DoF and Perspective Revisited

[Dick Roadnight]

A good way to think about perspective is to think about objects occluding other objects. When you dont move your head, of course everything remains static. If you move your head around, the occlusion of objects, the direction of lines, etc, change.
Christian

This is parallax, which may not, technically be "converging lines" perspective!

[Ray]

A single frame of the stitched image will have different FOV when different focal length lens are used. The FOV of a stitched image can be the same for any focal length and (here's the part you don't seem to believe) the perspective will be the same.

Of course I believe it. Any image of the same FoV taken from the same position, and taken with any lens on any format of camera will have the same perspective.

The only point I am making is that FoV is related to effective focal length. If the FoV of two images is different, despite the fact that both images have been shot from the same position, then the subject matter is different, the focal length of lens used is different and therefore the perspective is different because perspective has to relate to a specific subject. Or perhaps you would like to argue that the perspective of a non-existent subject can be the same as the perspective of an existing subject.

How does a person have perspective?

A person viewing a photographic image must view it from a specific position and that position will influence the sense of perspective in the image, as experienced by the viewer, but will not soley determine perspective. The image itself has its own perspective determined by the relative size of all identifiable objects within the composition.

In my examples above, from Angkor Wat, which show a pair of images of the same size, but the first one a small crop of the other, the size of the images (or prints) affects the experience of perspective in the viewer. Large prints can be appreciated from a greater distance than small prints.

Thierry has claimed that the perspective of my enlarged crop is the same as that portion of the image in the full scene from which the crop was made.  Well, of course it is. Two equal images are equal.

The interesting question is, how would anyone who didn't know beforehand that one image was an enlarged crop of the wider FoV image, be able to discern that the cropped image, of equal size to the whole image from which the crop was taken, be able to discern that the perspective was the same?

My assertion is, they wouldn't be able to unless they effectively altered their own subjective perspective of the full scene, by examining it up close, and making the the same visual crop in their mind that I had made in Photoshop, then retreating to compare it with the other image.

Sorry, your analysis is just not true.  I think the only way you will be able to realize this is to stitch several images together yourself.

The first image I ever sold was a 13 image stitch from 35mm film about 15 years ago. Sold it to the local City Council. I'm very much in favour of the practice of stitching. I'm pleased that stitching programs continue to improve. I really do understand that you can use any lens for stitching and get the same perspective in the same scene. I'm just uncomfortable with the statement that you can get the same perspective in two different scenes, or whatever the scene.

[BernardLanguillier]

We look forward to reading you comments... if you were not on the other side of the world, we could team up to do some comparisons, and save you the $800!

I have started a thread to compare the P65+ to the H3D11-60.

In the mean time, one example of a handheld 300mm image.

http://www.flickr.com/photos/bernardlangui...647680/sizes/o/

I find it to be just as sharp as the P65+ 100% crop samples just posted at outbackphoto, but I guess you will tell me than they messed up...

Cheers,
Bernard

[Panopeeper]

More examples: http://www.phototakers.com/forum/ftopic36480.html

These are not "more examples" but "examples". The drawing of the projection of a single point is neither an example for nor a demonstration of "perspective".

[Jack Flesher]

This is parallax, which may not, technically be "converging lines" perspective!

And parallax is part and parcel to perspective since it dictates how 3D objects will appear relationally when rendered in 2D...

[Jack Flesher]

In the mean time, one example of a handheld 300mm image.

http://www.flickr.com/photos/bernardlangui...647680/sizes/o/

I find it to be just as sharp as the P65+ 100% crop samples just posted at outbackphoto, but I guess you will tell me than they messed up...

Cheers,
Bernard

Resolution maybe, DR no...

[tho_mas]

I find it to be just as sharp as the P65+ 100% crop samples just posted at outbackphoto, but I guess you will tell me than they messed up...

yes  
f22 and f18 are certainly not the apertures to show the P65+ files at the best  

[Panopeeper]

In my examples above, from Angkor Wat, which show a pair of images of the same size, but the first one a small crop of the other, the size of the images (or prints) affects the experience of perspective in the viewer. Large prints can be appreciated from a greater distance than small prints.
...
The interesting question is, how would anyone who didn't know beforehand that one image was an enlarged crop of the wider FoV image, be able to discern that the cropped image, of equal size to the whole image from which the crop was taken, be able to discern that the perspective was the same?

Ray, honestly, the entertaining value of the faked lack of understanding of the basics is wearing off.

[Dick Roadnight]

I find it to be just as sharp as the P65+ 100% crop samples just posted at outbackphoto, but I guess you will tell me than they messed up...
Cheers,
Bernard

Looks OK, Bernard... but pictures look sharp if the kit can resolve the detail in the picture... not all pictures have detail that cannot be seen in a hand-held 300mm shot taken with a Small DSLR with an anti-aliasing filter.

It a good example of a shot for which pan-and-stitch is not suitable!

[cmi]

This is parallax, which may not, technically be "converging lines" perspective!

Sure. I could have said this for myself. But I think its still clear what I wanted to say

[Doug Peterson]

I find it to be just as sharp as the P65+ 100% crop samples just posted at outbackphoto, but I guess you will tell me than they messed up...

+1 Tho_Mas comment that the P65+ does not sing at f/22 as used in the Outback Photo article. See this test for an example of 100% pixel sharpness at varying f-stops with a P65+ using Phase One glass. Bare in mind that the f/22 shot will still make an impressive print, but for maximum detail it is not the best f-stop.
+1 the incredible dynamic range / virtually non-present noise of the 65+ are important to bare in mind when you're getting too far into pure-resolution discussions. You'll enjoy these when you rent the 65+ soon.

