Reading mode: Light Dark

Sharpest Focal Length vs Crop

Go to last post
J

Jack Hogan

Veteran Member
Messages
8,872
Solutions
17
Reaction score
4,220
Location
Toronto, CA
Skill testing query for you skilled photographers:

I am walking around casually with my D90 and 18-105 kit lens in aperture priority mode at f/5.6. Suddenly a great subject appears in the distance. I zoom in all the way and at 105mm I frame it perfectly.

But wait! The 18-105 is at its sharpest at wider focal lengths (see for instance here http://www.lenstip.com/182.4-Lens_review-Nikon_Nikkor_AF-S_DX_18-105_mm_f_3.5-5.6_VR_ED_Image_resolution.html ). Assuming that I would crop the image in post processing to look like the 105mm frame that I saw through the lens originally, and I only needed to display the image at a resolution of 1430x950 pixels (1/3 the resolution of the d90), what focal length would I have to shoot at to get the sharpest picture of my subject?

In other words, is it better to zoom in and get a higher resolution but blurrier image, or zoom out and get a sharper lower resolution one?
 
Zoom in.

There are numerous problems that you haven't thought of by taking a picture at 18mm and cropping it to simulate a 105mm photo:

1) It's a DSLR. It has a nice bit of depth-of-field, so you better pray that little pinpoint of a subject is exactly what your camera focused on if you are going to crop and enlarge.

2) Metering. The desired subject may be over or underexposed since at 18mm, the camera has to take a lot of things in the scene in consideration when metering.

--
http://www.flickr.com/photos/techoutsider
 
Extending the actual focal length nearly 6 times by cropping is never a good idea. I seldom use the 3x digital zoom provided by my camera - it produces crunchy results.

Don't worry, the 18-105 is an excellent lens and feel free to zoom to your hearts content :D. I've used it before too ;).
--
http://www.flickr.com/photos/techoutsider
 
Jack Hogan said:
Skill testing query for you skilled photographers:

I am walking around casually with my D90 and 18-105 kit lens in aperture priority mode at f/5.6. Suddenly a great subject appears in the distance. I zoom in all the way and at 105mm I frame it perfectly.

But wait! The 18-105 is at its sharpest at wider focal lengths (see for instance here http://www.lenstip.com/182.4-Lens_review-Nikon_Nikkor_AF-S_DX_18-105_mm_f_3.5-5.6_VR_ED_Image_resolution.html ). Assuming that I would crop the image in post processing to look like the 105mm frame that I saw through the lens originally, and I only needed to display the image at a resolution of 1430x950 pixels (1/3 the resolution of the d90), what focal length would I have to shoot at to get the sharpest picture of my subject?

In other words, is it better to zoom in and get a higher resolution but blurrier image, or zoom out and get a sharper lower resolution one?
Click to expand...
Your post implies that at full zoom, the lens reduces everything to a blur... :)

If this is true, I suggest you immediately invest in another lens.

Strangely enough, I've been told that for a kit lens, this one is very good.

Hmm?

I do believe that a tiny difference in quality by shooting at full magnification more than trumps the big loss of shooting at the lowest magnisfication... :)

And 1430 is not exactly a "thumbnail."

Dave
 
Jack Hogan said:
Skill testing query for you skilled photographers:
. . .

In other words, is it better to zoom in and get a higher resolution but blurrier image, or zoom out and get a sharper lower resolution one?
Click to expand...
It doesn't take a skilled photographer to know the answer to your question, and if you do a simple test you're sure to agree that zooming in produces far better images. Try this. Take a photo with the subject framed at 105mm, which due to being the maximum focal length, probably doesn't produce the 18-105mm's most detailed images. It might help if the image contains some text written in a very small font that might make it difficult to read later on. Then shoot one more photo from the same distance, either with the lens zoomed out to 35mm, or use an expensive, higher quality 35mm prime lens if you have one. Take that photo and compare the two on your computer's monitor, but you should be able to even see in the D90's LCD display that the image shot with the lens zoomed to 105mm shows much more detail and has much greater overall image quality. It just doesn't provide as much coverage as photos shot at 35mm. If done in moderation, cropping can be useful, but it is guaranteed to reduce image quality enough to be noticed unless you print very small or create small photos resized for web use.
 
Mr Roygbiv rainbow said:
You said you framed the subject perfectley at 105mm , why would you want to crop it ? Is it april the first today ?
--
Roygbiv
Click to expand...
The Zodiac signs have been rewritten, so from an, err, Astrological point of view, it may very well be April 1st.... :(

Dave
 
Here's an example. This was quick, so I didn't do much to meter everything. It's the same daisies at 28mm and 80mm (min max of my only zoom). I blew the exposure on the 28mm shot and darkened it in post. I think it's still pretty clear that the 80mm shot is sharper and clearer than the 28mm one. I'll swap it for the blown highlights original if someone wants to see the difference.

100% crops, the focus point for both was the center of this daisy.

80mm zoomed in:





28mm cropped to 80mm FOV:



 
I would like to raise a scenario related to the OP. Consider shooting macro where DOF is critical (and usually insufficient). Assume one needs 2.4MP to view on a monitor. I have a 12 MP sensor. If, instead of filling the frame, I step back and only fill half the frame (linearly) so that my subject is 3MP instead of 12MP. However, my DOF is significantly larger for the 3MP shot. I believe the 3MP photo of a 3D "bug" will look best (more detail because of better focus). We regularly discuss the excellent macro capabilities of P&S cams. I believe this is because of the larger DOF offered by small sensor cameras. If one had 48MP APS-C sensor and appropriate lens, the macro DOF would be further increased.
Am I out in left field here? Comments and enlightenment welcome.
Bert

PS: I agree with the other posts that optical zooming beats digital zooming for normal non-macro shooting. (which is why I do NOT own the otherwise excellent Panasonic LX5)
 
Good question Jack, you are a tease!

