Lens length for rendering

A have some questions about the camera length:

  1. What is it and what does it do?
  2. In the attached image below, notice that the drawing seems to open from the center to the corners. The modeler (in Blender) told me it was achieved by setting the camera length. How can I do the same in Rhino?

You mean ‘Lens Length’ – not ‘camera length.’ The longer the lens, the more you lose perspective (like a telephoto lens) and ‘flatten’ out the view. The shorter the lens, the more you increase your field of view (like a wide angle lens or fish-eye lens)

The lens length is accessed via the viewport label. Start in Perspective view, click on the viewport label, and select ‘Viewport Properties.’ The dialog that pops up has a field called ‘35 camera lens, Lens length = ?? mm.’ Try 30 mm for a nice / slightly wide angle view.

When an image “opens up,” it means that objects farther away get smaller and you are in perspective view, just like the real world. The opposite of that is an orthogonal view, like front or side, where there is no perspective distortion at all.

Do a google search for ‘perspective’ for more info

1 Like

Thanks man for the explanation!
Another question. What is the natural lens length for humans?

According to my sources at google, 45-50 mm is the closest approximation.

But this is not exact since human vision has no hard cutoff, the peripheries just get blurrier and the extreme edges are only sensitive to movement.

1 Like

Keep in mind that when Rhino (and Google) specifies the lens focal length as 50mm, what is really going on is that they are saying that a 50mm lens specified in Rhino will show the same field of view as a camera with 35mm photographic film would show if it used a 50mm lens, which is approximately the same field of view that a human eye would see.

In this day and age of digital cameras with various sizes of image sensors, each sensor size will have a different lens focal length appropriate for representing a human field of view. So you can’t just look at your point-and-shoot’s lens focal length label and expect it to be 50mm.

So, in a sense, the Rhino lens focal length specification is tied to an older photographic standard which is slowly fading. Right now there are still lots of people who have an intuitive feel for which focal length is appropriate for the view they desire based on 35mm film. Digital SLR cameras have image chips that are more or less the same size as a 35mm film frame so they will carry the 35mm frame of reference forward, but as digital photography evolves, who knows??? There may come a day when the Rhino standard is not traceable back to anything anyone is familiar with - except Rhino. :smile:

Last I have read 43mm and I posted a link at an other thread some days before.

Good point. @pascal Perhaps the “lens length” entry field can be made interchangeable with a “FOV” (field of view) field that can be specified by angle…


1 Like

Currently, the equivalent focal length with 35mm format is the de-facto standard for comparing focal lengths of different cameras with different sensor sizes. Point and shoot cameras are frequently advertised and even marked with the equivalent 35mm format focal length rather than the actual focal length. Technically, field of view angle would be better because it is independent of format/sensor size but it very rarely used.

1 Like

One thing that most forget when trying to emulate a human view and calculate an approximate focal length. Is that we have two eyes that visualize together for the most part and that would produce a wider view (than uniformly equal). More like the slightly offset ViewMaster imagery (stereoscopic) than the cyclops view of one lens.

fwiw, i’m not quite sure the “50mm on 35mmFormat most closely resembles what the human eye sees” makes much sense -or- shouldn’t really be considered when translated to a 2D image…

the fov of the 50, at least according to the theory, will capture the same cone which a human can have in tangible focus at any given time (our true view fields are much wider than a 50 however… i don’t think it’s 180º but it’s not too far off that…

anyway, when we look at a 2D image, we don’t always or exactly fill our focusable_cone(?) with that image alone… we generally look around at different parts of the image meaning we’re not looking at a picture made with a 50 using our similar 50mm fov… we focus in on individual parts of the image which effectively means we’re zooming in from the original lens (but we do that without the ability to change perspective-- which is one reason why a 2d image will always be ‘fake’ when compared to our natural vision… the perspective is locked in… this can certainly be used to positive effect but trying to capture ‘reality’ with a camera isn’t going to happen- so don’t try :wink: )

when making a render, it’s more important to consider perspective (where you would like the lens to be)… moving in close to an object makes is appear larger than the objects behind it -or- moving further away will make the two object appear more equal in size with less separation between the two (flatter)…

once the perspective is set, then the fov is considered in order to determine the crop or how much up,down,left,right info you’d like to include…

you can’t really force a FOV on a viewer unless you require them to view the image at a certain distance with their line of sight pointed at the middle of the image… without that happening, the viewer is going to look anywhere in the image they choose rendering your attempt at “this is a natural looking picture because i used a 50” - pretty much useless…

so really, when making a 2d image, the first question is 1) “what do i want to include in the picture” – followed by 2) "what perspective (where the camera is placed and/or what size relationships do i want) – followed by 3) “what fov should i use” -(which isn’t really a choice anymore, you just use fov to crop at this point…

in the case of joddys’ picture in the top post:

  1. the entirety of the floor plan
  2. camera is up top- the further you move the camera away, the straighter those side walls or flare will be… get closer for more flare…
  3. set fov to define the boundaries of the image.
1 Like

