I found culling only under spatial hashing for collisions. I'm referring to the kind of backface culling performed by 3D graphics libraries, where anything that need not be visible isn't rendered.
Does Chipmunk2D/Pymunk have any provision for not drawing objects that are not within screen bounds or does that user have to implement it themselves?
For example:
The red rectangle is the screen boundary. All blue objects should get drawn because they are within the screen. Green objects shouldn't be drawn.
I was hoping debug_draw() would have a culling functionality.
ps: btw, if I don't use debug_draw() for drawing, what is the other way of drawing? I don't see a draw() or release_draw() function. So would the user have to write code to individually iterate all objects and draw them? I guess that'd work fine because then the user can do a rectangle intersection test and decide which objects to cull. Perhaps debug_draw could be renamed to drawAll().
The debug draw method is mainly meant for debugging and quick prototyping, so more advanced features such as culling is out of scope for its implementation.
If you feel yourself limited by debug draw it might be time to transition to your own drawing code, where you have full control. It should be quite easy to emulate what debug draw is doing yourself, some of the example code do custom drawing.
Related
Is it a reasonable optimization to omit calls to ID2D1HwndRenderTarget::DrawBitmap() if the image will end up outside the visible area? If I implement the checking logic in the application that will cost some performance, so if the first thing D2D does is doing the same check then I'd rather not do it.
I had a test with my application which renders some UI part using Direct2D (and attaching renderdoc), seems it is a bit random.
I render a mix of Rectangles, Text, Path geometries (beziers) and Rectangle with a bitmap brush (which should be equivalent to your DrawBitmap call).
Then I capture a frame with all those objects visible, and another one panning my UI (using transform) so objects are not visible.
From there could check what is drawn or not:
Text is always culled
Solid color rectangles are not culled
Most of the times path geometries are culled, but sometimes not.
Rectangle with bitmap brush are NEVER culled
So it seems Direct2D is making different decisions depending on the types of elements you plan to draw.
Since rectangles are easily batched and cheap to draw, it seems that they are just drawn regardless.
Bitmap rectangles and text require more work, so it seems they are effectively culled.
Path geometry was looking to depend on how many polygon the geometry is tesselated to (I had a path that was translating to 26 primitives and it was not culled, another one translating to 120 and it got culled).
So you can either trust Direct2D that it will perform that optimization, but I would personally implement a quick rectangle to rectangle check just in case (it's not gonna hurt your performances as its an extremely simple operation).
I want to swap textures on the faces of a rotating cube whenever they face away from the camera; detecting these faces is not entirely equivalent to, but very similar to hidden surface removal. Is it possible to hook into the builtin backface culling function/depth buffer to determine this, particularly if I want to extend this to more complex polygons?
There is a simple solution using dot products described here but I am wondering if it's possible to hook into existing functions.
I don't think you can hook into internal WebGL processing but even if it's possible it wouldn't be the best way for you. Doing this would need your GPU to switch current texture on triangle-by-triangle basis, messing up with internal caches, etc. and in general - our GPUs don't like if commands.
You can however render your mesh with first texture and face culling enabled, then set your cull face direction to opposite and render your mesh with 2nd texture. This way you'll get different texture on front- and back-facing triangles.
As I think of it - if you have a correct mesh that uses face culling you shouldn't get any difference, because you'll never see a back-face, so the only ways it could be useful is for transparent meshes or not-so-correctly closed ones, like billboards. If you want to use this approach with transparency then you'll need to carefully pick the correct rendering order.
I am using Unity3D, and I have a function which is being called inside of OnGUI to lay out the various gui components of my application. Ordinarily, the labels and buttons are all inside of a certain Rect that I supply, which is centered on the screen.
No problem there... however, what I want to is sometime render the exact same gui elements, which can be dynamic, and thus not just put into a prefabbed texture, into a trapezoid-shaped area off to the side, looking as if that gui were actually on a flat plane, pushed away from the center of the screen, and rotated slightly. All gui buttons that were drawn in the function should still respond normally.
