Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
Best Technique to Test Corner Sharpness
Page <<first <prev 3 of 5 next> last>>
Mar 5, 2018 08:04:13   #
Joecosentino Loc: Whitesboro, New York
 
Print a test pattern and shoot it under good light a several f stops, in most cases with newer lenses my guess is it won't be much of an issue. If you have a lens that is shoot for the crop and crop out the softness

Reply
Mar 5, 2018 10:36:46   #
mikegreenwald Loc: Illinois
 
John_F wrote:
I would imagine your test shots would be affected by the f-stop in that that is your principle way to control depth of focus.


Right on! Brick walls are ubiquitous and free, and you don't need a model release! Look out for uniform lighting, and use a tripod for a sensitive test.

Reply
Mar 5, 2018 11:04:41   #
Retina Loc: Near Charleston,SC
 
tinplater wrote:
I started this thread because I was just curious about the technique of determining corner sharpness. If one takes a picture with the lens aimed at center, but focused on the corner, and it is sharp, should that lens be judged sharp in the corner (even though the center is not as in the images I provided above)?

If I were choosing among lenses for certain types of shooting where corner sharpness really mattered, I would naturally focus the test subject in its corner(s). Conversely, when assessing center sharpness, would you mainly focus for the corners and then see how the center does?

Reply
 
 
Mar 5, 2018 11:23:08   #
jcboy3
 
tinplater wrote:
Thanks but I don't understand the last sentence. If I want to take a SINGLE photograph and for the purpose of assessing corner sharpness, do I focus in the center of the target (meaning the corners are farther away than center) or do I keep camera centered on target but focus on corner (either manually or by moving the flexible focus spot)? It seems to me most people say corners are soft in images when in fact the corner may be in an entirely different focus plane than the center.


With a single exposure, you want to focus in the corner. That will let you know whether that particular corner is sharp, in and of itself. You will need 4 exposures (one for each corner) to assess overall corner sharpness. Then, you will need one exposure focussed in the center, to determine whether the corners are in focus when the center is in focus.

Reply
Mar 5, 2018 12:57:54   #
ORpilot Loc: Prineville, Or
 
Architect1776 wrote:
I use a newspaper.


πŸ‘πŸ»πŸ‘πŸ»πŸ‘πŸ». We used a newspaper to test our lenses ot Ohio University Photo Department. It worked in 1971 it should still work today.

Reply
Mar 5, 2018 13:07:30   #
tinplater Loc: Scottsdale, AZ
 
Again, what bothers me is a lens is pronounced "soft in the corners" (this is a frequent occurence) when one would almost expect that to be the case because the corners are farther away from lens center than the center of the image. How one quantifies this is beyond me? Focal length and distance from the object clearly come into play as well.

Reply
Mar 5, 2018 14:15:01   #
rgrenaderphoto Loc: Hollywood, CA
 
RWR wrote:
I recall that for the USAF charts I have, the recommended distance is 26 X the focal length, but any lens can only be optimized at one distance and there is a wide variation in lens design. There’s a reason I’ve never used a wall chart, but it could pass the time on a stormy day. Have fun with whatever you come up with!


So, for a 24 mm lens, that would be 480 feet? Seriously?

Reply
 
 
Mar 5, 2018 14:20:18   #
mwsilvers Loc: Central New Jersey
 
rgrenaderphoto wrote:
So, for a 24 mm lens, that would be 480 feet? Seriously?

That's a piece of cake. I could just set up a target at the other end of my basement.

Reply
Mar 5, 2018 14:33:29   #
therwol Loc: USA
 
rgrenaderphoto wrote:
So, for a 24 mm lens, that would be 480 feet? Seriously?


Okay. Math answer. 24mm x 26= 624mm = 62.4 cm = 24.6 INCHES (approx).

Reply
Mar 5, 2018 15:04:38   #
John_F Loc: Minneapolis, MN
 
I bungled. Where I wrote "focus" I meant "field".

