The water effect has some kind of issue when ran on Windows RT.

First this pops up.

https://onedrive.live.com/redir?resid=ABE7C5AEBE3DDDDD!5010&authkey=!AEYyXyT7Hj_v75o&v=3&ithint=photo%2c.png

And then this pops up after that when ok is clicked:

https://onedrive.live.com/redir?resid=ABE7C5AEBE3DDDDD!5011&authkey=!AOpdraZjAHiVxB0&v=3&ithint=photo%2c.png

If you were a real developer you would have read and understand the JavaScript SDK documentation. C2 has a component model that allows you to build your own 1) Behaviors, 2) Effects, 3) Plugins.

Read up on Intentional Programming... that is essentially what C2 is.

Probably because building equivalent applications on Linux takes 10x longer than using professional tools like Visual Studio on Windows.

Maintaining and supporting software on Linux is also 10x more expensive because you have to deal with many different versions of OS each that might or might not support various tools, as well as various different desktops and tools they may ...or may not support. Dealing with all kinds of possible version mismatch problems between different Java releases... open source projects etc... and you can say well just use Maven! But then after going through all the trouble you later find out Maven doesn't really work and you still end up with version mismatches in your dependencies.

And then to top it all off the market share of the Linux users is < 5% and so the amount of money that would be made from them would not come even close to the cost of building and maintaining a special version just for them.

Just a guess though... from my personal experience with developing software with Linux vs Windows.

Ok I do not think I am making myself clear. I am not displaying a single large image in a tile map. The image I used in the sample capx I linked in this post was just some random image I grabbed to use as an example... take this image for example as a tile map that when I use it to create paths across a 12800 x 10440 layout I get very bad performance.

skydrive.live.com/redir

Since my total video memory usage as reported by C2 is around 20 meg and the view port is 800x480 I would expect that this would work well on a modern smart phone or tablet.

I would think that there is at least a known ballpark number that would let know if the size I am using is far over the line or not.

The image was just a random picture I threw into the capx instead of putting in my real tile map images.

Is there a size of tile map that is known to be too big? As in too many tiles?

Is 12800 x 10240 just way too big?

In my real tile map the image is 640 x 1280.

Link to .capx file (required! If link is blocked remove the http and www parts):

skydrive.live.com/redir

Steps to reproduce:

1. Open Capx

2. Run in debug mode

3. Look at draw rect count for tile maps

Observed result:

One of the tile maps has nearly 80,000 draw rect even though most tiles are off screen and the CPU usage is very high even though there is very little going on. (Try clicking around to move the sprite)

Expected result:

Much lower CPU usage.

Browsers affected:

Chrome: yes

Firefox: don't know

Internet Explorer: yes (much worse)

Operating system & service pack:

Windows 7 SP1

Construct 2 version:

158.2

If that is true why would there be > 15,000 draw rects in a 800x480 space?

Just to be clear most of the tile map is empty. It contains tiles that are a road or path. This sits on top of another tile map that contains the ground tiles.

I have a tile map that is 12800, 10240 in size with 32x32 tile size.

This tile map shows over 15,000 draw rects in the debugger even though the view port is only 800 x 480.

Are all tiles drawn even though they are way off screen?

In the Tilemap editor there are little integers identifying each tile... in the Tilemap object in the Layout editor I need to see these same integers so I know where in the map editor a tile came from so I can line it up with the correct tile.

How do I get the profiler to stop sorting or whatever is causing it to jump around so I cannot really read it?

So instead of doing a scan potentially across 1000 objects with the "Pick instance with UID" condition looking for a matching UID... you would only scan for a matching UID on the objects in that Cell.

Not to just keep piling on (however if it would require about the same amount of work on your side...)

It might be logical to want to know what objects are in your current cell plus all the cells adjacent ... so you can spin up their AI (or whatever) as the player is approaching.