People aren't attracted to Construct because it's html 5 based - it's the great workflow.

I might be in the minority, but I was (and still am) attracted to Construct 2/3 because it is HTML5-based.

My target audience prefers browser-based interactives (education). I do see the point for those who are trying to make money strictly off of games, however. It is hard enough to make a living off of games, and to choose a tool that restricts you to only a small fraction of your potential market is financial suicide.

That being said, I would love to see a great 3D game built in Construct 3 that showcases HTML5's capabilities. The key is that enough people would need to play it and become inspired by it to encourage more developers to shift over to the platform so that more great games would be built using the tool. A lot of stars would need to align for this to happen.

In the end, I love HTML5 because pretty much every kid in school is carrying around a device with a browser, which makes the web such an awesome way to democratize the art form of games. I think Construct 3 has the potential to get there, but am worried about the financial risk to developers.

X3M, is there a current list of features and limitations for the B3D plugin? I am working with an artist on concept art now, and am planning to make a fully 3D game to showcase what can be done in C2/C3 with a 3D engine. We just need to know what the limitations are compared to the Unity WebGL exporter.

So far I have:

1. no particle systems

2. no mesh colliders (a.k.a. we can't use the terrain editor)

3. no "billboard" 2D sprites

Am I missing any other big ones?

cjbruce I think I've found a solution for this issue, I'll be posting it on my plugin thread in a while.

X3M

Sweet! Thanks!

After reading what I wrote, I just realized that we probably just need another macro to rename the screen files. With C3 currently exporting as .zip files anyway, it wouldn't be any extra work to make a macro that unzips the screen and renames the files, then unzips the controller and renames its files. Then just copy both of the resulting directories into a target directory on the server, and we would be all set.

Like I said, regardless of who it matters to, I don't want to add features that break other features.

Why can't other files be hosted alongside index.html? This sounds like an arbitrary restriction that could potentially break other C2/C3 features as well, so it seems the best solution is to allow this.

Ashley

No worries. I totally get that you can't afford to break the existing system. I have found a workaround for my client to meet their requirements.

As far as AirConsole goes, I wouldn't actually need to keep everything out of the root directory, I would just need to be able to merge two projects into the same directory:

AirConsole needs both a screen.html and a controller.html in the root directory. Up until now I have coded a very simple controller.html by hand and done all of the heavy game logic in the screen.html using Construct 2. I would like to use Construct 3 for both the screen.html and the controller.html so that I can do full mini-games on the controller while the main game is running on the big screen. The problem is that if I want to make and test a change on the controller, I have to go through a lengthy process of exporting both projects and deconflicting them.

The process currently works like this:

1. Make a modification to the AirConsole controller project.

2. Export the AirConsole controller project to a folder. This overwrites the existing c2runtime.js and data.js used by the screen.

3. Run X3M's macro to rename index.html, c2runtime.js, data.js, and all associated references to something different.

4. Export the AirConsole screen project to the same folder.

5. Rename the the index.html to screen.html.

6. Test the change on a server using the AirConsole API.

On my machine with my existing project, testing a single change on the controller takes about 8 minutes. It really slows things down, and I have had to keep the controller really simple as a result.

In order to make the workflow for AirConsole go faster, I would like to have a way to put the data.js, c2runtime.js, and potentially all of the graphics/sound assets in separate named folders: one folder called "controller/", and another called "screen/", and update all the references to and from these files in the project. All of the other files could stay as they are because it doesn't matter if they overwrite each other.

Could we possibly do an "Export to AirConsole Controller" and an "Export to AirConsole Screen" button?

Ashley,

Thanks for the prompt response! Neither my client nor AirConsole want offline support, so removing sw.js is not an issue for them. Would this be a lot of work to do on your end?

When doing work for both a private client and in creating a game for AirConsole, I needed to be able to create a different folder structure for the exported game root directory. In order to make things work, the only thing allowed in the root directory was the index.html file -- everything else had to be in a subfolder. Here is an example of what the root directory needed to look like:

index.html

[directory_of_game_files_where_I_could_specify_my_own_directory_name]/

I spent a lot of time editing the exported project files to accommodate a different directory structure. This used to be much easier in Construct 2 before data.js and runtime.js were split into two separate files. In any case, it would have been so much easier if there was simply an export option to specify the name of a subfolder for the project.

Would it be possible to add a field in the "Export Options" window where we could specify a subdirectory for export?

I just recently got started trying out Q3D again. So far so good, but my only issue so far is getting character animations in. As I'm using Maya LT, and there's no exporter to the js file type I'm kind of stuck. There is an exporter for blender, and full Maya, but I all my models are created in Maya LT. I'm going to try Babylon as well as i know it has a tool to covert fbx files to babylon.js

tunepunk, I'm not sure if you are in the same boat as me, but I ended up using .obj files for anything that didn't need custom materials.

First Impression:

A much more restricted (currently unusable) version of C2 in a browser on my mac.

I am *thrilled* to finally be able to work in Mac OS. It is an incredibly liberating feeling after spending years using either an old junker Windows laptop or VMWare Fusion (which was recently lobotomized by the Windows 10 upgrade).

I would love to be able to open up old C2 projects to see how they work in C3, but can't due to both the event limit and due to the lack of 3rd-party plugins (LiteTween, mostly).

I don't mind the fact that it is running in Chrome, as I will only use Chrome for Construct 3. The file saving issue bothers me quite a bit though. I do lots and lots of small projects, and not being able to specify local download folders could make things difficult and messy.

All in all, I see C3 positioning itself really well for the US high school chromebook districts. It should be a snap for students to use if they already have chromebooks. Unfortunately, my district is iPad-only, so we'll have to wait to see what happens with iPads/Safari over the next few years.

I can't get Super Ubie Island on consoles and i really wish i could.

This is a major issue when trying to develop bigger games.

This issues has me seriously looking into Game Maker Studio 2. But I really do enjoy the event system in Construct. Big reason why I am so on the fence. But I can't keep developing projects without a way to deploy to consoles

This thread has really got me thinking about the strength of Construct 2/3. Most of my work is expressly for the web -- I mostly do interactive simulations for my students, and have no need to port to any other platforms. I might well be in the minority here, but Construct 2/3 is the perfect tool for me: I can build quickly, deploy instantly, directly to my target platform.

I seem to have a fairly good idea but I have no clue how to make it work. I was thinking about something like "super crate box" but with more characters to make it harder. If you go and play it leave a reply and let me know how to make crates randomly spawn and how to make them change what you have as a gun....

Give it a go! There is nothing like a game jam to put the pressure on to finish. I also recommend trying to cut it to a bare minimum of features, then polish those as much as you can.

Random boxes should not be hard to do, but try manually placed boxes first.

https://www.scirra.com/TutorialViewFresh.aspx?ID=9456

This tutorial uses a top-down perspective, but you would build the level layout in 2D either way. In the tutorial, the buildings are randomly generated, but it is pretty easy to place them manually as well.

cjbruce Yes since it is the leading javascript 3D library ( Gets updated each week, it performs well, has exporters for several 3D softwares, etc ...)

X3M

Awesome!

It looks like I might need to port things over to Babylon. I'm looking forward to seeing what additional capabilities it has over Q3D.