In memory any image is fully decompressed, so each image will take pixelwidth x pixelheight x 4 bytes of image memory..

pixelwidth and pixelheight are the amount of pixels in the image editor inside the construct editor, for each frame of each animation offcourse if there are any..

Very large issue (for me at least):

While manipulating bones through the Q3Dbone-object (picked by name) , somehow the wrong bone is selected..

I'm not sure if this had to do with too many bones in my object or wrong hierachy or something else, but a bone is manipulated, which is nowhere near the bone I was aiming for and which has a completely different name..

I'm sure I'm pushing the boundaries of this plugin, but I'm very close to achieving what I want if I could only get this single bone manipulating to work as expected..

Edit: Maybe there are other limitations, like name-length, etcetera?

but i have no idea how to specify one instance

m not getting this point

If you want all instances to react at the same time you don't have to specify one instance..

If you want one instance to react, you should tell the computer which one..

If you know which one you mean, you should be able to tell the computer..

For example, select an instance by touching it, pick the nearest to something, etcetera..

If you don't specify a specific instance in the event the actions will be executed by all instances..

If you only want to move specific instances you should make it clear which ones, for example by giving them an instance variable or putting them on a specific layer or any other way to differentiate them from other instances and add the condition needed.

So what you have is a picking problem?

I'v never worked with a "rotate sprite" or anything like that, not even sure what it is..

Is there any way to tell the rotate sprites apart?

for example by adding a variable to both the sprite and the rotate sprite and checking that..

I'm guessing the multi-object enemies have some way for the multi-objects to know they belong together, couldn't that be used?

Without a better explanation on what you want to do exactly and/or an example either in capx or maybe visual, this is my best answer..

You are talking about grid based movents using the move to behaviour, so I guess that part is working..

Jumping on platforms would consist of first knowing when to do that (creating the conditions) and secondly attaching the right actions to do it..

I'm sure it can be done if the wanted effect/actions are clearer..

You could solve this by using else statements attached to your conditions..

But maybe using the button to add to a variable and setting the text based on the variable would be even easier..

I don't mean to say you should animate the jump, although you could, but for example:

The dummy sprite is used for the grid movement.. you could use this as a kind of shadow also, because it's always on the floor..

The animation-sprite (which could consist of a single frame) is moved in the y-direction based on the percentage of the distance travelled between start and endpoint of the moveto action..

Would it work to use a "dummy" sprite for the movement and an animation-sprite for the jumping?

What would be the issue moving the layer-objects in front of the player?

You are not using move to top or bottom so the z-order of the layer-objects should stay the same..

The issue might be (in the for each at least) that you are using

for overlapping : playerrange-object

for comparing the y : playerpin-object

for moving the player : player family

If the playerrange object is part of the player family, I would suggest using the player family in the overlapping event so at least the object referenced in the condition is the same as the one in the action..

How different are this thing and that thing?

Is the thing different on every layout?