I was rather hoping I could just specify some values in GUI.matrix to map the rectangle to a trapezoid, but my initial exploration seems to show that the gui elements don't appear to use homogenous coordinates, and everything still shows up as rectangular.
Is there any way to do this with Unity, ideally without requiring access to pro-only features?
Since now Unity3D GUI system isn't very flexible. The new GUI system is one of the features still not released in Unity 4 (we are all waiting for it).
From my point of view it has several problems, particularly:
You are forced to layout components using the flow of the code, instead of having a more declarative (or at least a more structured) way to do that.
It's quite inefficient (at least one draw call for button).
It isn't flexible at all. Add, Remove, Enable/Disable buttons can be come quick a painful operation when the number of buttons increase.
however, what I want to is sometime render the exact same gui
elements, which can be dynamic, and thus not just put into a prefabbed
texture, into a trapezoid-shaped area off to the side, looking as if
that gui were actually on a flat plane, pushed away from the center of
the screen, and rotated slightly. All gui buttons that were drawn in
the function should still respond normally.
This is quite hard if not impossible to obtain using Unity's GUI classes.
I see 2 possibilities:
Don't use GUI classes to do that. If your GUI is simple enough, you can implement your own (even 3d) buttons using for example:
A mesh (a plane or a trapezoid mesh) with a texture for the button background
TextMesh for drawing 3D text
RayCasting to check if a button has been pressed
Use a library that implements a more advanced GUI system like NGUI
When I ran into the same problem, I just used normal 3D GameObjects cubes with textures and called OnMouseDown(PC/Mac) or RayCasting(Android/iOS) on them. I guess that's how everyone does it.
I am making a GUI in OpenGL (more specifically lwjgl). I have tried hard to research different ways of doing this but I am having a hard time finding exactly what I want. I do not want to use any external libraries (only ones built in OpenGL, even trying to stay away from using GLUT) and I would like to have it work on anything that supports OpenGL (ex. Frame Buffer Objects don't work on older graphic cards).
I am making a 3D GUI with a scrollable panel as a component. The problem is I don't know how to draw a partial GUI component without doing a lot of calculations to only render part of it. I am making the components out of OpenGL primitives, not textures. I was hoping there is an easy way to do this like use multiple viewports. I don't really even understand what viewports are.
In short: I need to have a scrollable panel as a component overlapping other GUI components (since it will be a drop down menu) and not let any of the components in my panel draw outside my panel.
If you just want to prevent drawing pixels that are outside of a rectangular region (and I think that's what you're asking), than glScissor is exactly what you're looking for.
In lwjgl, you can find the function in org.lwjgl.opengl.GL11.
If you want to scroll a larger scene within a fixed region on the screen, the most straightforward way to go is by just modifying your projection matrix for the scroll position and redrawing the scene. If you are using gluPerspective to set up your projection matrix you'll have to convert it to a direct call to glFrustum; if you're using glOrtho it's much more straightforward.
Keep in mind that "scrolling" a perspective view has no one right way to do things - it depends on what sort of effect you want to achieve, and what particular sort of distortion you want near the edges of the overall viewport.
Every technique that I've found or tried to render outline in OpenGL uses some function that is not avaliable on OpenGL ES...
Actually what I could do is set depthMask to false, draw the object as a 3 pixels wide line wireframe, reenable the depthMask and then drawing my object. It doesnt work for me because it outline only the external parts of my object, not the internals.
The following image shows two outlines, the left one is a correct outline, the right one is what I got.
So, can someone direct me to a technique that doesn't is avaliable on OpenGL ES?
Haven't done one of these for a while, but I think you're almost there! What I would recommend is this:
Keep depthMask enabled, but flip your backface culling to only render the "inside" of the object.
Draw the mesh with that shader that pushes all the verts out along their normals slightly and as a solid color (your outline color, probably black). Make sure that you're drawing solid triangles and not just GL_LINES.
Flip the backface culling back to normal again and re-render the mesh like usual.
The result is that the outlines will only be visible around the points on your mesh where the triangles start to turn away from the camera. This gives you some nice, simple outlines around things like noses, chins, lips, and other internal details.