I need a little help understanding how to draw ellipses in perspective

[fauxfresh]

How do you know which direction the minor axis on an ellipse is supposed to be going?

I read that it should always align to the vanishing point like in this image but in this image it does not seem to be the case for any of the ellipses.


In figure 2.3 on this page the minor axis on the ellipses go towards the right vanishing point but not the left one.


Can someone explain this to me? I feel like its something really simple and I'm just not seeing it.

[John]

Hey there fauxfresh,

Keep in mind that we draw ellipses to illustrate a circle in perspective. Meaning it's perfectly round if we look at it straight on, and it's elliptical if seen on the side. (Which in the first image you provided illustrates). If the circle is indeed perfectly round, then we can also say that the circle can fit perfectly in a square..

The second image you provided does not illustrate a perfect circle. Couple of clues are: 1) both the minor axis isn't vanishing towards the horizon line. 2) Shape of the box is not a perfect square. The ellipse in question, if we look at it directly, should be oblong and not a circle.

In the third image, the minor axis is tilted towards the right vanishing point because the object is foreshortened towards the right, meaning, if we stretch the cylinder into the infinite, it's going to 'vanish' into the right vanishing point.

Hope this helps!

[Abnormal]

Same as above but in a nutshell XD.

" The minor axis divides the ellipse in half across its narrowest dimension.

It always points to the Vanishing Point that is perpendicular to the ellipse's surface being sketched.
This makes the minor axis like the axle of a wheel. " Scott R.

[fauxfresh]

Hey there fauxfresh,

Keep in mind that we draw ellipses to illustrate a circle in perspective. Meaning it's perfectly round if we look at it straight on, and it's elliptical if seen on the side. (Which in the first image you provided illustrates). If the circle is indeed perfectly round, then we can also say that the circle can fit perfectly in a square..

The second image you provided does not illustrate a perfect circle. Couple of clues are: 1) both the minor axis isn't vanishing towards the horizon line. 2) Shape of the box is not a perfect square. The ellipse in question, if we look at it directly, should be oblong and not a circle.

In the third image, the minor axis is tilted towards the right vanishing point because the object is foreshortened towards the right, meaning, if we stretch the cylinder into the infinite, it's going to 'vanish' into the right vanishing point.

Hope this helps!

Thank you for taking the time to answer my question!

If its alright with you could you explain a bit more about the second image? I am still a little confused. I do not understand what you mean when you say the "minor axis isn't vanishing towards the horizon line"

If I trace the minor axis it does eventually cross the horizon line, is the issue that it is not heading towards the vanishing point? I also do not know how to tell if the shape of the box is a perfect square or not. Is this something I should be able to figure out intuitively just by looking at it or are there perspective rules I can use to confirm if it is a perfect square?

[fauxfresh]

Same as above but in a nutshell XD.

" The minor axis divides the ellipse in half across its narrowest dimension.  

It always points to the Vanishing Point that is perpendicular to the ellipse's surface being sketched.
This makes the minor axis like the axle of a wheel. "  Scott R.

Thank you, that explanation really helps me understand it

Does that only account for images that are not one point perspective or does it still apply? For example the second image

[John]

Hmm..

EDIT: Forgot to mention, the minor axises (axi? axes?) for each ellipse should go though the center of both ellipses!  The 2nd image didn't illustrate that, ergo, not a cylinder! (Or at least not a cylinder with a circular face)
EDIT 2: As far as I know, in a two point perspective, any four sided shape with at least two lines parallel to each other can be considered as a perfect square. Of course, you can go to the extreme of making one parallel line a 100 feet and 10 cm with the other. I can still say it can be considered a perfect square in perspective, but a horribly distorted perfect square in perspective.
EDIT 3: Scratch my second edit! If you can fit an ellipse in a trapezoid, then it can be considered as a square in perspective.

[fauxfresh]

Hmm..
EDIT: Forgot to mention, the minor axises (axi? axes?) for each ellipse should go though the center of both ellipses!  The 2nd image didn't illustrate that, ergo, not a cylinder! (Or at least not a cylinder with a circular face)
EDIT 2: As far as I know, in a two point perspective, any four sided shape with at least two lines parallel to each other can be considered as a perfect square. Of course, you can go to the extreme of making one parallel line a 100 feet and 10 cm with the other. I can still say it can be considered a perfect square in perspective, but a horribly distorted perfect square in perspective.

Thank you so much for the explanation! 

I think I understand the concepts now. This is the video where the second image comes from and created so much confusion for me. https://www.youtube.com/watch?v=2GpQMxcHOEU


At 7:09 she connects the two ellipses into a cylinder but now I know that is just a mistake on her part because it is one point perspective and she should have placed another vanishing point.

[John]

Woot!