JJGP
Senior Member
I don't think the "sweet spot aperture" and "sweet spot zoom range" are hard fast rules, more like very general rules of thumb. May not apply to all lens makes, for instance my Sigma 100-300f4 gets pretty good results @ 300 mm.
jj
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
Hyperfocal distance and soft exposures. . .
- Thread starter Steve Dorsey
- Start date
jj
Steve Dorsey
Forum Enthusiast
I actually tried that.
I focused perfectly on the dish itself, and got the same result. A slight blur on the overall image. No crispness at all.
I'd shoot, then immediately check my on-camera screen by zooming in to maximum on the pic and checking out the detail. Then I'd change my settings and try again. I was never able to get a true, crisp image.
I'm learning that this is probably due to 2 things; atmospheric heat distortion in the air, and my poor knowledge and understanding of the "circle of confusion" concept. I'm doing my best to learn about both.
Aside from the heat distortion, I'd think that I should be able to get a great, crisp image from my nice 70/200 lens on a shot from that far away. Is this true?
Thanks for the post,
------------S
----------------------
Steve Dorsey
http://www.dorseygraphics.com
http://www.flickr.com/photos/dorseygraphics
I focused perfectly on the dish itself, and got the same result. A slight blur on the overall image. No crispness at all.
I'd shoot, then immediately check my on-camera screen by zooming in to maximum on the pic and checking out the detail. Then I'd change my settings and try again. I was never able to get a true, crisp image.
I'm learning that this is probably due to 2 things; atmospheric heat distortion in the air, and my poor knowledge and understanding of the "circle of confusion" concept. I'm doing my best to learn about both.
Aside from the heat distortion, I'd think that I should be able to get a great, crisp image from my nice 70/200 lens on a shot from that far away. Is this true?
Thanks for the post,
------------S
----------------------
Steve Dorsey
http://www.dorseygraphics.com
http://www.flickr.com/photos/dorseygraphics
gdanmitchell
Veteran Member
A few things to think about here.
1. Did you sharpen in post-processing? If you shoot RAW (and you should) you'll need to sharpen, and automatic settings are not always optimum.
2. How sharp do you need? Photos that make excellent prints at, say, 12" x 18" will often show some softness at 100% magnification on the screen.
3. The point about atmospheric haze is a good one.
4. Presuming you used tripod, MLU, remote release, and turned IS off?
5. There are some common misconceptions about DOF. One is that DOF is the distance range within which "stuff is in focus." It is not so simple as that. There is a plane of optimum focus - the distance of objects that you focus on. In front of and behind this (unless you are focused on infinity) there is what might be called "a zone of acceptable fuzziness" that gets progressively less in focus the further you go from the plane of optimum focus. DOF is actually a rather subjective thing - it depends on what level of fuzziness is OK, it depends on how big you will display/print the photo, and it depends on how much you'll have to magnify the original. An important fact here could be that by focusing on a distant object and then moving the focus ring you have achieved an image in which none of your subjects is actually "in focus."
6. A long lens is less amenable to using DOF to get stuff "in focus enough."
7. Sensor size makes a difference in DOF, in several ways that include the amount of DOF you'll get at a particular aperture.
8. At some point, you simply cannot get all the settings to work together enough to get subjects at different distances to be in focus enough. For example, if you keep stopping down you may get greater DOF but you lose maximum sharpness to diffraction blur. (Arguably this happens roughly at apertures smaller than about f/8 on a 1.6x cropped sensor body and smaller than about f/16 on a full frame body.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
1. Did you sharpen in post-processing? If you shoot RAW (and you should) you'll need to sharpen, and automatic settings are not always optimum.
2. How sharp do you need? Photos that make excellent prints at, say, 12" x 18" will often show some softness at 100% magnification on the screen.
3. The point about atmospheric haze is a good one.
4. Presuming you used tripod, MLU, remote release, and turned IS off?
5. There are some common misconceptions about DOF. One is that DOF is the distance range within which "stuff is in focus." It is not so simple as that. There is a plane of optimum focus - the distance of objects that you focus on. In front of and behind this (unless you are focused on infinity) there is what might be called "a zone of acceptable fuzziness" that gets progressively less in focus the further you go from the plane of optimum focus. DOF is actually a rather subjective thing - it depends on what level of fuzziness is OK, it depends on how big you will display/print the photo, and it depends on how much you'll have to magnify the original. An important fact here could be that by focusing on a distant object and then moving the focus ring you have achieved an image in which none of your subjects is actually "in focus."
