I know construct 2 sometimes exports animations by building sprite-sheets, as i've read in this blog post.

From what i understood, spritesheet exporting is hardcoded to only happen with sprites when a game is exported, and third party plugins that use the animation editor will still export a series of images. My question is if this has changed, and if i'm understanding things right. I really don't want / can't have the images exported as a spritesheet for my particular purposes, so i need to make sure it will never happen.

I create an entire webGL canvas above the construct window for my Q3D plugin, and perform lots of complicated rendering. Not sure how well it works on mobile, but it works fine in nodewebkit.

I think you can force chrome to use your gpu if you enable "ignore-gpu-blacklist' in chrome://flags .

not sure if this would help in your case but may be worth trying.

Is there a way to new Object Type Properties using the sdk?I haven't investigated but i haven't figured out a way to and was wondering if someone knew.

If there is no capability to do so, it would a nice addition to the sdk Ashley . Right now i have to make hacks to enable type wide features being set from the editor through the properties of an individual instance, and then having to set something that's part of the type rather than instance using the instance property in runtime, it's weird and feels wrong, and just adds confusion and sloppy workarounds that are confusing to the end user.

CSG operations are really complicated (and slow), i don't plan to add them to the plugin as even optimized native programs slow down with them, I don't think they'd be very viable in a browser for any use in games that warrants even trying to add them.

DrewMelton

Well your zordering with terrain would become non-existant if you decided to layer 2D sprites on the 3D background, everything would be above the background no matter what. Animations are not supported yet either. I don't think the current version would be something you could easily use for that particular game yet. I'd say wait for the updates before purchasing and see how those will suit your development style.

DuckfaceNinja

I'm waiting till I have more finalized version to make a manual, but i do plan to make one. Probably the next thing after the next update will be a manual.

DrewMelton

At the moment its difficult to mix 2D and 3D assets as the 3D window take up an entire layer, so you wouldn't be able to Z-order your construct sprites with the 3D content for example, as this would require some way of magically pushing them through the depth buffer. I don't think you'd be able to easily do what you're asking unless you were ok with the sprites always being above the background (i.e. nothing hidden behind walls) I don't know if this is clear, but imagine you had to put your background entirely on one layer, and your characters on a layer above. This wouldn't be a problem for some games, like lets say the dungeons in 2D zelda games, but in isometric it might look weird, i.e. not being able to go behind something you should, like lets say the small squares of dungeon wall in that picture, so the way you interact with terrain would need to be modified to "hide" that fact. Feel free to ask more questions i don't know if my answer is clear enough, as I've mentioned Q3D is difficult to use and doesn't hold your hand in anything at the moment, so you'd have to be ready to make a lot of low level systems and understand some of the limitations with how lights work etc. For example you can't always have more than 4 lights at a time effecting any one object; these are little weird things that are inherent to how shaders are handled on a GPU with ANGLE. You kinda have to be ready to scratch your head and work things like this out and understand there are limitations with the current plugin, and really with WebGL in general. Also the plugin is only made to work with browsers / Native Node-Webkit. Q3D makes 3D in construct possible, but certainly not very easy.

I'm planning to eventually (although this wont be anytime soon), create custom sprite objects based on the construct sprite objects that can be used with the plugin, but as of now there are other features i'm working on. Really if 15$ isn't much to you it would probably interest you to try the plugin, but if you feel uncertain about spending the money for something you may not use, I wouldn't recommend you purchasing it yet. If you feel uncomfortable working around limitations, or coding complicated code, or using an unfamiliar workflow, i wouldn't recommend it. I'm working to improve usability as I've stated, but updates take a lot of time, and realistically I can't make everyone happy.

QuaziGNRLnose

would you so kind to tell me how long it would take till update?

I already messed up multiple deadlines i set for release and now I'm back in school, so it'd be difficult to make any promises. I'm working on optimizing the collision system as it's not performing well enough to be usable yet, which means object recycling and more complicated sub systems are going to be needed, this will require time. I started work on how to do this today. There's a lot left to be done.

