So I'm working on a project, where I have a 3d cube-based world. I got all of that to work, and I'm starting the user interface, and the moment I start using spritebatch to draw a cursor texture I have, I discovered that XNA doesn't layer all the models correctly, some of the models that are further away will be drawn first, instead of behind a model. When I take out all the spritebatch code, which is just this:
spriteBatch.Begin();
cursor.draw(spriteBatch);
spriteBatch.End();
I find that the problem is fixed immediately. The cursor is an object, the draw method is just using spriteBatch.draw();
The way I see it, there are two solutions, I could find a way to draw my cursor and other interfaces without using SpriteBatch, or maybe there is a parameter in spriteBatch.Begin() that I could plug in to fix the issue? I'm not sure how to do either of these, anyone else encounter this problem and know how to fix it?
Thanks in advance.
This does not answer the question!
I'm not sure if you could (or should) draw 2D without a spritebatch, however I've had the same problem with 3D model rendering when using 2D spritebatch, and the solution on solution I've found on GameDev helped me solve this:
Your device states are probably wrong. This often happens when mixing
2D and 3D (for example the overload for SpriteBatch.Begin() which
takes no arguments sets some device states that are incompatible with
3D rendering. No worries though, all you have to do is to make sure
that the following device states are set the way you want them:
GraphicsDevice.BlendState = BlendState.Opaque;
GraphicsDevice.DepthStencilState = DepthStencilState.Default;
GraphicsDevice.RasterizerState = RasterizerState.CullCounterClockwise;
GraphicsDevice.SamplerStates[0] = SamplerState.LinearWrap;
Basically, you first call the SpriteBatch methods for 2D draws, then the above code (which should ensure proper 3D rendering), and then draw your 3D models. In fact, I only used the first two lines - BlendState and DepthStencilState and it worked as it should.
Have you had a look at this overload?
You can use the last parameter, layerDepth, to control in what order sprites are drawn. If you use that, make sure to check out the sprite sort mode (in the SpriteBatch.Begin(...) call) as well. Does this cover what you need to do?
Edit: Also note that this implies using the correct perspective matrix, drawing in 2D (I'm assuming you want your cursor to display in 2D on top of everything else), and after all the 3D stuff (it's quite possible to draw sprites at Z=0 for example, making objects in front of that obstruct the sprite).
Related
I am working on a small game where I only want to draw the objects (mesh) when it is inside an invicible box. I have gotten the clipping to work so that the mesh is only rendered inside the box (using the solution mentioned here: https://answers.unity.com/questions/1875660/urp-render-only-whats-inside-a-cube.html
The only annoyance now is that even when the mesh is clipped, the SSAO is still being rendered as you can see in the following image (in the red box):
I assume it is because the object is still contributing to the depth normals - but I am unable to find more information about this - or even if this is the actual issue.
Do any of you have a suggestion for how to prevent this from happening?
I am using Unity 2021.2.8f and URP v12.1.3 btw
The Postprocessing effect SSAO is applied to all layers seen (not culled) by your main camera. Try to put the object on a different layer and ignore it on your main camera.
You could also integrate an additional forward renderer (+ new camera) to your project, which does not use the SSAO effect and takes care of your object.
So for an assignment I have to morph a cube into a sphere. All that's required is to have a bunch of points along the surface of a cube (don't need to be connected or actually be on a cube, but has to form a cube, though connections would make it easier to look at) and have them smoothly translate into a sphere shape.
The main problem is I never learned how to make points in XNA 4.0 and from what I've seen it's very different to what we did in OpenGL (we learned the old one in a previous class).
Would anyone be able to help me figure out making the cube shape I need? Each side would have 10x10 points with the points on the edge shared by the surfaces of that edge. The structure would need to be easy to copy or modify since I would need to have the start state, end state, and the intermediate state to translate the points between the two states.
If I left out anything that could be important let me know.
First of all, you should familiarise yourself with the Primitives3D sample. It illustrates all of the rendering APIs that you need.
Here is how I would approach this problem (you can look up these classes and methods on MSDN and it will hopefully help you flesh out the details):
Create an array of Vector3[] that represents an appropriately tessellated unit cube around (0,0)
Create a second array of Vector3[] and and use Vector3.Normalize to copy in the vertices from your first array. This will create a unit sphere with vertices that match up with the original cube.
