How do I setup the collision based on a black and white picture or an alpha picture ?

youtube.com/watch

It looks similar to what i am trying to achieve but somehow i can't find how you do that in Construct 3 ?

I have a big sprite that i use as a background picture, i want to use an alpha mask to determine where the character can go or where he collide. Is this possible ?

Thanks a lot

You would need to preprocess it using Drawing Canvas to get the alpha, and set tiles on a Tilemap for collisions.

It's not a process you should plan on doing every tick.

You would need to preprocess it using Drawing Canvas to get the alpha, and set tiles on a Tilemap for collisions.

It's not a process you should plan on doing every tick.

Hi, thanks for the quick answer !

Drawing Canvas is the thing that launch when i double click on my Sprite ?

Do you know a tutorial somewhere that would explain the process ?

I am not using tilemap cuz i wanted to design my level on photoshop then export the level and his collision mask, don't you think it could work that way ?

Thanks and sorry for the super noob question i am new to construct, i just buyed my license : D

construct.net/en/make-games/manuals/construct-3/plugin-reference/drawing-canvas

It's a seperate object.

Construct 3 can't do per pixel comparisons every tick, you will have to make a stand in collision detector.

The only thing that is viable is the Tilemap object. Fyi it can be invisible.

There are a few templates In C3.

construct.net/en/make-games/manuals/construct-3/plugin-reference/drawing-canvas

It's a seperate object.

Construct 3 can't do per pixel comparisons every tick, you will have to make a stand in collision detector.

The only thing that is viable is the Tilemap object. Fyi it can be invisible.

There are a few templates In C3.

So this tilemap way of working is the only way is this right ? I can't do my level design in photoshop then export it as one picture. It's preferable to construct the block do a tilemap and do the level design into C3 ?

I am a senior illustrator but i understand nothing at the moment at Video game ^^

Thanks !

Actually the usual method is to do a scene by placing objects with their collision polygons, and bounding boxes as collision detection. Large composite images are not recommended.

construct.net/en/blogs/construct-official-blog-1/remember-not-waste-memory-796