6. A long lens is less amenable to using DOF to get stuff "in focus enough."
7. Sensor size makes a difference in DOF, in several ways that include the amount of DOF you'll get at a particular aperture.
8. At some point, you simply cannot get all the settings to work together enough to get subjects at different distances to be in focus enough. For example, if you keep stopping down you may get greater DOF but you lose maximum sharpness to diffraction blur. (Arguably this happens roughly at apertures smaller than about f/8 on a 1.6x cropped sensor body and smaller than about f/16 on a full frame body.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
Steve Dorsey
Forum Enthusiast
All very good input. Responses below. . .
I understand that the hyperfocal distance for a 70/200 lens is going to be VERY different than for a wide-angle. But that's OK because the subject I was shooting was 3700 feet away, so I didn't think this would really be an issue.
I'd like to revisit this spot at sunset to get a nice silhouette of the same shot. I'm hoping I can get nice, crisp edges when I do.
----------------------
Steve Dorsey
http://www.dorseygraphics.com
http://www.flickr.com/photos/dorseygraphics
The images I posted were taken from the RAW source and had no post processing at all.
Why would I want my subject to be anything other than as sharp as it can possibly be? I understand that some settings yield softness, but I'd like to have as crisp imagery as possible.
Yup - I think this was my biggest problem.
All those things were true.
This goes to my idea about hyperfocal distance. If I understand it correctly, it allows me to have "crystal clear" focus from a certain point into infinity. I have not found this to be the case.
I understand that the hyperfocal distance for a 70/200 lens is going to be VERY different than for a wide-angle. But that's OK because the subject I was shooting was 3700 feet away, so I didn't think this would really be an issue.
I'd like to revisit this spot at sunset to get a nice silhouette of the same shot. I'm hoping I can get nice, crisp edges when I do.
Subjective, but true.
I have a full-frame sensor.
The loss in sharpness with stopping down is the lesson I'm taking away from this experience. I have to learn more about this concept.
--
----------------------
Steve Dorsey
http://www.dorseygraphics.com
http://www.flickr.com/photos/dorseygraphics
Henrik Herranen
Senior Member
This is exactly what hyperfocal distance is NOT. It allows you to have "acceptable sharpness" for small prints.
As you've found out, atmospheric haze can also be a great contributor of problems - although only with longer focal lengths. E.g. it is never a factor with a 8 mm circular fish-eye lens, but it may be prohibitive with an object shot from a long distance with a tele like a 70-200.
Exactly.
Kind regards,
- Henrik
And if a million more agree there ain't no great society
My obligatory gallery: http://www.iki.fi/leopold/Photo/Galleria/
professorgb
Leading Member
I suspect that this is part of your problem. Your expectations are too high. As Dan says, hyperfocal distance doesn't guarantee "crystal clear" or "tack sharp" or any of that. In any photo, regardless of focal length and aperture, the only part of the image that is as sharp as possible is that which falls on the focus plane (and is in the center of the image, where lens distortions are the lowest). Before that, and beyond it, your image loses sharpness. How much sharpness is lost, and how rapidly, are a function of focal length, aperture, and subject-to-focal-plane distance.
The hyperfocal distance is the focus distance at which--given a specific focal length, aperture, and sensor size--you'll get the maximum range of acceptable focus. Of course, a lot of assumptions are hidden in here about the size of pixels on your sensor, the pixel pitch, the definition of circle of confusion/cone of confusion, and other such variables. But, the point is that no photo is going to be perfectly sharp from front to back without some serious software and a series of photos of varying focus points that run from near to far in the scene you're shooting.
gdanmitchell
Veteran Member
That is a serious problem. RAW images are not going to be sharp. You simply must apply sharpening in order to obtain the sharpest images that your lens can produce.
This is not a camera, lens, or RAW format flaw. It is simply the nature of the RAW file.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
gdanmitchell
Veteran Member
Reading more carefully, I see there is more to reply to...
Second, a sharp print will not look razor sharp on the screen at 100% magnification. There are a bunch of reasons for this. For one, an image from a 12MP camera viewed at 100% magnification on screen is equivalent to a print 60 INCHES WIDE. If you make a lot of those, it may matter - and you probably want to switch to MF or LF anyway. If you don't you'll find that you get very, very sharp prints at sizes at which you actually print.
Everything else is technically LESS IN FOCUS. If your plane of focus is at, say, 20 feet from the camera, subjects at infinity are LESS focused - though likely sharp enough to make a fairly good size print if you shoot at a small aperture. (But watch out for diffraction blur if you stop down too far.)