Create an array of VertexPositionColor[]. Fill in the colour data however you like.
Use Vector3.Lerp to loop through the first two arrays, interpolating each element to set positions in the third array. This gives you a parameter you can animate - you will have to do this each frame (in Update is probably best).
Create an array of indices (short[]) that describes a triangle list of the tessellated cube (and, in turn, the sphere and animation between the two).
Set up a BasicEffect for rendering. This involves setting its World, View and Projection matrices and maybe turning on VertexColorEnabled. If you want lighting, see the sample for details (you'll need to use a vertex type with normals, and animate those normals correctly).
The way to render with an effect is: foreach(EffectPass effectPass in effect.CurrentTechnique.Passes) { effectPass.Apply(); /* your stuff here */ }
You could create a DynamicVertexBuffer and IndexBuffer and draw with those. But for something simple like this, DrawUserIndexedPrimitives is much easier (here is a recent answer with some details, and here's another one with a complete example of BasicEffect).
Note that you should only create these objects at startup (LoadContent is a good place). During rendering you're simply using them.
If you have trouble with your 3D rendering, and you're drawing text or sprites, see this article for how to fix it.
I'm working on a project in XNA, and I'm using some rather large textures, which I'm loading into the game as Texture2D objects, and drawing on the screen much smaller than they are loaded in. The reason for this is that I need to draw them at various different sizes in different places, and, while I could do this with multiple textures, it is impractical to do so.
My problem is that XNA does not seem to antialias these Texture2D objects upon scaling them down. I have set:
graphics.PreferMultiSampling = true;
and also
spriteBatch.Begin(SpriteBlendMode.AlphaBlend);
But to no avail. I have also tried various things involving the filters in the GraphicsDevice but, to be honest, without really knowing what I was doing.
Obviously, I'm drawing this using a spriteBatch, but aside from that, nothing particularly interesting in the way I'm drawing it.
Here is an (enlarged) example of what's happening, and what I'm trying to make happen:
As you can see, XNA is not providing any anti-aliasing when rescaling the image. Is there a way I could force it to do so, in an effort to make the edges look cleaner?
Are you calling graphics.PreferMultiSampling = true; before GraphicsDevice is created? You must turn it on before it is created, either in the Game constructor or Initialize().
i want to make a 2D Point Light , in XNA, i was able to find some helpful information but is too advance for me , since i dont know nothing about Shaders
http://www.soolstyle.com/2010/02/15/2d-deferred-lightning
so my best aproximation is use a texture and use alpha blending, but im not happy with this result
so i was wondering, what is the most simple method to make a simple 2d point light?, if is there some Code examples, better
You can see a great example of how to do simple 2d lighting using XNA here at Shawn Hargreave's blog:
http://blogs.msdn.com/b/shawnhar/archive/2007/01/02/spritebatch-and-custom-blend-modes.aspx
"With multiplicative blending, I can draw a couple of rotated copies
of this sprite over my scene. Note how the light isn't just drawn as
white, but actually brightens up whatever scenery lies behind it:"
I'm totally new in game-dev and would like to know the best practice about above question.
Let me explain more.
I want to create 2D game with top-down view and with free movement (without snapping to the grid) just like any Zelda game on GameBoy.
How should I store map bounds? Is there a way to do this automatically? For example I have a texture with background and texture with foreground where black color should appear transparent and should allow to move in space of it.
Thanks in advance.
For easy 2D collision detection, you'll probably implement bounding boxes.
Basically you will create a rectangle that represents every Game Object. The coordinates and size of the rectangle will be the same as the Texture2D (it is common to make this a property on the given class). Every time you update the position of your Texture, you update the position of your bounding box.
Now to check for collision, just loop through your game objects and see if any of the bounding boxes intersect.
Once you get the idea, you'll see that its very easy to implement. XNA also provides some math helpers to abstract the math (though its simple addition and subtraction).
Try this link for a more in depth explanation with code examples: http://www.dreamincode.net/forums/topic/180069-xna-2d-bounding-box-collision-detection/