The generally accepted lens lenght closest to the human vision is indeed 43mm, in terms of magnification. If we consider the field of view it’s more like a 28mm lens.

nah… these are uncropped 28mm lens on a 35mm rangefinder… we can see wider than this…

likewise- these are an 80mm lens on 6x7 …this is what is considered a ‘normal’ lens… but again, that doesn’t translate to anything of much meaning except in all but the most technical of contexts… when you look at these, you’re focusing in different details and the rest of the image is in your peripherals… or, you’re not looking at these in the same way the camera did meaning there’s nothing ‘normal’ about lens choice… perspective is key-- focal length lets you work at certain distances in order to get the perspective you’re after and it’s completely ‘normal’ for you to either stand close to an object or further away… it shouldn’t be the other way around…

maybe another way to say what i’m trying to say…

focal length (or field of view) is nothing but a crop…

do this–
put camera on tripod.
shoot pic with 85mm lens
change lens to 35mm and shoot.

crop the 35mm version to that of the 85mm version and you’re left with the same exact picture

the different focal lengths did nothing to change anything about the image except for how it’s cropped (in other words, you can change a focal length after the picture has already been made simply by cropping in photoshop)

for comparison, do this:

  • shoot picture with 85mm lens
  • switch to 35mm lens and walk closer to the subject until it fills the same area of the frame

these are two completely different pictures now and it’s not because of focal length, it’s because of where you’re standing.

easier yet, forget about the camera and do the comparison in rhino… you’ll see what i’m saying.

Nah… that’s not true.
We can of course see wider than 28mm but the human vision is not an on and off thing. A good illustration can be found here: http://humaneyeproject.wordpress.com/2012/08/19/the-human-eye-vs-the-camera/
I’ve done numerous tests in real life; I often do renderings of plane interiors, even plane lavatories. I took real pictures, I put several people at the same place and have them descibing what they see when moving/without moving their eyes and their head.
It’s like designing a desk and looking at the reach data from Dreyfus and then comparing with what you get yourself.

hey @Marc

that blog post is giving incorrect information… that’s all there is to it…
funny though that there is an image to accompany the text which is right but the text doesn’t correspond to what the image is showing…

[quote=the blog post]Camera lenses can be categorised into three groups: Wide, Normal and Telephoto lenses. With the use of a narrower angle of view in the camera, objects seen are nearly of the same relative size, but the sense of depth is lost. However, while using a wider angle of view, the relative sizes of the objects seen are exaggerated and the objects near the edges of the frame becomes overly stretched.


the pictures are showing, exactly, what i said in my last post…

the lens has nothing to do with it… it’s because of where the camera is located… the pictures show it correctly- the camera is moving in and out… but the text is saying size relationships between objects are being caused by different lenses…
simply put, that is not correct.

the easiest and most accessible test you can do to see this with your own eyes:

• place your hand near your eye and notice your hand appears to be larger than the TV sitting over there…
• move hand to arm’s length and notice the TV is now bigger than your hand.

you see? focal length hasn’t changed at all in that test… according to the blog post, the size relationship should stay the same in this test since i’m using the same focal length but the relationship is changing … it’s because the distance from lens to object, the perspective, has changed.

Hey Jeff,
I’m not talking about the relative aspects of objects in the image being dependant of observer’s viewpoint and lens length…
I’m saying that on a 35mm camera, a 43mm lens length will provide an enlargement similar to the human vision.
And since we don’t see a defined square and we have two eyes and we perceive a little bit more than we see, we have a larger FOV than a 43mm lens on a 50mm camera. It’s more like 28mm or 24mm depending on the subject and the luminosity of the environment.

i guess there’s no real way to test or show this in a succinct manner until maybe we get some computers hooked up to our eyeballs :eyes: :computer:

but in anecdotal test form:
if i look straight ahead and place my right hand on my right ear, i can’t see it…
then, while continuing to look straight ahead, i begin moving my hand outwards to the right from my ear, i can see it when it’s ~2’ away… to me at least, my total field of view is way wider than that of a 24mm lens…

there isn’t any detail that far out in the peripheral but my eyes are still receiving perceptible information from that area…

i’m not sure what you mean by ‘enlargement’… are you saying that when looking through a camera with a 43mm lens that the objects will appear the same size as if we weren’t looking through the lens? because if so, the viewfinders also have various magnification (enlargement?) factors… or- i can look through a 50mm lens on camera-A and it looks different (smaller or larger) than a 50mm lens on cameraB… (but the pictures will be identical when printed at the same size)

same goes for the final image… you can enlarge a print for example… but that’s nothing to do with focal length.

Humans do not see by taking photographs with their eyes. Human vision, the combination of how the eye works and the brain works, is much more complex than that. A good comparison of the human eye and the camera is at http://www.lensrentals.com/blog/2009/03/the-camera-vs-the-eye

1 Like