If you refocus at infinity, your objects at 20 feet distance are now less in focus than they were before.
http://www.gdanmitchell.com/2007/04/12/sharpness-and-aperture-selection-on-full-frame-dslrs/
Using a 5D and 24-105mm f/4 L on tripod with MLU and remote release, f/8 and f/11 look virtually identical in 100% crops - and either would make very sharp prints at quite large sizes. f/16 looks ever so slightly less sharp at 100%, but this difference would almost certainly not be visible at all in an actual print.
Finally, go to a museum/gallery showing large photographs by great photographers and look at them closely. I think you'll discover that these photographs are not as "sharp" as you imagine them to be. There is a "myth of razor sharpness" that afflicts and confounds quite a few photographers today.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
As above, this is a big mistake. You must sharpen RAW images.
My first inclination is to reply with a question. If you want everything to be "as sharp as it can possibly be," why in the world do you not use sharpening?
Second, a sharp print will not look razor sharp on the screen at 100% magnification. There are a bunch of reasons for this. For one, an image from a 12MP camera viewed at 100% magnification on screen is equivalent to a print 60 INCHES WIDE. If you make a lot of those, it may matter - and you probably want to switch to MF or LF anyway. If you don't you'll find that you get very, very sharp prints at sizes at which you actually print.
It may have been "a" problem, but I think that the hyperfocal issue and, even more, the lack of sharpening are your real issues.
That's not what I wrote at all. If everything is working perfectly, there is a PLANE OF OPTIMAL FOCUS. (Notice that I did not use the word "perfect" even to describe this plane. "Perfect focus" is a dream...)
Everything else is technically LESS IN FOCUS. If your plane of focus is at, say, 20 feet from the camera, subjects at infinity are LESS focused - though likely sharp enough to make a fairly good size print if you shoot at a small aperture. (But watch out for diffraction blur if you stop down too far.)
If you refocus at infinity, your objects at 20 feet distance are now less in focus than they were before.
Not subjective at all.
At a given aperture you get narrower DOF on FF than you would on a crop sensor body, though you can safely stop down a bit more (close to two stops) before you see the effects of diffraction blur.
You might find this test interesting:
http://www.gdanmitchell.com/2007/04/12/sharpness-and-aperture-selection-on-full-frame-dslrs/
Using a 5D and 24-105mm f/4 L on tripod with MLU and remote release, f/8 and f/11 look virtually identical in 100% crops - and either would make very sharp prints at quite large sizes. f/16 looks ever so slightly less sharp at 100%, but this difference would almost certainly not be visible at all in an actual print.
Finally, go to a museum/gallery showing large photographs by great photographers and look at them closely. I think you'll discover that these photographs are not as "sharp" as you imagine them to be. There is a "myth of razor sharpness" that afflicts and confounds quite a few photographers today.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
gdanmitchell
Veteran Member
One more to illustrate three problems with your photograph.
I took your larger jpg and did a bit of work on it.
1. I applied normal sharpening. While sharpening your reduced size jpg is not going to provide the same image quality that I could obtain by sharpening the full size original, the differences seem clear.
2. Looking more closely at the photograph, it appears that your actual point of focus is in the trees some distance in front of the antenna. For whatever reason, your focus point does not appear to be on the antenna.
3. The photo also illustrates some other issues relative to the subjective impression of sharpness. For example, the dynamic range of the photo is somewhat compressed, and it looks better ("sharper") with some curves adjustments. Also, the high amounts of noise in the photo (take a look at the sky) contribute to the flat quality of the image. On top of that the color balance seems odd - although some colors (the foreground grasses, the blue mountains.) almost seem saturated, others (the sky) seem washed out.
By the way, note that even with this primitive sharpening you can now see the small runner on the trail on the upper right portion of the frame.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
I took your larger jpg and did a bit of work on it.
1. I applied normal sharpening. While sharpening your reduced size jpg is not going to provide the same image quality that I could obtain by sharpening the full size original, the differences seem clear.
2. Looking more closely at the photograph, it appears that your actual point of focus is in the trees some distance in front of the antenna. For whatever reason, your focus point does not appear to be on the antenna.
3. The photo also illustrates some other issues relative to the subjective impression of sharpness. For example, the dynamic range of the photo is somewhat compressed, and it looks better ("sharper") with some curves adjustments. Also, the high amounts of noise in the photo (take a look at the sky) contribute to the flat quality of the image. On top of that the color balance seems odd - although some colors (the foreground grasses, the blue mountains.) almost seem saturated, others (the sky) seem washed out.
