Orthographic projections are generally very simple : the angle of a line in the 2d projection always has the exact same relation to the angle the viewer is looking at. For example, all planes that are exactly perpendicular to the angle the viewer's eyes are pointing, are always completely invisible (dpixel's pipe edit demonstrates this principle); all planes that exactly face the angle the viewer's eyes are pointing, are completely visible (unless other objects obscure them directly) and unskewed. The same principle of a given angle appearing the same no matter where it is in the viewport, applies universally.
But more importantly, hopefully, it gives a clear visual on what faces will always be visible, and what faces will never be visible, relative to any given camera angle.
Let me get this straight ???:Rather, I would say that Orthographic projections are linear projections. That makes 'oblique is linear, but not orthographic' make sense easier.
Linear projection are orthographic projections.
Only, in the Ninja Gaiden screen, we're dealing with an oblique projection which is a linear projection but not an orthographic projection. Am I right so far?
dpixel's edit would be oblique projection:
While my original screen would be... orthographic projection?:
Ok, I think this is the definition I was looking for. I've saved both pictures to my hardrive by the way.Yes.
So this perspective:
(http://i.imgur.com/A0VtTAF.png)
...is equivalent to this one:
(http://image.noelshack.com/fichiers/2015/11/1425920099-sahand-prototype.png)
While this one:No. Sorry, I don't know how to render oblique projection with blender. Those two views I rendered are *both orthographic*. The first view has the object planes aligned perfectly with the camera, the second has them differing by 22.5 degrees in one axis.
(http://i.imgur.com/lOR6Zw9.png)
...is equivalent to this one:
(http://i.imgur.com/A7UKH37.png)
So basically, I have to choose either an oblique or orthographic projection and then stick with it, am I getting this right?.. Up to a certain point. It's pretty common to cheat perspective for gameplay purposes -- eg LOZ:LttP (http://tvtropes.org/pmwiki/pmwiki.php/Main/ThreeQuartersView) (TVTropes warning).
This would be orthographic:The image isn't orthographic (it's a photo taken with a real lense, meaning it is in perspective, probably with a degree of lense distortion involved as well)
(http://i.imgur.com/zq6Hswk.png)
My question is: based on my screen, just how much of the top should I see? What's the eye level here?Okay, I'll stop you there, because : that's up to you. Projections are not viewpoints. there are a range of possible oblique views, there are a much larger range of possible orthographic views (on a given scene).
Let me get this straight ???:
Linear projection are orthographic projections. Only, in the Ninja Gaiden screen, we're dealing with an oblique projection which is a linear projection but not an orthographic projection. Am I right so far?
dpixel's edit would be oblique projection:
(http://i.imgur.com/A7UKH37.png)
While my original screen would be... orthographic projection?:
(http://image.noelshack.com/fichiers/2015/11/1425920099-sahand-prototype.png)
In mathematics, orthogonality is the relation of two lines at right angles to one another (perpendicularity)...http://en.wikipedia.org/wiki/Orthogonality (http://en.wikipedia.org/wiki/Orthogonality)
This would be orthographic:
(http://i.imgur.com/zq6Hswk.png)
Oblique:
(http://i.imgur.com/JVSiWBp.png)
This is something else... everything looks distorted. Your cones turn into pyramids.
So to rephrase that, you need to decide the dimensions of your base unit (assumably that platform block), and the dimensions of your toilet roll. Then you can calculate the ratios between those dimensions, and you will have the information necessary to scale the two visible faces of your block. These scaled faces will exactly indicate the space that the toilet roll occupies, including how much of the top is visible. You can then fit the necessary ellipse onto the scaled top face and extrude a cylinder from it (assuming you want the toilet roll to be axis-aligned with other objects).
I think one of the rules should be: If the screen will scroll (which in your case looks like it will), than there shouldn't be a vanishing point unless rendering in 3D with openGL or something.
The rest would be up to you as how much of the top of stuff will be visible and how much of a side of stuff will be visible. And then keep it consistent.
(http://i.imgur.com/ADc6uaH.png)
The way I understand it, if there was a vanishing point, the cone itself wouldn't look like in your edit.Correct.
If there was a VP, the cone would be standing straight. It's perspective would change depending on the location of the VP.I'm not sure what you mean about standing straight, but yes the shape of the cone would change if there was a VP.
Am I right about this?
I don't know for everyone else, but I'm about as confused as I was at first.It can be confusing especially when you just want to pixel. But it's good to learn some fundamentals if you're going to do a whole game in a certain view.
@Cyangmou:Quote from: mou(http://i.imgur.com/ADc6uaH.png)
To sum things up: both of those are parallel projections. More specifically, they are orthographic projections. Even more specifically, the second one is axonometric.
Quote from: Wikipedia, http://en.wikipedia.org/wiki/Parallel_projectionParallel projection corresponds to a perspective projection with an infinite focal length (the distance from the image plane to the projection point), or "zoom".Quote from: Gary R. Bertoline et al. (2002) Technical Graphics Communication. McGraw-Hill Professional, 2002. ISBN 0-07-365598-8, p.330.Axonometric projection is a type of parallel projection used for creating a pictorial drawing of an object, where the object is rotated along one or more of its axes relative to the plane of projection.
If I wanted to use the 2.5D to build my tileset... how would I pull it off? Since a VP is implied, this would that I would need to create many versions of the same tiles with various angles depending said tile is in relationship with the VP. Seems like a chore to me.
The cavalier perspective seems the best bet for my game. How can I determine at what angle I should draw the slopes like you did?
If I were to use a cylinder using that projection:
(http://i.imgur.com/dKGVxhi.png)
How can I know the shape of the top ellipse of my cylinder?
Should I just put #3 on the floor, make it fit with it and use the same ellipse on the floor I would use on the top? Or should I use #1 or #2 for the top?
If the toilet roll is on its side like this:
(http://i.imgur.com/ODmmxbS.png)
Should I just draw a perfect circle and then have a 45 degrees angle at #1 and #2 to draw the rest?
You take the ruleset of the cabinet projection and construct an ellipse.
Make sure that you don't draw a circle. The ellipse is affected by the projection as well.
look at this - that's how a circle appears at a top-plane in cavalier projection:
(http://upload.wikimedia.org/wikipedia/commons/0/04/Ellipse_dessin_indus.png)
Can I post your spritorial at other places on the internet? The one about perspective? I'd mention it's from you of course.
Is this how things should look, especially "C" and "D"?
So we’ve talked about persepctive so far. I’m currently experimenting but I’m having some trouble still.
Let’s say I use an orthographic perspective and I want to rotate a cube in this perspective, how would I manage this?
So I took some time to really analyze this.
Suppose I want to rotate the rectangle on the “x” axis.
Would it look like this?
(http://image.noelshack.com/fichiers/2015/12/1426465597-cube-rotation.gif)
The second half of the animation looks right to me but not the first part, why is that?
Since there’s no vanishing point, it makes it more difficult to determine how the rectangle would rotate.Since there is no vanishing point, you can take your square and rotate it by constant amounts. Then you can just measure the coverage of the two visible faces. My diagram above illustrates this: when rotated by 45 degrees, the space occupied is 43px and half of this goes to the 'top' face, half to the 'bottom' face.