Reading mode: Light Dark

TSE 24mm II - field curvature

Go to last post
zerozeronine said:
That's very nice, Ben!
Click to expand...
Thanks, this was a set-up shot to test focus and tilt. I took another at f8 I believe.
zerozeronine said:
I think I understand the typical use of the tilt with landscapes (that is, you measure the vertical distance to the camera, and tilt the lens accordingly to make the ground plane in focus).

The question I have is what's the best workflow/mnemonic to figure out how much of an angle the wedge of focus will have?

For example, in your photo, you would try to get the flowers and ground in focus by measuring the distance to the ground, but how do you know whether that wedge is thick enough (by stopping down the aperture) to have the top of the mountain still be in focus?
Click to expand...
I never measure, I did a whole lot of practice in my yard to get a feel for this. Here is how I do it.

If at full height, and no tall subject near, 1 degree ( I think 0.9 technically) is right. I start with focus at distance then tilt until near subjects is sharp. I use live view and sometimes my Hoodman at 5X. !0 X on the 1DS-mk3 is too much as the pixels are too big.

If there are tall near subjects, tilt is not the way to go because there is no true plane of focus. Same for low shots with say a bush in the foreground. In these cases, I use hyperfocal, which is about 8 feet for f11. This will be very sharp from 4 feet to infinity. In these cases, I will also do one focused further out.

By the way, if you focus at infinity and f11, the near limit is 8 feet. Easy to remember, the near limit at infinity is about the same as hyperfocal, and at hyperfocal, the near limit is about 1/2 hyperfocal.

Also, I have found that when I really want some near object sharp, just make it sharp and stop down. This lens will not go all blurry on you at infinity like a 17-40 when you do this. Razor sharp near from corner to corner will really grab the image.

For low shots, (about 3 feet when I am setting) I do the same as tall shots but need about 2 degrees, and again, focus far and tilt for near. I always check the far focus again after tilting.

My eyesight is poor and focusing has been a challenge for me the last 30 years. I never trust what I see in the field so I do a lot of bracketing.
zerozeronine said:
(Of course, we can always look at the photo to see if things are in focus, but I'm looking for a way to get it roughly right before taking the shot.)

Kaz
Click to expand...
--
When you can't focus, nothing else matters
Once you can, everything else does.

http://ben-egbert.smugmug.com/

Ben
 
Symple said:
Rick Knepper said:
are you trying to persuade or dissuade us from buying? What is your premise?
Click to expand...
I think his premise was to get a job as a lens tester rather than as a photographer.

This is a waste of a great lens! I own one and would be pretty sad if this is what I came up with from it myself. Maybe the whole image is redeeming, but the parts are lame.

Maybe try the same series in a low contrast monochrome to finish the effect.
Click to expand...
or maybe he was providing helpful information and where to focus to get best overall sharpenss???

or maybe the TS-E is too much lens for you and you should go use a lens baby instead
 
rlshep said:
Why the sarcasm?

I would have thought working out how to get the best performance out of a new lens is perfectly legitimate, helpful for oneself and potentially for other people, and certainly not something to cause irritation to anyone else.
--
Lizzie
----------------------------------
http://www.lizzieshepherd.com
Click to expand...
Indeed. This would be your opportunity to click 'ignore user' on my comments, and submit a complaint as a representative of the sarcasm police here on the forum.

The 'best performance' part escapes me, and do I appreciated Rick's sincere query as to the premise of the original post, which is why I even bothered to quip.

Ben and Rick have been the better part of this forum for some time, and are thoughtful and civil; I am not it seems.

To me, this post looks to be another 'I just spent X amount of dollars on a lens and it still isn't good enough' post since there is no actual question or outcome desired indicated (maybe there has since?), and the images have no context or purpose in themselves other than to exhibit the properties of a given lens under certain conditions (lens review).

Are we to comment on how to build a better lens or what? Maybe I missed something like 'I am planning to use this lens to stitch multiple frames together to render distortion free architectural landscapes' - in which case I would recommend to use shift and or tilt sparingly with this lens, and instead pan using a nodal slide and stitch your images together in rows while still achieving the benefits of moderate movements.

By the way lizzie, without sarcasm, your gallery of images is absolutely gorgeous and you are a very skilled artist, and I still recommend you block me because I am irritating to most people, and obviously you too.
 
bronxbombers wrote:
nochrome to finish the effect.
bronxbombers said:
or maybe he was providing helpful information and where to focus to get best overall sharpenss???

or maybe the TS-E is too much lens for you and you should go use a lens baby instead
Click to expand...
Maybe. I wish he would have clarified at the outset.