By the way, note that even with this primitive sharpening you can now see the small runner on the trail on the upper right portion of the frame.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
Dave Jacobowitz
Well-known member
Steve,
Your problem is almost certainly 99% thermal. You took a picture (of the Stanford Dish, I presume?) at what looks very close to noon (shadows under trees give it away -- but the time in your EXIF data tells me that your camera's clock is not set properly!) on a day when it was 90 degrees F outside. Especially off of that dry grass, you're going to get shimmer.
Come back again in the morning, shortly after sunrise. The air will be cool and still, and you'll get much more interesting and dramatic light as a bonus. Evening might work as well, though the wind kicks up in the evening in the Stanford hills.
The next thing that is hurting you, as others have commented, is the f-stop. Higher numbers mean more DoF, but you eventually get into the realm of diffraction limiting. The sweet spot for most landscape lenses on a crop body, in my opinion, is in the rand of f/8 to f/11.
The next thing is a little sharpening of the RAW image.
But really, thermal shimmer is the big deal in that photo.
dave
--
just another obsessive amateur
Your problem is almost certainly 99% thermal. You took a picture (of the Stanford Dish, I presume?) at what looks very close to noon (shadows under trees give it away -- but the time in your EXIF data tells me that your camera's clock is not set properly!) on a day when it was 90 degrees F outside. Especially off of that dry grass, you're going to get shimmer.
Come back again in the morning, shortly after sunrise. The air will be cool and still, and you'll get much more interesting and dramatic light as a bonus. Evening might work as well, though the wind kicks up in the evening in the Stanford hills.
The next thing that is hurting you, as others have commented, is the f-stop. Higher numbers mean more DoF, but you eventually get into the realm of diffraction limiting. The sweet spot for most landscape lenses on a crop body, in my opinion, is in the rand of f/8 to f/11.
The next thing is a little sharpening of the RAW image.
But really, thermal shimmer is the big deal in that photo.
dave
--
just another obsessive amateur
Dave Miner
Forum Enthusiast
What was the temperature that day?
Rick Knepper
Forum Pro
I started getting acceptably sharp images from back to front after I scrapped the hyperfocal technique.
I read a series of articles by this fellow:
http://www.trenholm.org/hmmerk/HMArtls.html
The following link is the 2nd article in the 4 part series of articles. The rest can be found at the above link.
http://www.trenholm.org/hmmerk/SHBG02.pdf
To paraphrase, basically, focus off the Farthest point in your image (someplace in infinity) and stop down. Note the use of the term acceptable sharpness. You may be able to squeeze an x/100th of a stop better focus in the corners using the hyperfocal technique but until you master that, you can squeeze off a shot at the mountain range and stop down to f8 or smaller and take home an acceptable shot.
Not sure if the mountain in the pic is even at infinity for a 200mm lens.
I read a series of articles by this fellow:
http://www.trenholm.org/hmmerk/HMArtls.html
The following link is the 2nd article in the 4 part series of articles. The rest can be found at the above link.
http://www.trenholm.org/hmmerk/SHBG02.pdf
To paraphrase, basically, focus off the Farthest point in your image (someplace in infinity) and stop down. Note the use of the term acceptable sharpness. You may be able to squeeze an x/100th of a stop better focus in the corners using the hyperfocal technique but until you master that, you can squeeze off a shot at the mountain range and stop down to f8 or smaller and take home an acceptable shot.
Not sure if the mountain in the pic is even at infinity for a 200mm lens.
Hello.
I went out today to experiment with all that I've been reading about
hyperfocal distance and taking crisp shots of landscapes.
I'm using a Canon 5D with a 70/200 EF lens. In this example, I was
shooting a radio telescope (a large dish) that was about 3700 feet
from my location. Plenty of distance to shoot using hyperfocal
distance according to the article I read.
I was never able to get any focus no matter how I tried. Hyperfocal
distance tutorials will tell you to select your focal distance and
then re-focus your camera to a distance that is somewhat closer to
you. If you are using a small aperture (higher f-stop) then your
depth of field will (effectively) exist from 1/2 your focal point to
infinity.
I'm pretty sure that I'm understanding this properly.
It didn't really matter what f-stop and ISO I was using, my pictures
came out fuzzy. No clear focus at all.
