3D weirdness

mathsyperson · 2008-04-27 13:05:47

Here's something I just realised when idly staring at my ceiling trying to comprehend group theory.

Stand roughly in the middle of the room (which should be cuboid-shaped. Switch rooms if it isn't.), and look at a corner above you.

It should have 3 angles meeting up with each other - two on walls and one on the ceiling. The angle on the ceiling should appear to be a bit more than 90 degrees.

However, this is true for all four of the ceiling-corners, and thus your ceiling is a quadrilateral whose angles sum >360 degrees. Crazy ceilings.

MathsIsFun · 2008-04-27 13:16:20

Ahhh... the missing element is the turning of the head. You should see all corners at once I think

mikau · 2008-04-27 16:01:20

Yeah, if you could get all 4 angles in view at once, you should have a trapezoid that does not defy the rule.

You do present an interesting point though.. how do we know what the angles do when we're not looking? They might be stealing candy!

MathsIsFun · 2008-04-27 16:07:53

mikau wrote:

...They might be stealing candy!

Of course! That explains the mystery!

Dragonshade · 2008-04-27 16:50:05

its true if you consider it to be a relativistic space. XD.

integer · 2008-04-27 22:39:01

spherical trigonometry!

You are viewing/measuring the angles in a plane normal to your line of sight.

The sum of the angles of a triangle on a sphere is greater than 180 degrees.
Thus, the sum of the angles of a quadrilateral on a sphere will be greater than 360 degrees.

The brilliance of your observation:
You can "prove" to some that the sum of the angle of a square is greater than 360 degrees.
Of course, those are then same persons who believe in the "proofs" that 1 = 2 and 1 = 0.

mathsyperson · 2008-04-27 22:49:23

integer wrote:

The sum of the angles of a triangle on a sphere is greater than 180 degrees.

I knew that bit, from the puzzle about the man who walks S, E, N and ends up in the same place, before seeing a polar bear. (The angles of his turns would be 90, 90 and >0)

What confuses me is that the ceiling is a plane, and so surely spheres shouldn't enter into it?

Dragonshade · 2008-04-28 03:51:44

ceiling is not a plane, its a fragment on a sphere, since it's parellel to the ground.

mathsyperson · 2008-04-28 06:38:40

Well yes technically, but the sphere is so huge (and so has curvature so tiny) that the ceiling could reasonably be approximated as a plane.

I wouldn't think that you'd get such a massive discrepancy from the expected 360 degree sum just because the ceiling might be very very very slightly curved.

John E. Franklin · 2008-04-28 15:27:45

You think that's weird?
Try taking a nice pair of binoculars
into the car with you when you are
not the driver. Take some hilly and
curvy roads. Your observations
will be unforgettable. I won't even
try to explain it because it blows
the mind.

Dragonshade · 2008-04-28 18:00:06

mathsyperson wrote:

Well yes technically, but the sphere is so huge (and so has curvature so tiny) that the ceiling could reasonably be approximated as a plane.
I wouldn't think that you'd get such a massive discrepancy from the expected 360 degree sum just because the ceiling might be very very very slightly curved.

Hmm,you got a point... isn't that some kind of visual fallacy, then?

John E. Franklin · 2008-04-29 03:41:51

Okay, well the binoculars thing reminds me of Lorenz transforms
when things get shorter at the speed of light.

But the angle in the ceiling is to be expected because an angle in a plane will
be viewed at its smallest at the true angle, but from other angles it can be
as large at 179.9999 if you are almost in the plane, like a fly on the ceiling.

John E. Franklin · 2008-04-29 03:45:36

Woops, correction, there are places where the angle actually gets smaller than the true angle too I just noticed while walking around looking at my thumb and index finger turning it all around.

John E. Franklin · 2008-04-29 03:47:20

If you are "inside" the triangular perpendicular prism formed by the angle, then
the angle can get larger, but if you outside this triangular space, so "outside the angle", then the angle can get smaller.

John E. Franklin · 2008-04-29 04:09:18

Woops, there are really 4 triangular prisms;
make an X from the original V angle by extending
the lines to cross. Now I think now if you
are inside the 2 opposite prisms with the same
size angle, where one is the original one, then
the viewed angle can get larger. But if you
in 1 of the other 2 opposing angles, call them
the "left" and "right" extensions from the original
prism, then the angle can get smaller.
This is just my observation, no mathematical
conclusions yet, sorry. And it may be wrong.

Last edited by John E. Franklin (2008-04-29 04:09:48)

Math Is Fun Forum

#1 2008-04-27 13:05:47

3D weirdness

#2 2008-04-27 13:16:20

Re: 3D weirdness

#3 2008-04-27 16:01:20

Re: 3D weirdness

#4 2008-04-27 16:07:53

Re: 3D weirdness

#5 2008-04-27 16:50:05

Re: 3D weirdness

#6 2008-04-27 22:39:01

Re: 3D weirdness

#7 2008-04-27 22:49:23

Re: 3D weirdness

#8 2008-04-28 03:51:44

Re: 3D weirdness

#9 2008-04-28 06:38:40

Re: 3D weirdness

#10 2008-04-28 15:27:45

Re: 3D weirdness

#11 2008-04-28 18:00:06

Re: 3D weirdness

#12 2008-04-29 03:41:51

Re: 3D weirdness

#13 2008-04-29 03:45:36

Re: 3D weirdness

#14 2008-04-29 03:47:20

Re: 3D weirdness

#15 2008-04-29 04:09:18

Re: 3D weirdness

Board footer