Hi, i'm the developer of the Q3D plugin.

I understand the dream you have. I love the design of constructs programming interface, and would love to have it for something 3D, and that's why i'm currently developing the plugin. Right now it's very basic and many people misunderstand the complexity of 3D games and expect making one will be anywhere as easy as making a 2D one. Everything Ashley is saying is completely true, 2D engines are much more widely accessible because of the fact drawing is such a basic skill, and the fact most math you're going to use for common things doesn't really require anything beyond high school trigonometry. He's not really insulting anyone's intelligence by claiming most people construct is aimed at wont have studied the more complicated math and sub systems involved yet. He's simply stating construct is meant to appeal to a demographic which doesn't really overlap the demographic who are capable of working in 3D.

I think i'll be able to somewhat achieve the goal of "easy" 3D using construct, but there are always caveats. People will need to understand that lighting and shader building, preprocessing of lighting, the vector math and linear algebra and every complicated trick which is situation specific and complicated about 3D cannot be simply tucked away under the hood as an automatic process which the engine will take care of. In a way my goal is to move towards doing that with future updates for Q3D, but as it stands now the plugin is meant to enable 3D and nothing more. Doing what you're asking for in its entirety is very much easier said than done. 3D games require careful engineering in order to perform properly in a variety of different situation. There are a lot of challenges i face in developing a plugin to make 3D work for an audience that will have very high expectations about performance and ease of use.

I think you aren't really looking deeply enough into the repercussions if scirra officially focus on making their tool support 3D. A lot of people could unanimously agree that Construct is an incredible 2D game engine, really the best around. If they were to focus on adding in 3D or a new 3D tool, that would mean they really wouldn't be able to keep making it the best 2D game engine around. They would have to put an incredible amount of effort to get something okay, and people would likely be unhappy because the reality of 3D game development would clash with their expectations. There's no point for them to stop working on a fantastic 2D tool, to make an alright 3D one, which would really never compete very strongly with better established 3D tools like unity or engines like Unreal.

I think the devs should focus on making the Construct SDK better so that more people can make construct do more things than it currently can. Right now extending functionality through the editor isn't possible, and we're left to basically only have some basic object parameters to control everything. the ability to do more complex drawing and operations in the layout, launch popup windows to organize properties better akin to custom "image editors" etc. would be a great step in making higher quality plugins possible.

The problems lies not with google play but with consumers of mobile games being accepting of very shallow games with some shallow element they appreciate, or of games that can be easily imitated. If a game is easily imitated it will be easily imitated. If someone will play anything because it has pikachu in it vs. something of quality with original content, then it's not really the fault of google play. They serve popular demand, and mobile gamers aren't know to look for quality, they play whatever is popular.

[quote:3dj3s36f]- Easy loading of 3D models in .obj, and three.js JSON format

Does this mean they can be loaded at runtime as well? Lets say if I use Node-Webkit and I could have folder for the objects and load them from there?

Also do you have any plans to put this plugin to Scirra store? I don't have PayPal.

If I put it on the scirra store the price would be elevated, so far there are no plans to do this.

You load objects through filename if they are in the files folder. I haven't used Node-Webkit extensively for loading, but i'm pretty certain what you're suggesting is possible. If you can provide a valid file it will be able to load it. Sorry for the late reply your post got past me somehow.

well we are not all novoice , i am certain it would be useful for me at least , i may not understand everything , but if a code work u dont allways need to understand the full depth of it , and we dont learn more whitout new challanges so thats my opinion . u can pm me and i am sure i will find some usefull stuff if u may , i shall keep it for my self and only use some code if I may ?

coz I want to apply physics suspentions on the wheels .

http://web.archive.org/web/200804101716 ... dc2001.htm

This is (roughly) the technique used to simulate the tank physics in the demo. Reading this is simpler than reading through the cap and it's actually meant to be explanatory instead of a pile of virtual spaghetti. These are programming related questions which the plugin does not seek to answer.