Truthfully, all sarcasm aside, you are more correct about the TS-E and the lensbaby than I would like to admit, it is too much lens for me. With some luck I will grow in to it. When people have a question or a statement for the members, why don't they just ask/say it [like you did with the lensababy tip]?

I'll try, but since I have the shift/pan/stitch thing mostly covered now I will honestly ask [yeah yeah, start a new thread I know, but since I will get flamed either way]:
  • I have been trying to focus in the dark with my 24 TS-E II, but it just hasn't been working out for me? I have been having an easier time finding out where to focus in the daylight and twilight, but with this focal length it seems especially difficult to focus accurately without a point light-source to set focus with, any useful tips? I have tried live-view, a flashlight, and of course the viewfinder, but combined with movements I am having a difficult time getting accurately focused images. I can get sharp images, but only with point light sources like the example I have included. Also frustrating is the fact that this lens focuses well past the indicated infinity setting, and in the dark it is easy to bump out of focus, and even a slight turn past infinity on the TS-E II can render everything out of focus. Thanks in advance everyone for your valuable information on how to help me improve my use of this lens for my future night shots where I may want to have star trails visible behind a large rock bluff that itself has very little light to focus on for the foreground and tilt; oh yeah, I can't really get there until it is already dark, so waiting is not and option. Thanks again for the useful tips.


 
Rags_and_Old_Iron said:
  • I have been trying to focus in the dark with my 24 TS-E II, but it just hasn't been working out for me? I have been having an easier time finding out where to focus in the daylight and twilight, but with this focal length it seems especially difficult to focus accurately without a point light-source to set focus with, any useful tips? I have tried live-view, a flashlight, and of course the viewfinder, but combined with movements I am having a difficult time getting accurately focused images. I can get sharp images, but only with point light sources like the example I have included. Also frustrating is the fact that this lens focuses well past the indicated infinity setting, and in the dark it is easy to bump out of focus, and even a slight turn past infinity on the TS-E II can render everything out of focus. Thanks in advance everyone for your valuable information on how to help me improve my use of this lens for my future night shots where I may want to have star trails visible behind a large rock bluff that itself has very little light to focus on for the foreground and tilt; oh yeah, I can't really get there until it is already dark, so waiting is not and option. Thanks again for the useful tips.
Click to expand...
Thats a tough one. I tried to do a milky way shot at Mount Rainier last year and had no focus target, however it was the same problem with an AF lens.

I eventually used the moon which while not in my final image was a good focus target. If you have nothing for a target, its pretty tough. But anything at infinity can be used and then just reframe.
Rags_and_Old_Iron said:
Click to expand...
--
When you can't focus, nothing else matters
Once you can, everything else does.

http://ben-egbert.smugmug.com/

Ben
 
Hello,
Interesting matter.

I have many experience with true wide-angle lenses on traditional sheet-film cameras, 4x5" to 20x25cm. I actually use often PA-Curtagon 4/35mm on Canon 5DII.
Your test is interesting, and I could guess the results.
Some remarks:
  • there is a new version of the Canon TS 24 (II)
  • this type of lens is very complex, not only for achieving widest angle, but also for making this in retro-focus design
  • mechanical constrution is very delicate; very small error in Tilt setting can cause problems
  • Live view on 5DII with autofocus does not work properly; here is a manual stting: why not set disdance manually instead of focusing (in this case, not close-up)
  • retro-focus lenses tend to loose quality at borders, have usully more distortion thas true wide-angle (eg Biogon)
Cheers,
Pol
 
Ben

I'd expect to see more issues with distant subjects and also shifted fully to left or right (assuming landscape orientation) - I'm sure someone will shoot me down in flames for saying so, but generally this is where WA or UWA is likely to start to fall down in my experience.

