1
\$\begingroup\$

This article argues that,

50mm prime lenses, on the other hand, have a maximum aperture of somewhere between f1.2 and f2, letting you hand hold photos in much lower light levels (they also give you a much brighter viewfinder, making it easier to see the subject in low light)

For example, this landscape was taken with a 50mm prime lens by the light of the setting sun. The light was very low and I wouldn't have been able to hand-hold the camera if the maximum aperture was limited to f4 or f5.6.

enter image description here

Generally, while shooting landscapes, it is expected to have everything in sharp focus. It means that we need to use to use a higher f-stop to increase DoF and reduce blurring thereby. However, the author advocates using 50mm at a lower f-stop just to let more light falling on to the sensor in low-light. The image above doesn't exhibit any lack of sharpness apparently. But, I wonder if it is really the case or because of the scaled down version of the original image. Will the lack of sharpness become evident for a full-size, high resolution image?

\$\endgroup\$
2
  • \$\begingroup\$ Remember that most wide aperture prime lenses won't perform at their best when they're at their widest aperture. Usually to get the sharpest results, you need to be at perhaps 2-3 stops smaller than the lens's widest aperture \$\endgroup\$ Commented Aug 12, 2016 at 13:12
  • 1
    \$\begingroup\$ @laurencemadill On the other hand, you can stop an f/1.4 lens down two full stops and still use a shutter time half as long as with an f/4 lens used wide open. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 17:10

5 Answers 5

1
\$\begingroup\$

Does shooting landscape at lower f-stop produce reasonably sharp images?

Even at very wide apertures when the nearest part of the scene within the field of view is several hundred feet away, as is the case in your example photo, focusing at or beyond the hyperfocal distance can be used to keep everything in acceptable focus. If the nearest object in the scene is further away than half the hyperfocal distance then you would focus somewhere just short of twice the distance to the nearest object.

Will the lack of sharpness become evident for a full-size, high resolution image?

Just because the focal length, aperture used, and the shooting distance may allow for most everything to be within the limits of acceptable focus doesn't guarantee a sharp image will be the result. In such a case as described in the question the largest contributing factor to blur is likely not the size of the circle of confusion or the resulting depth of field or even failing to use the lens' "sweet spot" at all. Rather it is probably the amount of camera movement allowed by the photographer holding the camera by hand.

So in the case of the article cited in the question the author is balancing the need for a short enough shutter speed to prevent unacceptable camera motion blur with the need for an aperture that allows for acceptable sharpness.

Look at it this way: You can have a 50mm f/1.4 lens that is theoretically perfect at f/8 and a film that is a theoretically perfect recording medium at ISO 25 but if you're handholding the camera and only have enough light to use a shutter time of 1/4 second with that combination you're not likely to wind up with an image that is acceptably sharp by anyone's definition. If you use f/2 and 1/60 second you'll almost certainly wind up with a much sharper image that might be acceptably sharp to you.

Since more detailed information is not given, we'll have to guess at a few things. Let's assume the photo above was captured on 35mm film by a 50mm f/1.4 lens. If the photographer used f/2 that means the lens was stopped down a full stop from wide open, yet it still allowed a shutter time 1/4 as long as an f/4 lens used wide open would have required and a shutter time 1/16 as long as an aperture setting of f/8 would have required for the same film speed and lighting conditions.

At 50mm and f/2 using 35mm film to produce an 8x10 print viewed at 10 inches by a person with 20/20 vision (which gives us the CoC of 0.03mm) the hyperfocal distance is 136 feet. That means everything from 68 feet to infinity would be acceptably sharp. If the nearest point in the vista is further than 68 feet, then you would focus at slightly less than twice the distance to the nearest object appearing in the frame to maximize the depth of field of the entire vista.

enter image description here

\$\endgroup\$
9
  • 1
    \$\begingroup\$ @BrandonDube care to explain why they are inaccurate? \$\endgroup\$
    – kazanaki
    Commented Aug 12, 2016 at 14:39
  • \$\begingroup\$ @kazanaki blog.retrorefractions.com/?p=4 \$\endgroup\$ Commented Aug 12, 2016 at 15:06
  • 2
    \$\begingroup\$ @BrandonDube They're certainly not so wildly inaccurate that they don't illustrate the point, which is that distance to subject is a significant factor in DoF, and that if the distance to subject is large enough, everything can be reasonably sharp even at f/2. \$\endgroup\$
    – Caleb
    Commented Aug 12, 2016 at 15:20
  • 1
    \$\begingroup\$ @BrandonDube The viewing conditions are specifically laid out in the answer. In the case of the photo in question, 3, 6, or 400 feet makes no difference: everything in the photo is a greater distance than that from the camera. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 15:43
  • 1
    \$\begingroup\$ @AlanMarcus But still useful for comparison when only one of the variables is changed (i.e. different aperture with the same focal length, focus distance, and circle of confusion). Also very useful for demonstrating the theory of hyperfocal distance even if not always as accurate predicting it. Understanding the criteria for using a specific circle of confusion is critical, which leads directly to understanding the effect of display size, viewing distance, and even the visual acuity of the observer on the calculation of the CoC. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 16:58
0
\$\begingroup\$