Here's an example of a series of images I took as samples to compare
the best settings to use. They are unaltered and shown at 100%. The
first is the whole image (I reduced its size to fit) and the second
is samples of a part fo that image taken at various f-stops and
speeds.
Any ideas here? How can I shoot super-focused crisp images?
Thanks,
------------------S
----------------------
Steve Dorsey
http://www.dorseygraphics.com
http://www.flickr.com/photos/dorseygraphics
Vernon D Rainwater
Forum Pro
Steve, I don't have provisions to display an adjusted image, however; the primary issue with your image is that it can be greatly improved by applying proper sharpening that helps to reduce the "haze" and to enhance both sharpness and contrast.
I did a test sharpening with the image and find that it accepts rather heavy sharpening and does not seem to show any ill effects from the process.
In fact, I went somewhat "heavy" with sharpening and it (to me) looks much better.
Suggest you try this: I am using Photoshop 7 so i used this method. Select the highlights using Ctrl+Alt+shift+~ to select highlights -- then, invert to select the shadows (darker) parts of the image and apply USM 30-30-0 with this setting.
Next, use Ctrl+D to release the selection of shadows and again apply USM to the entire image using 30-30-0 and see what you think of th image after this combined sharpening. This method helps to NOT burn out highlights by applying a higher(one time) USM setting to the entire image.
As discussed, there is "obviously" atmospheric effects in the image.
--
Vernon...
gdanmitchell
Veteran Member
If you have CS3, after converting from RAW convert your photo into a smart layer.
Select smart filter and try settings of 150 and 1.0.
Select Unsharp Mask (USM) and try 15, 50, and 1.
These are very generic settings, but they are a decent starting point with full size files.
In general, a very slight s-curve that drops mid-dark levels and raises mid-light levels via curves layer can also often be effective with flat images.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
Select smart filter and try settings of 150 and 1.0.
Select Unsharp Mask (USM) and try 15, 50, and 1.
These are very generic settings, but they are a decent starting point with full size files.
In general, a very slight s-curve that drops mid-dark levels and raises mid-light levels via curves layer can also often be effective with flat images.
Dan
--
---
G Dan Mitchell
SF Bay Area
Blog: http://www.gdanmitchell.com/
Gallery: http://www.gdanmitchell.com/wpg2-3/
rhlpetrus
Forum Pro
I've never seen anyone using hyperfocal distance with a telephoto lens ...
1) It's for WA lenses mostly. Teles have less DoF, so HF distance is farther away, moreover you pick up haze and thermal oscilations of atmosphere, causing image to look not that sharp, even at precise focus point.
2) DoF and HF distances ar computed assuming you'll view it enlarged to 8x12" (2:3 ratio, originally computed for 8x10" view-camera format) from about 2ft = 60cm distance. If you use 100%, image is about 40" wide or more (depends on sensor and monitor resolution), read on about that.
3) For a fixed sensor size, there's the largest disk (diameter) on sensor which, viewed as above (2), shows as a point to our eyes. That's the circle of confusion used in calculations. For APS-C it's about 0.020mm, for FF it's 0.030mm, APS-H's in between. That defines how much OOF a point in an image can go before it is seen as OOF.
4) There's no guessing, use DOFmasters calculator to compute HF distance for your lens and aperture, then use that info in the field:
http://www.dofmaster.com/dofjs.html
--
Regards, Renato.
http://www.flickr.com/photos/11435304@N04
OnExposure member
http://www.onexposure.net/
1) It's for WA lenses mostly. Teles have less DoF, so HF distance is farther away, moreover you pick up haze and thermal oscilations of atmosphere, causing image to look not that sharp, even at precise focus point.
2) DoF and HF distances ar computed assuming you'll view it enlarged to 8x12" (2:3 ratio, originally computed for 8x10" view-camera format) from about 2ft = 60cm distance. If you use 100%, image is about 40" wide or more (depends on sensor and monitor resolution), read on about that.
3) For a fixed sensor size, there's the largest disk (diameter) on sensor which, viewed as above (2), shows as a point to our eyes. That's the circle of confusion used in calculations. For APS-C it's about 0.020mm, for FF it's 0.030mm, APS-H's in between. That defines how much OOF a point in an image can go before it is seen as OOF.
4) There's no guessing, use DOFmasters calculator to compute HF distance for your lens and aperture, then use that info in the field:
http://www.dofmaster.com/dofjs.html
--
Regards, Renato.
http://www.flickr.com/photos/11435304@N04
OnExposure member
http://www.onexposure.net/
Keyboard shortcuts
- f
- Forum