Confused about perspective
TaniaGomesArt last edited by TaniaGomesArt
One thing you need to have into consideration is that one vanishing point (or set of vanishing points) works for things that have the same alignment. The lines that go to the vanishing point are lines that are parallel in real life. But if an object is not in the same alignment, it will have a different vanishing point on the same horizon line.
Those 2 streets don't seem to be parallel to one another, it seems that they go in different directions. So they will not have the same 2 vanishing points. I'm sure this was mentioned in one of the svs courses I did, I think it was Basic Perspective by David Hohn. And those building in the middle of the 2 streets don't seem to be square in their base, which complicates things a bit more because not even their sides are parallel.
Ultimately, yes, you may have several vanishing points in the same perspective. It's still a 2 point perspective, but for each objects alignment, you will have 2 specific vanishing points in the horizon line.
Jeremy Ross last edited by
Hi @Urvashi-S, one thing that helps me a lot with perspective is to draw the vanishing points and horizon line directly on photos. It’s pretty amazing to see everything line up, kinda makes you feel like you’re in the matrix.
Whew, that's a tricky one! It is definitely 2 separate one point drawings within a 2 point perspective system. All objects within a scene can have their own vanishing point, and those points will move along the horizon line. know that in photos the lens that is used can wreck havoc on trying to line up vanishing points in a easy to understand way. Wide angle lens vs telephoto will yield VERY different results for a scene.
This image is further complicated with a triangular building right in the middle so there is that to deal with as well. Let's ask the master: @davidhohn what do you think?
This is a bit tricky -- but honestly you all have given great info on how to approach a perspective situation like this.
First, does this photo primarily use a 1 point, 2 point, or 3 point system?
Keep in mind we always experience the world using a 3 point system.
The application of 1 or 2 point is an artificial construct applied to a scene to simplify it. So, while a photograph is always fundamentally capturing a 3 point system sometimes it can be just as effective to evaluate it using a 1 or 2 point.
Okay so let's begin:
First does this photo justify a 3 point system?!
I drew some vertical lines (orthogonals) on the photo. While the vertical (and also parallel) sides of the buildings do seem to taper slightly toward the top of the image, in my opinion they don't do it dramatically enough to justify a 3 point system.
Cool. A bit simpler!
So, is this a typical 2 point system?
I'm going to ignore the FlatIron building at this stage (it's a big triangle) and focus on the buildings to the left, right and behind the Flat Iron. These are more rectangular buildings with 90 deg corners.
The Horizon line (HL) is indicated by the purple line toward the bottom.
Note how all the sides the buildings with windows facing the viewer (generally) follow the brown horizontal lines. If those brown lines converged toward the HL on either the left side of the image (the left vanishing point) or the right side of the image (the right vanishing point) this would indicate a typical 2 point system. These lines don't converge. They remain parallel to the HL. This image does not employ a 2 point system.
So then what's the deal with those two streets (5th Ave and Broadway)?
I've found it's important to understand a city photo from above as well as street level, when trying to fit it into a 1, 2 or 3 point system. Check out this shot from Google Earth:
A nice, simple two point persecutive system applies neatly to a city scene when the buildings are on a tidy rectangular grid. When all the streets and buildings are perfectly 90 degrees to each other. @Jeremy-Ross shows this perfectly in the example he posted above.
An important concept to remember is that in linear perspective all groups of parallel lines (orthogonals) converge on their own vanishing point on the horizon line (HL). So each of these sets of parallel lines (blue, green and purple) will converge on their own distinct blue, green and purple vanishing points (VP).
Note what the location of the Flat Iron Building does. 5th Ave and Broadway angle away from each other, each street forming it's own 1 point perspective system. The blue line of Broadway, as compared to the purple lines of 5th Ave.
Combine that with the fact that the Broadway face of the Flat Iron Building itself doesn't line up parallel the street. The green lines show it's at a slightly different angle. (Who built this thing‽)
By looking at the ariel view in Google Earth above we can start to better understand the street level photograph below.
The 5th ave face of the Flat Iron is parallel to the street and to the buildings on the opposite side of 5th ave. All the purple horizontal lines of those buildings converge neatly on a vanishing point (VP) on the HL.
The buildings on the left side of Broadway are also parallel to each other and the street. The blue horizontal lines of those buildings converge on their own VP on the HL.
The left face (Broadway) face of the Flat Iron is parallel to itself, but not to the Broadway street or the buildings on the opposite side of the street. So all the green horizontal lines of that face of the Flat Iron will converge on their own VP on the HL.
The result of this careful analysis, is a photo that creates a perspective grid that looks like this
To the uninitiated artist, a confusing tangle of lines -- but if you step through each set of orthogonals carefully you can see the underlying order in the image.
know that in photos the lens that is used can wreck havoc on trying to line up vanishing points in a easy to understand way. Wide angle lens vs telephoto will yield VERY different results for a scene.
This is too true! For this reason, when students are starting out I strongly recommend that you work from photos you've taken. From locations you have physically been in. That way you'll be better prepared to understand the potential distortions a camera lens can create.
Jeremy Ross last edited by
Fantastic @davidhohn! This should be a blog post!
@davidhohn so informative
TaniaGomesArt last edited by
@davidhohn Great explanation. This was definitely a tricky one and it was great to see this break down to understand what kind of perspective it was (1 or 2 points).
Frost Drive last edited by
Forget 1 point, 2 point, 3 point perspective mindset.
Even if all you drew is a minecraft character walking in a white void. That minecraft character has WAY more than just 3 vanishing points.
I actually have a video I made before on perspective. And I teach it in a simpler different way than the 'traditional' way.
Urvashi S last edited by
@TaniaGomesArt Thank you so much for that simple explanation! I was very confused as to why all the lines didn't seem to converge on the same points.
Urvashi S last edited by
@davidhohn Thank you so much for that detailed explanation! My head feels clearer. Does that mean, in order to create a realistic cityscape or buildings( I'm studying to be a concept artist), I could use a perspective grid like this, with multiple vanishing points on the horizon line?
@Urvashi-S Glad it's helpful. Most "realistic" cityscapes will employ multiple different sets of Left and Right VPs. This is because typically the world doesn't exist in a simple, tidy grid.
You can always tell the cityscape drawing of someone new to perspective because everything is just a little too perfectly parallel with each other, or at 90 degrees to each other. (Hey, we've all been there!)
So, yes you could certainly use something like this.