As to the viewfinder brightness: The appeal of the SLR (single lens reflex) is due in no small part to its viewfinder system. The modern SLR maintains the lens aperture wide-open for viewing and composing. Thus the photographer is presented with a view that is at maximum brightness. Additionally, with the aperture open wide, depth-of-field (DOF) is at minimum. A shallow DOF is desirable for focusing as focusing is more critical. During the actual picture taking, the mechanism closes down the aperture to a pre-set f/number. This action expands the zone of DOF. Naturally, the larger the maximum aperture, the brighter the viewfinder view.

Let me add that landscape images likely only portray objects at an infinite distance. We set the focus at infinity ∞. Unless these vistas contain foreground objects, DOF is moot.

All that being said, there is truth and there is lore when it comes to lens sharpness. The old masters of landscape called themselves “Group 64”. They set their aperture to f/64 to gain maximum DOF and maximum acuity. The DOF part is correct, the acuity part is iffy.

A a rule-of-thumb: The camera lens is at maximum sharpness when stopped down about two f/stops. This is because at max aperture the outer perimeter of the lens is being used, this region has the most severe curve (figure), imaging forming rays from this region often go astray reducing acuity.

Thus as we stop down, acuity increases. Not so fast! As we stop down, twin demons of diffraction and interference begin to degrade the image. Due to the nature of light, some of rays hit the blades of the iris with a glancing blow. These ricochet and bleed into the path of the image forming rays. As we stop down the percentage of these glancing blows increase. The net result, the resolving power of the lens decreases as we stop down. We measure lens acuity in lines per millimeter. We are talking about the ability to distinguish spaces between closely ruled lines.

The Rayleigh Criterion conveys the theoretical resolving power of a lens as RP=1392 ÷ f/number. This value is usually given for green light; other colors will have a slightly different value.

Table of RP lines per millimeter vs. f/number: f/1 = 1392

f/1.4 = 994

f/2 = 696

f/4 = 348

f/8 = 174

f/16 = 87

f/32 = 44

f/64 = 22

Note: f/8 value and above is higher than is pictorially useful for film.

How does this affect the final image? Generally we view small images and likely the RP is of little importance. However, when we make and view large images for display, this stuff becomes important. Especially true if the viewer is positioned near the enlarged displayed image.

\$\endgroup\$
2
  • \$\begingroup\$ Despite the remark in the quote, I don't think the OP meant to ask anything about viewfinder brightness. \$\endgroup\$
    – Caleb
    Commented Aug 12, 2016 at 17:50
  • \$\begingroup\$ @Caleb Since the question doesn't specify, if one is manually focussing the viewfinder brightness would be rather significant. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 18:07
0
\$\begingroup\$

However, the author advocates using 50mm at a lower f-stop just to let more light falling on to the sensor in low-light.

Photography is almost always a compromise. Choose parameters that allow for pleasing depth of field and short enough exposure time. Fast lenses give you more options here.

And no, landscape does not have to be sharp everywhere.

The image above doesn't exhibit any lack of sharpness apparently. But, I wonder if it is really the case or because of the scaled down version of the original image. Will the lack of sharpness become evident for a full-size, high resolution image?

Even if you insist on Group f/64 style sharpness in the whole image area, I would not be worried about DOF too much with near infinity image with no foreground. There are other factors, though. Most fast 50 mm lenses suffer from spherical aberration, field curvature and other aberrations that reduce resolution wide open. If you want to achieve the best possible resolution, it is important to know your lens and find the right balance with the aperture/ISO/exposure setting.

One more thing - I don't know how old this article is, but some of these images look like shot on film and compared to film, modern cameras can achieve shorter exposures thanks to better high ISO image quality and presence of image stabilizer. These things together can give you possibly 5 stops advantage compared to film and no image stabilization. So there are other lightweight options for landscape photography than 50mm/1.4 lens today. Shooting with 50mm is still fun, though.

