GDevelop's monthly recap − August 2016

[ddabrahim]

@victor

But AGK is not free, so it’s not a solution

Well, I don’t really know about any good free alternative.

The point is that, even free software need regular updates but you and 4ian just haven’t got time to update GDevelop regularly.
If it even possible what I’m saying it would have the benefit of using an engine that is updated regularly by paid developers and if it possible with the events to keep them as a collection of Tier1 code that can be edited, extended and created by anyone then you and 4ian need to do nothing but keep the IDE running and hopefully the community going to update the events so people know nothing about code can still use all the features.

GDevelop has to stay as a free alternative to paid softwares

It doesn’t have to be GDevelop, just keep GDevelop what it is and make this one a separated product and call it different, AGKDevelop maybe and make it free or charge money for it as you wish.

Of course some people might say there are better frameworks out there for free, but the point is, the frameworks require C++ and compilers and SDK’s and what not, AGK Tier1 require nothing but it own compiler and that’s the point. If you can manage somehow to generate Tier1 code and send it to the AGK Compiler or just open it in AGK and do it from there, it would solve all the problems and combine the power of the GDevelop IDE with the power and regular updates of AGK.
If it done right, nothing going to beat this marriage.

Yes, AGK cost money, but it is on sale all the time, you can get it for 50% less in every 3 months if you wait and it is nothing compared to what AGK has to offer:
-available for Windows, Linux, MacOS and the Pi
-Tier1 code (Basic) can be exported to Windows, Linux, HTML5 and Android without requiring any SDK’s, can also export to MacOS and iOS requiering xCode
-Tier1 code can be tested on any device through Wifi (absolutely love it)
-Tier2 (C++) can also be used in case you prefer but it is require messing with compilers and SDK’s as usual.
-packed with all features most user dreams about

Free or not, I just can’t see GDevelop going anywhere unless you and 4ian cooking something but if you don’t, this is the only way I can see GDevelop stay above the water in some form.

[utunnels]

Maybe, just maybe, instead of adding more new features, you should focus on fix the bugs and polish the template games.
Provide the status of the community and wiki, a better template is easier to make new users play with it for a while.
For example, the platformer template, the Martin Mario thing can’t even die, the moving platform makes the player character jitter, and the game doesn’t have a title screen, score screen and game over logic. I know there are other tutorials and examples, but usually a new user needs the template to start.
Personally I don’t want to make a game scratch unless I have no better choices. It is far more easier to mod a game than making one. Even though I ended up rewrite all logic bricks of the platformer game, I still started from the official template. The good thing is GD has a decent IDE. But I doubt any user may find the logic part easy to use, they need to either ask on the forum(provide they are not ignored) or find a working example, so why not make the default one better?

[Kink]

On this point, everyone can create and enhance new templates. In my opinion it’s not on what does the developpers should focus, the community could work on this.

[AuShadow]

I plan on making an updated tutorial for beginners within the next month or so, just have to find the time, in just over a fortnight missus and kids are going away for a little bit so will have time then.

[ddabrahim]

@utunnels

instead of adding more new features, you should focus on fix the bugs and polish
For example, the moving platform makes the player character jitter

This is exactly my point. New feature or bug fix, improvement doesn’t matter. 4ian and Victor simply haven’t got time to maintain the actual engine. Also when it comes to new features, GDevelop is also limited by the libraries it using (SFML, Pixi.js, Cocos2D.js…etc). There is many things just can not be done because of the limitation of the libs.
That’s why I think the best would be if the GDevelop IDE would become an independent project / product and made to be used with other engines and tools so we can use the power of the IDE and event system with other engines with regular updates and active development. In my opinion AGK would be a very good try.

[Lizard-13]

I’m sure you love AGK (extremately sure), but as a free alternative to anything GD can’t be a non-programming wrapper tool for a paid software, I mean: I would still support GD here in the forum even if it turns in another kind of software, even as another engine’s tool, but only if the second is also free / open source (Godot, Phaser, Cocos, I can’t think any other good engine right now). Otherwise good bye, it lost its soul, do you get it?
But I don’t want to play along with you, it doesn’t mean I want GD to turns in any other kind of software

So, let’s see, I’m expecting right now. Everybody here is waiting to see what happens, as you, but without complaining for this or that, or insulting the works of others, the man is working!

