Talk:Angle of view
From Wikipedia, the free encyclopedia
 |
This article is within the scope of WikiProject Films. This project is a central gathering of editors working to build comprehensive and detailed articles for film topics on Wikipedia. If you would like to participate, you can edit the article attached to this page, or visit the project page, where you can join the project and/or contribute to the discussion. | ||||||||||
|
|||||||||||
|
|||||||||||
|
|||||||||||
I'm not an expert, but the first sentence of the article seems incorrect to me. Surely all the sample images took up roughly the same area on the 35mm film? Soundray 21:14, 16 Feb 2004 (UTC)
- I don't think it's right either, and I don't remember what page I copied that from (they should both be changed). As I recall, I was talking with someone who was trying to explain to me how to define the term without using a circular definition, and used the chance to dump in some photos I'd taken. I believe the definition as is is clumsy and misleading, if not outright wrong (and the fault is, of course, mine--look at the version history). Koyaanis Qatsi 00:15, 17 Feb 2004 (UTC)
This page could really use a chart converting 35mm focal lengths to angle of view. Also, the formula to do this, with the film size as a parameter.
Contents |
[edit] Digital cameras
I just added the term 'sensor' as the digital alternative for 'film'. But steve's digicams says "Consumer digicam focal lengths are usually stated in terms of their 35mm film equivalents." So should the given focal length then be recalculated for the formula to work?
By the way, that page also says "For digital SLR cameras with interchangeable lenses it's more difficult as different cameras have different size sensors." Which complicates matters. DirkvdM 10:15, 2 February 2006 (UTC)
- I haven't used many digital cameras, so I don't know. But for the two Canon cameras that I do have, the real focal length is printed right beside the lens. Using this focal length, I calculated the angle of view of my Canon ZR90 camera to be 37 degrees. The angle of view I measured earlier was 36 degrees, so the calculation was correct (or at least accurate). Also, the size of the sensor (i.e. CCD) can be found on most detailed webpages about the digital camera. Bowlhover 13:45, 4 February 2006 (UTC)
[edit] Error!
In the "Calculating a camera's angle of view" section there is an error:
In the first formula (ƒ) should be "focal distance" instead of "effective focal length". Effective focal length is something like 28 for wide-angle, 50 for standard, 200 for telephoto objectives (35mm film), but that's not what it is about. It is about how far the object is from the camera (film or sensor), e.g. 50cm or 5m, or 23m.
The thing is, however, that I don't know how to relate this to macro, and other objectives, so someone should correct this article who knows what he is speaking about.
- The thing you're calling "effective focal length" is the 35-mm-field-of-view-equivalent focal length. Different beast. Dicklyon 14:58, 12 June 2007 (UTC)
-
- I agree with Dicklyon, the effective focal length describes the FL of your main lens + any optical elements you hang off the front or back (macro dioptres etc); it's like the gross or total FL of all the lenses in that system. It is correct for that formula, because AoV is the same at any object distance. (87.102.18.32 21:18, 27 August 2007 (UTC))
[edit] Optical Axis
My understanding is that the Horizontal and Vertical Angle of View are traditionally measured across the Optical Axis (centre of the lens / image) to avoid errors caused by Barrel Distortion or Pincushion Distortion. That way the AoV formula also works accurately for non-rectilinear lenses.
Currently the illustration does not show the diagonal, horizontal and vertical AoV arcs 'meeting' as they pass through the optical axis. The description (definition) should read something like the Vertical Angle of View is measured from the centre of the top edge of the image to the centre of the bottom edge of the image ... etc.
83.100.174.227 00:29, 27 August 2007 (UTC)
[edit] Still vs. motion picture 35mm film.
The frame size is needed to calculate the angle of view. A still 35mm film camera has a VERY different frame size than a 35mm motion picture camera (normal, not vistavision).
In a still camera, the film moves sideways and in motion picture cameras, the film moves vertically. That is why there is a vast difference in the size of the frame.
Therefore any reference to a 35mm camera MUST make it clear if this is a STILL or a MOTION PICTURE camera. Just saying a "35mm camera" or "35mm frame" is WRONG!!!! ~~~~ Robert Elliott (talk) 03:50, 3 January 2008 (UTC)
- I inserted "still" in the one place I could find that might remotely have been considered ambiguous. If there are others, just fix them. Dicklyon (talk) 04:44, 3 January 2008 (UTC)
[edit] Difference between Field of View & Angle of View in photography
Despite popular misuse (and I do it too!), saying that Field of View is interchangeable with Angle of View is simply not true! The funny thing is that knowing your Angle of View is relatively useless in practical photography, which is probably why Field of View is the more familiar term.
Field of View means simply how much of this scene is included in my picture. A larger FOV can be obtained by either increasing the AOV, or increasing the camera-object distance.
"Field of View" and "Depth of Field" are measures of distance (so many feet or so many metres) - they are concerned with objective dimensions. Angle of View is not dimensional, angles are dimensionless. FOV size increases proportionally with distance whereas AOV does not.
