Printable View
I am trying to make a simple app where you can "paint" active objects into a frame and then save the flattened image of the actives to a new active object. I want to be able to manipulate the result as a new active, so I don't want to simply store a ton of positional data for each painted active. I also want to use this on iOS.
Can anyone tell me of a way to approach this? I feel like the Surface object is the right extension, but I fear it isn't iOS compatible.
Seem lik you need to take a screenshot somehow, but i cant find any objects that let you do this for iOS.
To be clear, I want to capture a specific area of the screen, not the whole screen. The idea of the app is to make a donut construction kit for my 3 year old daughter. She would be able to put any topping as active objects onto the donut and then I would save it out as a new object so she can make it roll around or do something silly in another part of the app. I would want to make the donut and its children into a static image, but ignore the space around it.
So, can I assume this is not possible or is it just that no one has ever tried anything like this? I have seen countless apps on iOS that do similar things to what I am describing. I am just wondering if there are any extensions I can use in F2.5 to do the same.
Storing the positional data is probably the only way, unfortunately.
Your best bet would be to minimize the amount of objects. Like one image for the dough base, another for a layer of chocolate and then smaller actives for sprinkles, candy or whatever. You could use Active Picture Objects to set the hotspot to the donut center in order to make rotating all the donut components easier. As far as I know you can't modify an Active Object's hotspot unless you use Load Frame, which is afaik, only supported on windows.
Is it possible to use RGBat and Replace color in iOS?
If so, there could be another way,
but can be complicated... yet maybe fun to setup.
Idea as follows:
after you do the painting stuff (with paste to backdrop or whatever)
fire a "scan_screen" loop on your drawable area/canvas
and store all RGB values of pixels in array or something
then, use a preset active object, same size as canvas,
one that you already prepared-pixelated with a set of color you surely don't use in your drawing tool (if applicable)
something like this:
Attachment 18947
and store its individual pixel-color-at-position (in an array)
so that you know which color is in any position
Then replace its single pixel colors basing on the "map" you created,
with the new colors stored in your array
this should bring up an active with the drawing you need
Not sure if "transparency" (transparent color) for "empty" zones
would result in consistent results with collisions..
Now, please don't tell me iOS doesn't support RGBat and Replace color... XD
edit___
OK, there's another problem... RgbAt only gets pixel colors inside actives,
to get pixel color in screen you would need an extension unsupported in iOS
...so you wouldn't have to "paste to backdrop"
but instead "paste as actives", and then getting pixel color of foremost object at location...
...starting to get a bit too unrealistic >:)
Thanks guys. I was afraid this might be a bit too much for the iOS runtime at this point. Although, it would be great if Clickteam could add a feature like this to the iOS runtime. Really, it would be ideal if all runtimes had some way to capture a portion of the screen/frame and store it to a clipboard or to a frame of an active object. And I mean a native feature, not an extension. :) Manipulating a frame buffer is a pretty common thing in coding today, so I hope it is at least considered for MMF3 at the very least.
mobichan - Have you tried using Screen Capture Object and loading the results into an Active Picture? I've done an iOS version of the Screen Capture object here: http://community.clickteam.com/threads/88760-iOS-Screen-Capture-extension-v1-2-for-Fusion-2-5?highlight=ios+screen+capture
It would be great if the Active Picture had a way to load from clipboard as this would speed things up - assuming the image doesn't need to be saved to a file for use later.
Even better would be a way to load an image into an Active, so can use fine collision detection.
Good luck!
I saw something similar on youtube where you drew your own spacesh but i have no idea on how to do it, maybe you can get ideas from watching it
https://www.youtube.com/watch?v=LtklXO6Dr8A
Ayre: sorry, as I said I need to capture a specific part of the screen. Not the whole screen. I think the best solution right now is to just use a crap ton of actives. Otherwise I can't get the detail level I want to give users. I'm actually a little curios now if F3 will have a painting canvas similar to F2.5 but with a lot more features and possibilities for mobile.