[ddabrahim]

@Lizard-13

Everybody here is waiting to see what happens, as you, but without complaining for this or that, or insulting the works of others

I don’t think anything going to happen actually and I didn’t meant to insult anyone.
I’m only talking about facts that even 4ian has mentioned. He was thinking about to make GD to be a wrapper for other engines exactly because of the same reasons I’m talking about… But then he gone silent and finally end up implementing Cocos2D.js…

@Lizard-13

I’m sure you love AGK (extremately sure)

I’m not mentioning AGK only because I love it, but because the Tier1 language requires no technical knowledge and possibly would require no extra actions to take other than may open the project file exported from GD and press export->Android.
But I’m pretty sure if 4ian and Victor would contact TGC with the plan to make a wrapper they would be more then happy to pass the specifications of the compiler to them and so they can use it directly form GD and this way users don’t even need to see a piece of code ever (unless decide to use it from GD).
Also, it free for the Pi, completely:
appgamekit.com/agk-pi/
And packed with everything:
appgamekit.com/documentation/commands.html

If you know any free alternative just say it.
Cocos2D and the others are not real alternatives as they are requiring technical knowledge and extra actions to take even after it is exported from GD.

But nothing going to happen anyway I’m just wasting my time here
Good Bye

[victor]

Just to make things clear. We are still making changes to GDevelop, it’s just that we didn’t have the time to do the release process (I have the time but I don’t have access to the website nor the Ubuntu repository).

The last weeks, I’ve been working on improving the code of GDevelop so that it eases its maintenance and remove possible bugs.

[Zapata]

Hello, i’m a new user of Gdevelop without any programming knowledge whatsoever and as such I probably represent an absolute majority of Gd users, the reason is obvious.

So keeping that in mind I wanna make a proposition to help draw users like me into Gdevelop and make it a worthwhile experience and maybe even make us care about this project. I’m not sure if what I’m gonna suggest is even feasible and I’m probably being too naive, but anyway…

I’m talking about a community project of sorts that is tailored to fresh users, as in non commercial project meant to serve as a place where people can meet, learn, cooperate, maybe even form teams along the way and all of the potential benefits could be funneled into Gdevelop project, if a decent game would to come out of it, that would imho benefit the project, considering that having good games in Gdevelop’s potfolio just might be the most important thing… imho.

Of course I understand the amount of nuances that go with it, but would anyone be interested even discussing the posiibility of such a project? We could discuss all of different ways it could be handled, since I’ve seen one too many times them failing, for number of reasons, but I think those problems are avoidable if done right.

[ddabrahim]

@Zapata

I’m talking about a community project…anyone be interested even discussing the posiibility of such a project?

We have tried the same thing years ago:

GD open game project?

Would be anybody interested in an open GD project (including 4ian) to make a commercial game or tech demo mainly to show the capabilities of GD? We would have a game idea and a todo list. Anybody who interested, can choose anything from the list to make it in GD and contribute to the development of the project, including resources such as art and sound. Once we have a complete game, and good enough to be released as a commercial game, can be released by CompilGames on steam or desure for free …

The main reason it failed because we could not decide who would lead the project, how to work together and what kind of game to make. All kind of ideas was flying around and finally I have just found my self on my own wit my own project.
I was developing it for some time and hope the community join at least with discussions of how to get things done technically, in theory with ideas…etc but eventually the whole idea of the open project is failed…

I would be interested to join an open-source GD project, but it would make no sense to start an other discussion about what kind of game to make. If you have anything in mind, just get started and take the lead, upload it to github and eventually people may contribute.

[Zapata]

Thanks for a quick reply, I read the thread, I think one of the main issues with these kind of projects is “fixation on the end goal”, as they say - journey is more important than the destination.

I’ll try to clarify what I mean, it’s like desinging a game, where every nuance matters, when we’re talking about making a team out of people on the “internets” we run into a main problem here - human factor, people can’t care to work with somebody they don’t know, even friends can’t always get along, forget strangers getting along.

It’s imho the main problem here, before you try to make anything ya need an actual team, not just a bunch of strangers with a goodwill. First, people gotta get to know each other better, they gotta figure out if they are comfortable to be a part of a team before even discussing a particular project.

That’s why in the initial post I wrote “educational purposes”. Let’s look at what I think is the most common way of how people go about learning Gdevelop - you’re fresh to Gdevelop, first thing you do is well… learn how it works, how’s that usually going? Ya watch videos, ya read tutorials, you browse forums, etc., people usually do it alone, since there isn’t any other convinient way to do it. What I propose is to establish first and foremost a “learning platform”, and not only for learning but also for, importantly - meeting other people. I think that this is very much needed starting point for a community project to have a chance of going places. Imho ofc.

And even if nothing will be produced, it’s still might serve the purpose, it might just help with easying people into Gdevelop and might increase retention rate and whatnot. If you say - well, go ahead, make it happen, I think this requires the backing of devs and be presented in a convincing manner, not just a post by a newcomer.