Either way, I think the original post shows you can get great detail fully shifted for distant subjects with the right aperture/focusing.
--
Lizzie
----------------------------------
http://www.lizzieshepherd.com
 
That tends to be my starting point when using the 17TSE with much shift. However I suspect I need to refine this somewhat if I want to do the same with a 1.4x attached. Needs more experimentation - generally I've used it without TC so not an issue.
--
Lizzie
----------------------------------
http://www.lizzieshepherd.com
 
I've never actually tried 'ignore user' and have not felt the need to date! I was not irritated by your post - just didn't feel there was a need to be uncivil. I knew where the OP was coming from because of previous threads on here - perhaps not clear to everyone but asking a polite question tends to get a better response in my experience! ;-)
But all that aside, thanks very much for your kind words...
--
Lizzie
----------------------------------
http://www.lizzieshepherd.com
 
mattr said:
**** England said:
mattr said:
http://www.luminous-landscape.com/tutorials/focusing-ts.shtml
Click to expand...
Click to expand...
I agree. To me this paragraph doesn't make much sense.
Click to expand...
These are the critical sentences in the article:
mattr said:
It is necessary for the lens not to be a flat field design to accommodate ‘shift’. > Imagine what would happen to your focus when you shifted left or right if the edges of > the lens focused further away from the centre! When shifted the centre focus would > jump forward, magnification would change ever so slightly making stitching difficult.
Click to expand...
For a WA TSE lens during shift stitch, the edge of the center frame is further away, but still in focus if the lens were flat field. When shifted to the max this more distant object is now in the center of the shifted image and thus is now behind the plane of focus (we are assuming max wide aperture, thus minimun DOF). The shifted frame is not refocused, but simply shifted and shot taken. In practice smaller apertures are normally used to increase dof, so this becomes less of problem, unless you are pixel peeping as in the OP.
Mike K
 
rlshep said:
Ben

I'd expect to see more issues with distant subjects and also shifted fully to left or right (assuming landscape orientation) - I'm sure someone will shoot me down in flames for saying so, but generally this is where WA or UWA is likely to start to fall down in my experience.
Click to expand...
Right my example was not very informative with respect to the problem posed. As you know, I don't often shoot portrait, and most landscape oriented shots need a bit of vertical shift to correct the horizon. Since I don't do panos, I have only done left or right shift when practicing. I also tend to keep my lens rotated for vertical shift when in landscape mode.
rlshep said:
Either way, I think the original post shows you can get great detail fully shifted for distant subjects with the right aperture/focusing.
--
Lizzie
----------------------------------
http://www.lizzieshepherd.com
Click to expand...
--
When you can't focus, nothing else matters
Once you can, everything else does.

http://ben-egbert.smugmug.com/

Ben
 
Mike K said:
Mike K said:
It is necessary for the lens not to be a flat field design to accommodate ‘shift’. > Imagine what would happen to your focus when you shifted left or right if the edges of > the lens focused further away from the centre! When shifted the centre focus would > jump forward, magnification would change ever so slightly making stitching difficult.
Click to expand...
For a WA TSE lens during shift stitch, the edge of the center frame is further away, but still in focus if the lens were flat field. When shifted to the max this more distant object is now in the center of the shifted image and thus is now behind the plane of focus (we are assuming max wide aperture, thus minimun DOF). The shifted frame is not refocused, but simply shifted and shot taken. In practice smaller apertures are normally used to increase dof, so this becomes less of problem, unless you are pixel peeping as in the OP.
Click to expand...
Hmm... I still don't understand the argument.

The TS-E lenses have a larger image circle. You could just use them on a camera with a large flat sensor that covers the whole image circle. I think nobody would argue that some field curvature would be necessary with such a camera.

Shift in the TS-E lenses simply substitutes for a larger sensor camera by providing a mechanism how the smaller sensor can sample different parts of the image circle. Why would field curvature now be necessary?
 
mattr said:
If you really want to shoot scenes fully shifted with distant detail at the edges of the extended image circle, f/11 and focusing slightly closer than infinity is probably the best compromise for diffraction versus edge performance.
Click to expand...
Leave it to mattr to get to the "heart of the"... :p

Good work, and thanks for the samples. I don't have the patience to test my gear that thoroughly (when I got the Sigma 50/1.4 and was first to post, I almost pulled my hair out over all the permutations involved), but using your advice will help in the field when opting for a full shift.

--
-CW

よしよし、今日も生きのいい魂が手に入ったな
 
One last try with this explanation.

Imagine the TSE lens has a perfectly shperical or perfetly flat field. These are hypothetical examples to emphasize the point.

The camera is at position named X. You are focusing straight ahead on an object 10 ft away (call it A). Thus the focal distance is X-A is 10 ft. We are at maximum size aperture so the dof is quite small. There is an object called B, on the the edge of your image that is perpendicular to XA. Thus if your lens were perfectly flat field, both A and B would be perfectly in focus. For the sake of simple math our TSE lens is the 17mm with almost 90 degree view, so XB is at approximately a 45 degree angle from XA. Thus the distance of XB is the square root of 2 or about 14 feet.

