough on it own. I would probably subscribe only when I’m about to ready to publish my game to mobile devices and only for that amount of time until it done if there is nothing else offered for the subscription.

I have the feeling you do need to offer something for free in order to get this started. For example you can have a build project for free and if you need more than you need to pay. Or free builds take longer for example and need to pay to remove limits.
I believe this is the only way it could work for you and for us.

Also, PayPal for subscription would be nice. I don’t want to use debit card online.
I have recently looked in to electronic banking to get a top-up card because I want to pay some online courses that don’t accept paypal unfortunately but the fees they charge for every single transactions plus service charge and the fact some of them honestly note that they CAN NOT guarantee the security of you money is ridiculous.
PayPal makes online purchase so much safer and simple on my end. I know it makes things more complicated on your end, but hey, customers are first and always right. Right?

If you don’t offer PayPal you need to offer a lot to convince me to use my debit card to subscribe. I haven’t decided yet on electronic banking, it is just a joke, as I plan to use it mostly on small subscriptions like 10-20 euros here and there, because of the transaction fees it would almost double my annual expenses.

**uck this community.

It is a pretty important decision to make if it worth wasting time and money to set all this up or not.
Even if you are not interested you could share some opinion.

Would you use it Yes:No?
Do you think beginners would use it Yes:No?
If you are interested, how much would you pay for such service?

For God sake.

@4ian
In my opinion you should go for the money in 2018. Maybe less people would use it but what difference would it make?
Make the core free and open but charge a licence fee for the new editor.
I’m willing to pay approx 30 euros for the new editor. You need to make your decision now while the editor is not complete and the parasites don’t have it all.
With C3 require a monthly fee, even a one time payment model could work for GD.

Desktop version → 30 euros
Free for students and contributors (contributors compile for them selves).
Web version free with the option to pay 5 euros monthly to export using the build service.
Also introduce plugins so you can offer new feature in plugins and charge money for each additional feature, also allow people to make and sell their own plugins.
It would make you money while still receive contribution (in hope of making money) WIN-WIN for everybody.
Instant C3 killer if you ask me.

Marry Christmas and a Happy new year.
See you all next year.

Not interested. I am more focused on pc now.

I won’t use the build service for Android, but new users might love it. I’m afraid of 4ian designing the build service and becoming sad if very few people use it.

4ian, are you willing to charge a one-time fee like ddabrahim mentioned? Were you saddened by that suggestion, or are you considering it? I would pay a single fee for every major version, like 5.0, 6.0, 7.0, but not a subscription. You could give free updates for at least one year, and then release a major version that requires another payment.

I won’t buy Construct 3 because it asks for a subscription of $100 USD per year and I cannot own it forever. You could charge $50 as a one-time payment for each major version, and release a new major version every year, but people would always fully own the version they buy, even if they never buy the newer ones. You should only add bug fixes during the second year, and then abandon it.

Example of the idea:

Version 5.0:

2018 - full support
2019 - only bug fixes
2020 - abandoned, but previous buyers still own it and can still sell games they make with it.

Version 6.0:

2019 - full support
2020 - only bug fixes
2021 - abandoned, but previous buyers still own it and can still sell games they make with it.

You could also try creating a Patreon page. The Godot engine is fully open-source with the MIT license, but they have one full-time developer and one part-time developer because they receive over 5,000 USD per month from Patreon. That would allow you to fulfill your vision of an open-source masterpiece while still receiving money.

I tried Godot and it was very difficult! My plans that are easily achieved with GDevelop are much harder with Godot!

I have this piece of advice:
Completely stop supporting the native platform, and use that extra time to make the best HTML5/JavaScript 2D engine. I hope people are not angered by that suggestion. Our developer probably has college classes and a job to focus on, so splitting time between HTML5 and Native might not be wise.

Thanks for your feedback and your ideas

I’m still trying to find a good way to have a business model around GD, while keeping it open-source. As you mentioned, a one time payment is a solution, but currently a good advantage of GD is that it’s free. Another is to build services around GD, for example a build service for Android and maybe Windows/macOS/Linux (advanced users could still do it manually so you’re not locked in).
This could also be specific new features like for example a Debugger for games, accessible from a one-time fee that would unlock all the advanced features.

Also one time fee vs subscription is an interesting debate, but both can be done : you could either subscribe to pay every month or pay one-time for an unlimited use.

Still thinking! My goal is to keep the whole software open-source but find a sustainable model for developing GD.

The native platform is very useful in my opinion. And also about the fees, the whole reason behind I using gdevelop is because it has all the features i need and doesn’t require fees like other engines. It would be a huge disappointment if you implement charge to it. I know it might sound selfish but i as a new indie dev , do not have any income to start with and i need a free good engine like gdevelop to really get started. If you need money, you could setup donation sites for us to donate to you.