Several have written that the corners are somewhat further distant than the center. It is true that the actual distance of the corners is greater than rhe perpendicular distance, but those are not the optical distances. Every lens has central axis on which the origin of the radii of curvature of all lens surfaces lie. Every lens has 4 cardinal optical planes: two principal planes and two focal planes, one for lens front and one for lens rear, the distance between the principal and focal planes is the focal length - these planes are mathematical constructs. Subject distances are conveniently measured from the front principal plane so one doesn't have to deal with negative numbers. If you set your lens to 10 feet there will be an optical plane at 10 feet in which everything will be in focus. If the lens optic axis is perpendicular to the wall, the entire wall will be in focus. Of course, depth of focus comes into play if the wall is not perfectly flat. Some lenses do show corner softness and this arises from lens elements are not perfectly spherical.

John_F wrote:
I would imagine your test shots would be affected by the f-stop in that that is your principle way to control depth of focus.

Reply
Mar 5, 2018 16:10:20   #
Retina Loc: Near Charleston,SC
 
John_F wrote:
...the actual distance of the corners is greater than rhe perpendicular distance, but those are not the optical distances. Every lens has central axis on which the origin of the radii of curvature of all lens surfaces lie. Every lens has 4 cardinal optical planes: two principal planes and two focal planes, one for...

Nature sure was smart when she decided on a curved eyeball.

Reply
 
 
Mar 5, 2018 16:42:15   #
wdross Loc: Castle Rock, Colorado
 
tinplater wrote:
All the talk about "soft corners" just wondering what is the best way to asses/test the corners? Assume you have a test chart or a nice big brick wall..something flat. Do you focus on the center then examine the corners? Or do you move flexible focus spot to the corner and shoot, or do you use the center focus but focus at corner, keep focus and then go back to center and shoot? Focusing at the center would seem to bring into play the fact that the corners are farther from the lens center than the center portion of the image.
All the talk about "soft corners" just w... (show quote)


I would use a newspaper on a wall and use the camera's leveling system. Focus at the middle of the newspaper, take a shot, and examine the corners for chromatic aberration, distortion, etc. Change the aperture and shoot again. Examine the corners again. Continue until you have satisfied your interest. Lenses are designed for a relatively flat plane at the point of focus. The corners may be better in-focus a little in front or a little behind that center focus point but should be of relative insignificant especially compared to the subject of interest in the main part of the frame.

Reply
Mar 5, 2018 19:13:32   #
Jim Bob
 
tinplater wrote:
All the talk about "soft corners" just wondering what is the best way to asses/test the corners? Assume you have a test chart or a nice big brick wall..something flat. Do you focus on the center then examine the corners? Or do you move flexible focus spot to the corner and shoot, or do you use the center focus but focus at corner, keep focus and then go back to center and shoot? Focusing at the center would seem to bring into play the fact that the corners are farther from the lens center than the center portion of the image.
All the talk about "soft corners" just w... (show quote)


Shoot a flat object wide open.

Reply
Mar 5, 2018 19:22:34   #
tinplater Loc: Scottsdale, AZ
 
John_F wrote:
I bungled. Where I wrote "focus" I meant "field".

Several have written that the corners are somewhat further distant than the center. It is true that the actual distance of the corners is greater than rhe perpendicular distance, but those are not the optical distances. Every lens has central axis on which the origin of the radii of curvature of all lens surfaces lie. Every lens has 4 cardinal optical planes: two principal planes and two focal planes, one for lens front and one for lens rear, the distance between the principal and focal planes is the focal length - these planes are mathematical constructs. Subject distances are conveniently measured from the front principal plane so one doesn't have to deal with negative numbers. If you set your lens to 10 feet there will be an optical plane at 10 feet in which everything will be in focus. If the lens optic axis is perpendicular to the wall, the entire wall will be in focus. Of course, depth of focus comes into play if the wall is not perfectly flat. Some lenses do show corner softness and this arises from lens elements are not perfectly spherical.
I bungled. Where I wrote "focus" I meant... (show quote)


Thanks for the information...unfortunately math and I do not get along well. I wish I could meet with you in Mickey's Diner, share a heart attack on a plate, and have you tutor me through this. But if I understand you, a lens should be able to have corners exactly as sharp as the center despite being at different distances from the sensor?

Reply
Mar 5, 2018 19:39:27   #
therwol Loc: USA
 
Jim Bob wrote:
Shoot a flat object wide open.

The problem with some lenses, and I have one, is that the corners may remain soft at any f/stop. Also, many lenses, particularly older designs, sharpen up considerably after being stopped down a bit. The wide open test is only valid if you plan to shoot wide open.

Reply
Page <<first <prev 3 of 5 next> last>>
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.