[victor]

Just to inform that a new version of GDevelop is gonna be released this week. This will be a beta version as we made some internal changes to the codebase of GDevelop. You can already read the changelog here: https://github.com/4ian/GD/releases/tag/4.0.93

[sleeper_service]

Awesome, looking forward to it!

[utunnels]

@ddabrahim

So it is not an inventory then, basically it is just an array that I can use to store strings and quantities like:
“apple” = 10
“pearl” = 2
“AK47 ammo” = 127
and how to enter and use this information is totally up to me.

If so, it might should be called something different, for example in 001 the same thing is called a “Collection variable” and it seems the same thing to me except, in 001 we have a visual editor to add the strings with default values if we wan it to but other than that, the same thing. It helps with making an inventory, but it is not an inventory. Only my opinion

I don’t think I get it.
But it seems we have similar features in variable editor. So what’s the benefit exactly we have now?

[ddabrahim]

I don’t think I get it…So what’s the benefit exactly

With the inventory extension you can setup your own inventory more easily than using a bunch of variables and structures.
Of course it offer nothing that would be not possible with using variables, but somewhat more transparent way especially for people new to this. By using a single event you can simply add/remove items, check the number and existence of an item and flag an item as “equipped”. In case you have a working inventory it might don’t worth it to make one from scratch just because of the inventory extension, but in case you are about to make one now, it may be a bit easier and faster to setup an inventory with the inventory extension.
Also more optimal I guess.

It may mislead some people though because when you read “inventory” people may expect something complete that you can find in RPG Maker and 001 but it is not the case, it is basically just an array and some events dedicated to help with making an inventory but definitely not a complete solution.
This is why I was suggesting to call it something different, but since it is become “inventory” anyway, it may would be practical to develop it further and give people at least a basic, fully functional inventory system.
Nothing fancy:

Let say, we would have an action "open inventory “name”. When we run this, let say

If key "I" is pressed : open inventory "name"

it would open a basic inventory “prefab” something like the inventory in Minecraft, many little squares inside a big square and this inventory would read the content of the selected inventory and display the content. Then we can equip, use, move, drop items as usual.
Of course, which item do what would depend on the user, we could have an event like this:

if item "name" clicked in inventory "name" : do whatever you want

I believe with such a simple condition, we could do many different things, like:

if item "medkit" clicked in "player" inventory : Do+20 to Variable(player_health) if item "apple" clicked in "shop" inventory and Variable(player_money) >= 20 : Do-20 to Variable(player_money) and Do-1 to item "apple" in "shop" inventory and Do+1 to item "apple" in "player" inventory
And tons of other things really.

Finally, of course it would also require some sort of table to be able to easily set what image or sprite represent which item.
“apple” = apple.png or object
“AK47 ammo” = ak47ammo.png or object
And so on. When the inventory “prefab” let just call it this way, load the information, list of items from the selected inventory “array” then it could just simply look in the table what image or sprite object representing which item and use it to display in the inventory.

I don’t think it would be that difficult to implement and it could make the inventory extension more practical and complete. But I’m also sure about that people would start complaining about inventory look, they want it to look different and about missing features and features they want to work differently. Different games may require different inventories, to look different, to work different so I don’t know if something like this a “built-in inventory” worth the effort after all.
It is just an idea for Victor and 4ian, as always

[Wendigo]

This year is coming to it’s end. I’ve seen that victor and florian were quite buisy pushing updates to GD on github during christmas, thanks a lot for spending your freetime.

May I ask what your plans for the upcoming year will be?
The Trello Roadmap has still some “Ideas” left but the “Doing” section is currently empty.
While there are some interesting things in the ideas section I’d like to add two points I consider really important:

• Better interaction between the different extensions (i.e. using physics inside a tilemap)
• Achieving feature parity between Native and HTML5

[4ian]

Hey! I’ve updated the roadmap with the latest changes.
Basically for now I’ve been working on improvements and extensions that can be used for the Lil BUB game I’m working on. In particular, I’ve implemented time scale that can be changed per layer: trello.com/c/Xk7whysX/36-per-layer-time-scale
And spent a lot of time trying to improve HTML5 games performance (for example, fixing lags in Android that would freeze the game for 0.5/1s quite often). I’ve made on card for it on Trello, for the record: trello.com/c/MFZbRgZB/38-perfor … on-android

On my side, the big ideas for the upcoming year are:

• to finish the Lil BUB game, this will surely still bring some improvements and fixes to GD
• then see if some things can be simplified. The software is quite massive and I’m sure some part could be improved or even simplified, improving the whole quality of the software.

Nothing more specific for now but we’ll see! I’ve been doing a few interesting research to see if the whole editor could be updated with web technologies so that is more easy to make it evolve.

[Rex]

Can’t wait for the update! Having the ability to pause layers are so vital for my game and i would definitely appreciate it.