This is exactly why I think my suggestion extended with the idea of wild master to pay for every major release could work.
Have a think about it.
It would be open-source but people need to pay for convenience. On the desktop it is IDE and later export options.
You could charge a one time fee for every major desktop release as wild master suggested.
If people don’t want to pay, that is fine but then they need to compile from source for them selves or use the web version which you can take advantage of through subscriptions like a build service, game hosting, assets, additional features or even ads. And if you can introduce an easy plugin system which would allow us to really easily add features using JS, and sell these plugins people would more likely contribute in my opinion in the hope of making money.

You made GD free and open-source with the hope people going to contribute with code and features. Some may did, but in my opinion it is failed if we are looking at the big picture. It time to try a business model and I truly believe my suggestion could work simply because it works for everybody. Offline users, online users, free users, contributors and developers. Everybody wins.
GD could evolve faster with a business model imo rather than waiting for free contribution.

I thought of an idea for a compromise and I want to know what all of you think.

We all want GDevelop to continue as an open-source project because we don’t want the source to be lost after 4ian dies. Will Construct 3 have a legal plan to pick new owners when the current owners die? Will the new owners care about the customers as much as the old owners? Does Clickteam Fusion have a legal plan? Does GameMaker have one? Some of us are worried about our games not fully being ours if we use closed-source engines.

BUT, we also know that 4ian only has a few moments each week to work on GDevelop because we’re not paying him.

So, as a compromise, I have an idea that might mostly satisfy all sides. He could release one major version every one or two years, and its source code will not be released. After a new major version is released, he could release the source code of the previous major version.

It would happen like this:

1. Version 7 is released and costs 50 dollars. The source code is not released.
2. Since version 7 is being sold, the source code of version 6 is released for free.
3. People will still be allowed to use version 6 to sell games, but it won’t receive updates.
4. People who cannot afford to pay will always have a free version to use.
5. Without funding, new versions of GDevelop might never even be created, so people who cannot pay shouldn’t be upset. The older version will still be great!
6. If 4ian decides to leave the project, he could release the final source code when he says goodbye. If he dies, his legal documents will say that the source code of the final version must be released to the public. Even if he decides to give ownership of the final source to his son, and his son becomes so diabolical that he ruins the product, we will still have the source of the previous version to fork and improve. Hahahahaha!

Is this a bad plan? I’m not joking when I say I am afraid of the owners of engines dying. I own many engines, and I feel fear when I think about creating a beautiful game and being unable to preserve it for the future when the owners of the engines die, so I constantly come back to GDevelop. But I also want 4ian to become rich because he deserves to be recognized for his tremendous gift to us, so people who want the newest features should buy the major versions! I will gladly pay for this business model! It would be the best 2D engine that is also a public resource for future generations, and the newest portion of its source code will be only temporarily private to financially maintain the developer’s life!

I’m not 100% sure it a good idea to release the source code only to the previous version because this way people can not contribute to the current version. The point of my suggestion would be that the web version of GDevelop would take over the “spirit” of being a free and open software that is open to contribution and free to use because it is important to 4ian after all GDevelop is made using free and open frameworks and libraries. Holding the source code back would be against everything an open and free software stands for.

But, because it would be the web version that is free to all and hosted by 4ian he can still offer additional features for a small subscription fee that would cover the cost of the hosting such as a build service, royalty free assets, remote preview to mention a few possibilities that could be web version only. I think it would be more friendly than the subscription only plan of C3 but it would be still sort of carry on being a free and open software.

But, the idea of make the source of the previous version available remind me what 3DRad was doing. What they did was, every time a new major version released, you had to buy it and the old version become available free for everyone.
So, you sort of paid for getting the latest updates and features 6-12 months earlier. That could be a solution too but it would require to offer ground breaking updates and features to motivate people to buy the new version. This is what 3DRad failed to do very well and I’m not too sure 4ian could do it in part time. But I think it would be nice addition if all major releases require a one time payment and the old version become free to all. So in this case, GD5 would require a one time payment and GD4 would be the free version. When GD6 release, GD5 become free to all. But, the web version carry on being up to date all the time completely for free and offer services that may even make the desktop users to subscribe and use the web version instead of purchasing the next desktop version. Such thing could be the build service, remote preview, royalty free assets, Facebook API integration to mention a few.

But I think, the source code need to be shared otherwise GD can’t be a true open software and this way I believe even your concerns regarding the ownership in case of death would be solved as the new owner could not take away the version already shared for free only the one that is about to become free. And because the most recent source code is shared at all times, in case there is no new owner, someone might be able to take it over and continue it development but most importantly as long 4ian is alive GD can receive contributions and be a free and open software while also offer paid options.

As a side note, GDevelop 5 beta 12 is available compilgames.net

As a Christmas gift, you can find a Preferences menu where you can choose the new Dark theme: twitter.com/Game_Develop/status … 4362702854
If you’re programming a bit, you can create new themes: github.com/4ian/GD/tree/master/newIDE#theming

Merry Christmas to everyone!

Thanks for your thoughts about GD business model!