Doug Peterson
__________________
Head of Technical Services, Capture Integration
Phase One, Leaf, Canon, Apple, Profoto, Eizo & More
National: 877.217.9870  |  Cell: 740.707.2183
Newsletter: Read Latest or Sign Up
RSS Feed: Subscribe

[JeffKohn]

+1 Tho_Mas comment that the P65+ does not sing at f/22 as used in the Outback Photo article. See this test for an example of 100% pixel sharpness at varying f-stops with a P65+ using Phase One glass. Bare in mind that the f/22 shot will still make an impressive print, but for maximum detail it is not the best f-stop.
+1 the incredible dynamic range / virtually non-present noise of the 65+ are important to bare in mind when you're getting too far into pure-resolution discussions. You'll enjoy these when you rent the 65+ soon.

Looking at DxO Mark comparison of D3x vs P65+ paints a difference story. Setting aside high-ISO performance and pixel count, these two sensors are actually pretty similar in performance. The P65+ has a slight edge in some categories, but not others (including Dynamic Range - where the D3x comes out on top, probably due to cleaner shadows). DxOMark doesn't measure/score on resolution, and the MFDB will have an advantage there, even at the pixel level due to not having an AA filter. But if you apply the appropriate capture sharpening to the D3x shot and compare at the pixel level, it's not going to be the huge night and day difference that some MF proponents would have us believe.

[BernardLanguillier]

It a good example of a shot for which pan-and-stitch is not suitable!

True, but also a good example of a locale/subject where a back is of little use...  I don't believe that many humans have the physical ability to climb there with a 500mm Mamiya lens together with a pro level anti-vibration gyro device...

Cheers,
Bernard

[BernardLanguillier]

Resolution maybe, DR no...

How can you tell? I have significantly increased the contrast of this image because that is how I think it looks good.

Shadows are totally clean and could be pushed a lot. Of course you'd have to do it before sharpening for best results.

Cheers,
Bernard

[Nick Rains]

The only point I am making is that FoV is related to effective focal length. If the FoV of two images is different, despite the fact that both images have been shot from the same position, then the subject matter is different, the focal length of lens used is different and therefore the perspective is different.

Sorry Ray, this is quite incorrect.

Focal length, FoV have absolutely nothing to do with perspective or parallax. The relationship between objects in any scene is fixed when viewed from a single point. Change the viewpoint and the perspective, parallax, whatever, changes but only if you move the point.

Remember that focal length and FoV are not the only two variables here, and are only loosely connected.  FoV depends on the sensor size and image circle of the lens design. Focal length is, at it's simplest, just a measure of the distance from the lens centre to sensor plane when the lens focuses at infinity. A 100mm lens on a large film/sensor (10x8) is a wide angle. A 100mm lens on a GH1 is a telephoto. How then can changing the sensor (by changing cameras) affect the perspective?

http://jamesmskipper.tripod.com/jamesmskip...erspective.html

These images make the point clearly.

[Jack Flesher]

How can you tell? I have significantly increased the contrast of this image because that is how I think it looks good.

I know because I've shot both and worked extensively with the files. And anybody else that has worked extensively with both will tell you the same thing...  

[BernardLanguillier]

I know because I've shot both and worked extensively with the files. And anybody else that has worked extensively with both will tell you the same thing...

OK, fair comment. I assume that you used C1 Pro to convert both these files? Did you do side to side comparisons?

If it is the case, that might save me 800 US$ if these were available.

Thanks.

Cheers,
Bernard

[Ray]

Ray, honestly, the entertaining value of the faked lack of understanding of the basics is wearing off.

I often find that disputes about issues result from the issues not being clearly defined in the first instance, so people can find themselves actually arguing about different things although they might think they are arguing about the same thing.

How about the following definition.

As a theoretical construct and property of lenses, unrelated to the making of a picture or composition and therfore excluding the perspective of a potential viewer, the theroetical perspective through the lens is independent of its focal length or the format of the camera it is attached to, and is determined only by the distance between the lens and the theoretical subject.

However, when creating a photographic composition, which most photographers try to do now and again, the focal length of the lens and the format size of the camera will both influence the distance to the subject(s), the size of the subject within the composition and the number of individual subjects or objects within the composition at varying distances from the photographer.

How's that as a piece of clarity?  

[Panopeeper]

How about the following definition.

• It is not a definition but an explanation.
  
  
• It relies on your incorrect interpretation of perspective (it is circular).
  
  
• How about sticking to the existing, working meaning of perspective? You are free to use rayspective to your purposes (or devise another, new term for that), but I don't think that photographers will change their correct interpretation of a customary term for your sake.

[Ray]

So yes if you are refering to our common sense - we tend to totally neglect our own effect (walking around!) on perspective and only see what different lenses do  - then yes, in common sense, a wideangle gives "a nice perspective", but if we take that apart and really try to understand HOW it works, we need to rethink some of our common phrases, and eventually come to realize that it works somehow different then common sense suggests. And yes, if you where saying different lenses give different viewing experiences (after all, we have them for a reason) if you refer to THAT well there is no dispute that this is correct.

To end here, I read your last posts and tried to understand how you where thinking about it, I hope this makes somehow sense to you.

Christian,
Regardless of whether a wide-angle lens produces a nice perspective or a horrible perspective, it produces a different image to that produced by a longer lens (from the same position) and the objects in that different image will relate in a way which produces a changed perspective of the viewer, will they not?