- The most important point is that objective dimensions (size and distance) are much easier to measure/estimate than relative angles. Angles of view may be useful to surveyors and astronomers, allowing them to calculate unmeasurable distances, but if I ask you to imagine the difference between a 18 degree elevation and a 26 degree elevation, you would be pretty hard pressed. However if I ask you to imagine looking at a 10ft wall from a distance of 30ft and then at a distance of 20ft, it gets a lot easier to visualise.
- Focal Length, Image Format, Minimum Focal Range, etc are always given as dimensional values - usually mm. Aspect Ratios describe proportions between rectilinear lengths, not angles and arcs. Converting between angles and dimensions involves more complex mathematics such as arctangent tables which you can't do in your head. For practical photography, these conversions are completely unnecessary and only add to confusion. The FOV formula is so simple you can do it in your head and it gives you useful information.
83.100.138.9 (talk) 02:56, 15 March 2008 (UTC)
[edit] FOV formula in practical photography
In landscape, architectural, underwater, macro and aerial photography, sizes and distances can be so large or critical that 'stepping back a bit' and 'zooming out a bit' are not enough to give the photographer the FOV he wants. Therefore photographers have devised a user-friendly FOV formula - as opposed to the AOV formulae given in the article - for use when 'setting up a shot' on location with minimum fuss.
Where
- FOV = Field of View as the dimension of the Objective Frame,
- F = Focal Length of the lens,
- I = Image (D, H or V dimension of the Sensor) and
- D = Distance between object and camera:
FOV = D * I / F
D is going to be the most variable factor, so if you let D = 1, then you get a FOV:D ratio of Objective Frame-Size : Distance. This ratio is constant for this prime lens on this camera. Note that the AOV is also derived from I and F (2 arctan (I / (2 F))), but FOV size increases proportionally with distance whereas AOV does not. With practice, and familiarity with the focal lengths of your lenses, this method of setting up a shot is extremely quick and efficient. Even zoom lenses have minimum and maximum AOVs and 'sweet spots' measured as focal lengths.
- Eg1 If I use my 28mm lens: I/F = 36mm/28mm so my FOV:D ratio is 9/7 which means I need to stand at least 70ft away from a 90ft building to get it all in frame.
- Eg2 Using a normal lens, 36mm/50mm gives me FOV:D ratio of 72%. If I want to photograph a 70mm-ish butterfly full frame with my normal lens, I need to get the camera about 100mm from the insect. However, I also know that the minimum focal range of this lens is 300mm, so I need to change the lens to get this shot. If I do this before I shove the camera at the butterfly, it's more likely to still be there when I'm ready to take the picture.
Lens Manufacturers use the Diagonal AOV reference because it corresponds the diameter of the Useful Image Circle and is unaffected by Aspect Ratio. However, FOV traditionally refers to the Horizontal FOV, because this is most frequently the one a photographer wants to use. Aspect Ratio? easy: if my AR is 4:3, my vertical FOV is going to be 3/4 of my horizontal FOV. (That doesn't work with AOV!)
83.100.138.9 (talk) 02:56, 15 March 2008 (UTC)
[edit] Angle of Coverage
When comparing the actual size/area of the lens' Useful Image Circle to the actual size/area of the camera Sensor, a (dimensionless) "Angle of Coverage" is just useless. The "back focal length" of a lens gives you this information in a useful form, but as this property of a lens system is certainly not adjustable and conforms incredibly precisely to the BFL defined by its Lens mount system, I see no reason to reinvent it here. It would make more sense to talk about a "Field of Coverage at the Focal Plane", but that is exactly what a lens' "Image Circle" is. Even in View Cameras, it is the size of the Lens' Image Circle which is important to Shift, not the angle of divergence of its peripheral rays - and the Tilt angle is relative to the optical axis (the angle of the lens' focal plane is always perpendicular to its optical axis) so again a "Back AoV" just doesn't come up.
83.100.138.9 (talk) 02:56, 15 March 2008 (UTC)
[edit] Non Rectilinear Lens - Radial Distortion
Quantifying Radial Distortion in Non-Rectinlinear Lenses is no mystery if you realise that Radial Distortion increases with distance from the optical axis. The corners of a rectangular image are furthest away from the optical axis and thus most distorted. As long as you measure Vertical and Horizontal FOV through the optical axis, ie between the mid-points of opposing sides of the frame, you will get accurate Focal Length readings. Radial Distortion of a lens can be quantified objectively by measuring the Diagonal FOV empirically and comparing it to the calculated (rectilinear) dFOV for the lens' focal length. Barrel distortion stretches the dFOV, whereas PinCushion Distortion shrinks the dFOV in proportion to the hFOV or vFOV. It is that simple. For the 'worst case' measure of radial distortion, you would compare the shortest FOV (usually the Vertical) to the longest (always the Diagonal). Of course marketing 'specs' for lenses usually employ different ways of measuring radial distortion in order to exaggerate how 'rectilinear' the lens is. As with all serious photographic equipment, price is a better indication of quality than marketing material.
83.100.138.9 (talk) 02:56, 15 March 2008 (UTC)