Yes indeed! It’s important I guess to keep games working on a game engine that is open-source (to keep being open to contributions, improvements, maintainability in case I don’t have time etc…).

Having a free web version with additional features for a small subscription fee looks like something that is not going to disrupt users but can still work. I’d like to try this. Not sure how to handle the desktop version. For now it’s the full “featured version” as web does not handle export nor custom resources (but this could evolve). Could have the same export options as now + the paid build service like the web version.

The dark theme looks great, thanks a lot for that

The problem is that if both version share the same features there would be little reason to use the web version. There need to be something that set the web version and the desktop version apart. My idea was that to make the web version the free option and the desktop the paid option exactly because the desktop version is the fully featured version and you can more easily offer paid features to the web version and have a business model on both while GD can be still considered a free and open software after all.

But If you would like to offer both version for free and offer only a build service for a small fee, an other option could be is to make both version completely free for making/exporting HTML5 games and add a build service to the web version to export to Android, iOS, FireOS and a build service to the desktop version to export to Windows, Linux, Mac which would require a one time fee or monthly/annual subscription. So the desktop would be targeting desktop, and the web version targeting mobile with remote preview I think it would be very important to offer remote preview for mobile in the web version and both of them can be used to target HTML5 completely for free. I know there is no reason to make the web version do not export to desktop and make the desktop version do not export to mobile, but the way I see it in case you want both version to be equally fully featured and exist next to each other there must be something that set them apart from each other. Desktop → Desktop and Web → Mobile could be used as an excuse for having both exist.

An other option could be just popped out of my head as I’m writing this, probably it is not what you want but I just leave it here for you.
The web version could be a sort educational version with lots of templates and free assets like 3D Gamemaker was:
youtube.com/watch?v=1_sy7D3HmZ4
There would be no option to use custom assets but using only the ones included and you can charge a monthly fee for more assets and templates to choose from. This way both the desktop and the web version could target the same platforms using the same build service and both have it place in the world while nothing stop you from evolving the web version to become a fully featured version one day to replace the desktop version if/when it is become reasonable step to take.

You probably want both the web and desktop equal but in my opinion it is just not possible, this is why Scirra decided do not offer a desktop version for C3. It would ruin the web version instead they turned the web version in to a hybrid that also works offline but still require an annual fee to access all the features. It is something that you can also consider except you would offer a truly free plan not just a limited trial like C3 does.

new version keeps crashing when I try adding new scenes and associating external events with scenes. adding and associating external layouts works just fine.

Strange, nothing should have changed on this part.
Can you send me a screenshot/video/GIF of the crash? Can you send me the game that you’re using and the steps to reproduce the issue? Thanks!

It happens even starting a new project, here’s what the console reads:

• Loading builtin extensions… BuiltinObject, Sprite, BuiltinCommonInstructions, BuiltinCommonConversions, BuiltinVariables, BuiltinMouse, BuiltinKeyboard, BuiltinJoystick, BuiltinScene, BuiltinTime, BuiltinMathematicalTools, BuiltinCamera, BuiltinAudio, BuiltinFile, BuiltinNetwork, BuiltinWindow, BuiltinStringInstructions, BuiltinAdvanced, BuiltinExternalLayouts, done.
  main.f6af7bbf.js:1 * Loading other extensions… PlatformBehavior, DestroyOutsideBehavior, TiledSpriteObject, DraggableBehavior, TopDownMovementBehavior, TextObject, AdMobObject, PanelSpriteObject, AnchorBehavior, PrimitiveDrawing, TextEntryObject, Inventory, LinkedObjects, SystemInfo, Shopify, done.
  main.f6af7bbf.js:1
  Pixi.js VERSION - ✰ WebGL ✰ pixijs.com/

main.f6af7bbf.js:1 TypeError: Cannot read property ‘wrappedInstance’ of undefined
at t. (main.f6af7bbf.js:1)
at wt (main.f6af7bbf.js:1)
at commitLifeCycles (main.f6af7bbf.js:1)
at t (main.f6af7bbf.js:1)
at _ (main.f6af7bbf.js:1)
at x (main.f6af7bbf.js:1)
at batchedUpdates (main.f6af7bbf.js:1)
at Q (main.f6af7bbf.js:1)
at je (main.f6af7bbf.js:1)
at HTMLDocument.r (main.f6af7bbf.js:1)
e.(anonymous function) @ main.f6af7bbf.js:1
main.f6af7bbf.js:1 Uncaught TypeError: Cannot read property ‘wrappedInstance’ of undefined
at t. (main.f6af7bbf.js:1)
at wt (main.f6af7bbf.js:1)
at commitLifeCycles (main.f6af7bbf.js:1)
at t (main.f6af7bbf.js:1)
at _ (main.f6af7bbf.js:1)
at x (main.f6af7bbf.js:1)
at batchedUpdates (main.f6af7bbf.js:1)
at Q (main.f6af7bbf.js:1)
at je (main.f6af7bbf.js:1)
at HTMLDocument.r (main.f6af7bbf.js:1)