smebor

Wow thanks for showing off your project

—

I've been trying to fix that for a while, but I haven't been able to yet.

hgr

setters/getters using actions/expressions

QuaziGNRLnose, where is examples of touch triggers?

In the examples.zip you can download on the first page, there is a viewport example which uses the mouses position + a view-port condition to change the color of mouse overed objects. You could modify this to use a touch position instead of a mouse, and only fire when "on touch" or something similar.

—

just have a bit of touching up, but this is a busy week for me so im not sure it'll be released this week.

A slope? What do u mean like a 3D ramp shape?. That would still be a box in editor, it knows nothing about the model

—

Not sure what you mean by "without an ID". Do you have the latest version of the plugin ? There are many actions for setting material properties over the whole object, you only need an ID if you're using model materials.

About the touch triggers, what exactly do you mean? touch as in clicking on a model ? You can do this with the Q3D Viewport object, look at the examples.

Ziggums

Sadly the construct 2 SDK doesn't allow for me to make this plugin display the actual model in the editor, however you see a simple orthographic view of the XY plane with a proxy of the 3D object (box/sphere/AABB) that somewhat represents it, and is properly transformed by the rotations. When you run the project in the default view settings you can see a similar result as what's in the editor. It's as good as i can do with Construct 2, but hopefully when Construct 3 comes out i'll be able to render stuff in the editor and have a fly-around camera.

I put the editor and the corresponding project running (without any events or anything to move the camera from its initial settings) side by side in the following screenshot so you can see what you do / don't see.

If i was using a model, i could change the shaded boxes to be wire-frame, and i'd see only the wire-frame box/sphere in the editor, but then the model and wire-frame object at runtime. The colored objects are a point light (green diamond) and spotlight (red cylinder with pointer)

If you use Q3D for rendering, it implements a proper Z-Buffer and would correctly render things based on X-Y-Z without the ordering trouble . It has a 2D sprite object you could use so its actually works well for isometric stuff too, and also gives you the freedom to add 3D stuff if you have to and use 3D collision detection, you could also use normal maps on 2D planes and add lights too. There's a bunch of features that would help with this type of project, since ISO is really mostly just 3D game development with 2D sprites instead of models.

irina

It'll depend what you do / want. Why don't you just try it to see? I can tell you its possible to do what you want, but you'll need to try playing around to understand what might go wrong / the steps you need to take.

In anycase if you can make it in construct, you can always just add models over the 2D stuff and then hide the layer with the 2D stuff.

3D Physics plugin is done, should be released in a day or two

I'll be devoting my time to skeletal animation next!

irina

Theres no issue having the Q3D layer between C2 layers, the only thing is you'll need to use "inside mode" which is slower on weak hardware. But in your case It shouldn't matter, you can have Q3D above everything and use behaviours anyway since they can operate with all the sprites/layer invisible/under Q3D. Instead of making the grid 2D in construct i'd just make it a textured model in Q3D, and fix the camera in orthographic mode.

Yea it breaks, it's really annoying and a terrible design.

from what i understand though, construct saves properties based on their name, so you can change the order without breaking future/past projects using an older version of your plugin, but you'll need to manually go into the run-time file to change all the references. So from a user perspective it's not big deal, but annoying for a developer to do the swapping.

irina

If you want to make a game with 3D graphics and 2D gameplay, i recommend you use invisible sprites to do the gameplay if you want to use 2D built-in construct behaviours, and then pair 3D graphics to those sprites by use of Q3DModels/Q3DSprites and containers.

If you want 3D terrain, you'll need to devise a way to do it, probably best done using a heightmap so you could find the 3D position of an object from its 2D coordinates if its on the ground. There are lots of tutorials on the web on how to make and use heightmaps in a game using standard data structures like arrays. Q3D handles the rendering and boilerplate for 3D work, but you still have to code gameplay on your own, and the internet is full of resources on how to do this which are independent of engine.