\$\endgroup\$
10
  • \$\begingroup\$ My 50mm f/1.4 is the lightest lens in my bag. The only thing that would conceivably be lighter would be a plastic 50mm f/1.8 or a pancake lens. Wider angle lenses made of the same materials tend to be heavier due to the necessity of their retro-focal design unless designed for use on a camera with a very short registration distance. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 18:15
  • \$\begingroup\$ I would still consider lenses like the 35mm IS from Canon lightweight. \$\endgroup\$
    – MirekE
    Commented Aug 12, 2016 at 18:21
  • \$\begingroup\$ I said "there are other lightweight options" in my post. Not "lighter". \$\endgroup\$
    – MirekE
    Commented Aug 12, 2016 at 18:36
  • \$\begingroup\$ It is lightweight at 11.8 oz., but it's not lighter than the EF 50mm f/1.4 at 10.2 oz. It is also one stop slower. The 35mm f/1.4 lenses are 20.5 oz. and 26.8 oz. respectively. the-digital-picture.com/Reviews/… \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 18:36
  • 1
    \$\begingroup\$ Yeah, I misread that as "lighter". Still, in the context of the article, we're talking 35mm film for all of the landscape images from Argentina. You can't just crank up the ISO with low speed film. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 18:38
0
\$\begingroup\$

However, the author advocates using 50mm at a lower f-stop just to let more light falling on to the sensor in low-light.

That's a pretty big just. A lens that gathers more light does, in fact, let you shoot at a higher shutter speed, which means that you can hand-hold the camera in lower light than you could with a slower lens. This is a fair point. On the other hand, many kit lenses have image stabilization that makes it possible to hand-hold the camera at a slower shutter speed than you could with a non-stabilized lens.

The image above doesn't exhibit any lack of sharpness apparently. But, I wonder if it is really the case or because of the scaled down version of the original image. Will the lack of sharpness become evident for a full-size, high resolution image?

Even the example image doesn't seem tremendously sharp; it may be that some sharpness was lost in scaling down or compressing the image. In general, though, the thing to know is that depth of field increases as the distance to the subject increases. When the distance to the subject is great, the depth of field will be very large even at large apertures. Is it possible that objects in the distance will be noticeably less sharp? Of course. But depending on the circumstances, it's also possible that the decrease in sharpness won't be detectable.

Overall, I think you may be reading too much into a single point in a larger article about the advantages of 50mm prime lenses. Yes, you can shoot a landscape at large aperture, and that may give you more flexibility than you'd have with a slower lens. Other advantages include low price, image quality, narrow DoF, and low weight.

\$\endgroup\$
-2
\$\begingroup\$

Every lens has a "sweet spot" for sharpness. It's usually in the middle of the range of f-stops (f8 to f11, generally). As you move further away from that sweet spot, detail becomes more blurred, as shown in the image below. Whether this reduction in sharpness is perceivable depends mostly on viewing distance relative to image size. You will absolutely see the difference at close range. At longer range, the lack of sharpness becomes less noticeable.

Here's an article that describes the phenomenon, and how to find your lenses' sweet spots.

Sweet spot example from Digital Photography School article linked below

\$\endgroup\$
12
  • 1
    \$\begingroup\$ How does your answer apply to the specific use case outlined in the question? That is, a landscape photo taken at large distance from the nearest items within the field of view? \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 16:51
  • \$\begingroup\$ The "sweet spot" will give the sharpest results possible no matter what the distance from lens to subject. I was answering the question "Will the lack of sharpness become evident for a full-size, high resolution image?" Depending on viewing distance, the lack of sharpness will either be noticeable, or not. Any landscape taken from long distance (like the example image the OP used) at the sweet spot of the lens will have plenty of depth of field to keep everything in frame in focus. \$\endgroup\$
    – digijim
    Commented Aug 12, 2016 at 17:09
  • \$\begingroup\$ But you've totally left out the two most important parameters of the question! The need for a wide aperture due to limited light and a handheld camera and the effect of far greater than hyperfocal distances on the depth of field. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 17:14
  • 2
    \$\begingroup\$ This would be a more useful answer if the second image was shot also at f/3.5, but from a greater distance. Then the OP would see that the plants and trellis behind the clock would be nearly as sharp as the clock, illustrating the point that OP was actually asking about. \$\endgroup\$
    – Caleb
    Commented Aug 12, 2016 at 17:46
  • 1
    \$\begingroup\$ The question primarily asks, "Does shooting landscape at lower f-stop produce reasonably sharp images?" You're obsessing on one aspect of the expanded description to the exclusion of the main thesis of the actual question. \$\endgroup\$
    – Michael C
    Commented Aug 12, 2016 at 20:17

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.