We now shift the lens all the way and object B now lies in the center of our image. We do not refocus but simply take another shot after shifting. However our lens is focused at 10ft in the center of the frame, and XB is 14 feet. Thus the effect is that B is now 4 feet behind the focal plane and out of focus.

How about the spherical field example? XA is 10 ft and defines the focus distance. B is 14 ft away from the camera X, at an angle of 45 degrees from XA, and is 4 ft behind the focal plane. Lets shift the TSE lens so that B is now in the center of the image. B is still 4 ft behind the focal plane, thus whatever misfocus was in the first image is still in the shifted second image. The shift stitch should align in focus and in magnification.
Mike K
 
Mike K said:
We now shift the lens all the way and object B now lies in the center of our image. We do not refocus but simply take another shot after shifting. However our lens is focused at 10ft in the center of the frame, and XB is 14 feet. Thus the effect is that B is now 4 feet behind the focal plane and out of focus.
Click to expand...
No, B will still be in focus if the lens has a flat field.
 
Symple said:
  • I have been trying to focus in the dark with my 24 TS-E II, but it just hasn't been working out for me? I have been having an easier time finding out where to focus in the daylight and twilight, but with this focal length it seems especially difficult to focus accurately without a point light-source to set focus with, any useful tips? I have tried live-view, a flashlight, and of course the viewfinder, but combined with movements I am having a difficult time getting accurately focused images. I can get sharp images, but only with point light sources like the example I have included. Also frustrating is the fact that this lens focuses well past the indicated infinity setting, and in the dark it is easy to bump out of focus, and even a slight turn past infinity on the TS-E II can render everything out of focus. Thanks in advance everyone for your valuable information on how to help me improve my use of this lens for my future night shots where I may want to have star trails visible behind a large rock bluff that itself has very little light to focus on for the foreground and tilt; oh yeah, I can't really get there until it is already dark, so waiting is not and option. Thanks again for the useful tips.
Click to expand...
well that's kind of a tricky one

did you try liveview M mode ISO whatever max the camera goes to and f/3.5 and a very long exposure and exposure sim mode?
 
mattr said:
Mike K said:
We now shift the lens all the way and object B now lies in the center of our image. We do not refocus but simply take another shot after shifting. However our lens is focused at 10ft in the center of the frame, and XB is 14 feet. Thus the effect is that B is now 4 feet behind the focal plane and out of focus.
Click to expand...
No, B will still be in focus if the lens has a flat field.
Click to expand...
A in the center of the frame as well as B at the edge of the frame were in flat field focus for the unshifted shot. A is centered and 10 ft from the camera and B is 14 ft at the edge of the frame. After the lens is fully shifted, B is now centered in the frame and still 14 feet from the camera. However we have not refocused the camera, and the focal plane is still at 10 ft. The focal plane is at 10 ft, B is at 14 feet, thus B is 4 ft behind the focal plane.
Mike K
 
Mike K said:
mattr said:
Mike K said:
We now shift the lens all the way and object B now lies in the center of our image. We do not refocus but simply take another shot after shifting. However our lens is focused at 10ft in the center of the frame, and XB is 14 feet. Thus the effect is that B is now 4 feet behind the focal plane and out of focus.
Click to expand...
No, B will still be in focus if the lens has a flat field.
Click to expand...
A in the center of the frame as well as B at the edge of the frame were in flat field focus for the unshifted shot.
Click to expand...
Yes.
Mike K said:
A is centered and 10 ft from the camera and B is 14 ft at the edge of the frame.
Click to expand...
Yes.
Mike K said:
After the lens is fully shifted, B is now centered in the frame and still 14 feet from the camera.
Click to expand...
Yes. And B is still in focus. Why would shifting the sensor in the focal plane change whether B is in focus?
Mike K said:
However we have not refocused the camera, and the focal plane is still at 10 ft.
Click to expand...
Yes, the center of the image circle is focused at 10ft. Shifting the sensor doesn't affect it.
Mike K said:
The focal plane is at 10 ft, B is at 14 feet, thus B is 4 ft behind the focal plane.
Click to expand...
If the lens is flat field, B at 14 ft is in focus because it is not in the image center (sharp at 10 ft). You said that yourself above. The position of B in the image circle did not miraculously change when the sensor was shifted.
 
At least we have clarified where our perceptions differ.

When we run across a flat field 24 tilt shift we can test this out.
Mike K
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top