Although I can't be bothered to answer your original scenario directly, I have often wondered the same about whether to shoot on my cheapo (but respectable) Tamron 70-300 at 300mm (bit soft), or cut back a bit to 180mm, where its reputation for good optics remains intact.

Your question inspired me to find out. Again, I'm too lazy to actually post results, but if you really need 300mm, then I agree with others that its best to shoot at that rather than enlarge. However, if you don't require the full resolution of your camera, as you suggested in your original post (in my case 6MP rather than 12MP) then there is no difference in sharpness.

The only real world difference is the change in depth of field, which can be a positive or negative depending upon your artistic vision/desperation to have some. IMHO it's would be worth adding this level of control to the toolbox.

cheers
Flakey
--
flakey
 
There are so many of these types of questions, I can't stand not replying any longer. Sorry to single you out.

Stop asking, go out and take some photos and compare them. I say this because that is how I would answer this question. And I would have the correct answer long before waitng to hear other people's guesses in a forum.

Why rely on other's guesses when you can take two simple photos and compare them and decide for yourself which way is better? And realize that depending on the situation it may be best to take both shots if you have time or bracket your shots with sequential shooting and choose the best shots when you see them on your monitor.

You can't take two photos with different adjustments and compare them? I just don't get it.
 
BertIverson said:
I would like to raise a scenario related to the OP. Consider shooting macro where DOF is critical (and usually insufficient). Assume one needs 2.4MP to view on a monitor. I have a 12 MP sensor. If, instead of filling the frame, I step back and only fill half the frame (linearly) so that my subject is 3MP instead of 12MP. However, my DOF is significantly larger for the 3MP shot. I believe the 3MP photo of a 3D "bug" will look best (more detail because of better focus). We regularly discuss the excellent macro capabilities of P&S cams. I believe this is because of the larger DOF offered by small sensor cameras. If one had 48MP APS-C sensor and appropriate lens, the macro DOF would be further increased.
Click to expand...
BertIverson said:
Am I out in left field here? Comments and enlightenment welcome.
Click to expand...
Moving back does not increase DoF, if you subsequently cancel the increase by enlarging the image. DoF is always image viewing scale dependant, and if the scales are the same, the DoF will be the same. Moving back and then cropping in will leave you with less image quality, of course, because there are fewer pixels describing the image.

Whether the loss is tolerable, or not, will depend more on your personal tolerance levels than anything else. It is quiet possible it will be... a 2.4 MP monitor is quite a big one, and normal viewing would be at some distance from that, which will naturally reduce scale.

However, the experiment you describe is an easy one to set up. Why don't you carry it out for yourself and let us know how it goes?
--
Regards,
Baz

"Ahh... But the thing is, they were not just ORDINARY time travellers!"
 
flakin stephen said:
The only real world difference is the change in depth of field, which can be a positive or negative depending upon your artistic vision/desperation to have some. IMHO it's would be worth adding this level of control to the toolbox.
Click to expand...
I suggest you ask yourself if there really is a difference in DoF.

Remember, if you enlarge an image by cropping you are magnifying it, which means you are also magnifying any blur in it.
--
Regards,
Baz

"Ahh... But the thing is, they were not just ORDINARY time travellers!"
 
I don't understand why anyone buys a poor performing lens and doesn't sell it right away. There are lots of affordable lenses out there. When I buy a lens that fails to live up to my expectations it goes straight onto Gumtree, if it fails to sell there it goes onto ebay.
 
Barrie Davis said:
I suggest you ask yourself if there really is a difference in DoF.
Click to expand...
Yes, I'm with Bert (see earlier post).

Same viewpoint, both f11
firstly, 300mm scaled down to match the second shot





70mm, 100 % crop from original





For artistic reasons I prefer the first, but there's plenty of times when you run out of depth of field. It's no different to using a smaller sensor.

Always happy to educate!

cheers
Flakey
--
flakey
 
flakin stephen said:
Barrie Davis said:
I suggest you ask yourself if there really is a difference in DoF.
Click to expand...
Yes, I'm with Bert (see earlier post).

Same viewpoint, both f11
firstly, 300mm scaled down to match the second shot





70mm, 100 % crop from original





For artistic reasons I prefer the first, but there's plenty of times when you run out of depth of field. It's no different to using a smaller sensor.

Always happy to educate!
Click to expand...
You state that both pictures are from the same viewpoint, so you didn't rescale by moving.
--
Regards,
Baz

"Ahh... But the thing is, they were not just ORDINARY time travellers!"
 
You must log in or register to reply here.

Similar threads

C Sean
Bringing back old primes and other focal lengths!
2 3
Replies
47
Views
4K
Gabe H
G
Satyaa
Focal length vs. resolution
2 3
Replies
42
Views
4K
MrBrightSide
MrBrightSide
U
Zoom vs Prime Focal Length Playing with Lens Comparisons.
Replies
1
Views
3K
Ian Stuart Forsyth
Ian Stuart Forsyth
S
Cropping in post to mimic different focal length.
Replies
13
Views
6K
John Sheehy
John Sheehy
M
Canon 100-500 vs 200-800 vs rf 100-400 vs 600 f11
2
Replies
21
Views
6K
R2D2
R2D2

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