Ad
Ad
Ad
Pages: [1] 2 3 »   Bottom of Page
Print
Author Topic: Why do people oversharpen?  (Read 11549 times)
MrSmith
Sr. Member
****
Offline Offline

Posts: 917



WWW
« on: April 13, 2013, 08:49:35 AM »
ReplyReply

I have noticed this trend recently of over sharpening to the point of nasty artifacts and smooth areas taking on an orange peel effect. Why do people do this? Particularly amateurs with images that only get viewed on the web.
Can they not see what they are doing to their images?

I do a small amount of sharpening at output* depending on lens used then leave it up to the client to sharpen for their intended use as sharpening for print will mean the image is too gritty for small web use.

*usually just the normal/soft look from the old capture one in C1-7
Logged
PeterAit
Sr. Member
****
Offline Offline

Posts: 1961



WWW
« Reply #1 on: April 13, 2013, 09:15:44 AM »
ReplyReply

I suspect it's two things  -  at least. One is the overemphasis on sharpness in photographs, the "sharper is always better" mantra that, while widely believed, is of course false. Then there are less knowledgeable photographers who don't understand that sharpening can be overdone to the point of the kinds of artifacts you describe. "Hey, if LR lets me slide the sharpness control all the way over, it must be OK!"
Logged

Peter
"Photographic technique is a means to an end, never the end itself."
View my photos at http://www.peteraitken.com
hjulenissen
Sr. Member
****
Offline Offline

Posts: 1683


« Reply #2 on: April 13, 2013, 11:32:17 AM »
ReplyReply

perhaps their camera or some web-based scaling does this by default?
Logged
Schewe
Sr. Member
****
Online Online

Posts: 5512


WWW
« Reply #3 on: April 13, 2013, 12:52:25 PM »
ReplyReply

I have noticed this trend recently of over sharpening to the point of nasty artifacts and smooth areas taking on an orange peel effect. Why do people do this?

Because they don't have a clue how to sharpening properly...
Logged
petermfiore
Sr. Member
****
Offline Offline

Posts: 584



WWW
« Reply #4 on: April 13, 2013, 12:58:40 PM »
ReplyReply

Because they don't have a clue how to sharpening properly...

And that's the truth.


Peter
Logged

PhotoEcosse
Sr. Member
****
Offline Offline

Posts: 640



« Reply #5 on: April 13, 2013, 02:16:04 PM »
ReplyReply

Given that you suggest that the culprits are those whose images you say "only get used on the web", then that may be a large part of the answer. We all know that no normal computer monitor is capable of displaying a photograph well (in any meaningful sense) but I suspect that you are correct that many photographs never get printed.

For those that do, however, I further suspect that many of the images that we see on web pages and in chatrooms like this are Jpeg versions of photographs that may have been correctly sharpened for the production of a large print, but then compressed for the web in a way that caricatures the sharpening.

Certainly I agree that many of the images that get posted here do look grossly over-sharpened and also over-saturated. But that may not be how they look when outputted to print as the authors intended.
Logged

************************************
"Reality is an illusion caused by lack of alcohol."
Alternatively, "Life begins at the far end of your comfort zone."
digitaldog
Sr. Member
****
Offline Offline

Posts: 9192



WWW
« Reply #6 on: April 13, 2013, 02:20:52 PM »
ReplyReply

If you over-sharpen an image visually, then back off, the result is, the image looks less sharp. That fools the brain. I think the trend is to go the other way hence over sharpening. But the bigger issue is sharpening visually for any output other than to the screen is fraught with issues! What looks butt ugly on-screen can look fantastic on output. WYS isn't WYG.

http://www.creativepro.com/story/feature/20357.html
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3805


« Reply #7 on: April 13, 2013, 06:01:47 PM »
ReplyReply

I have noticed this trend recently of over sharpening to the point of nasty artifacts and smooth areas taking on an orange peel effect. Why do people do this?

Two reasons;
  • 1. Because they can.
  • 2. Because the sharpening tool doesn't warn them when they go too far.

I'm not trying to be a Smart Aleck here, this is serious. Nothing is stopping them, so why would they?
The follow-up question then becomes; If they knew when to stop, would they?

Quote
Can they not see what they are doing to their images?

Actually, that's part of the issue. When is enough, enough? The sharpening tool will not warn you, not how much more is needed, and not how much it is overdone. It could, but it doesn't. So we'll have to do it by eye, and we're really not that good in pinpointing the optimum setting by eye. One can gain experience by trying, but maybe that is not such an efficient method to find out.

Let's address the issue in an analytical way.

The maximum possible sharpness, without artifacts, is reached when two adjacent pixels (or an edge) jump(s) exactly from the minimum possible brightness to the maximum possible brightness, e.g. from 0 to 255 in an 8-bit channel. That assumes the original subject was such a sharp edge, with maximum contrast. No smooth transition, just an abrupt jump. That is only possible when these pixels or edges happen to be aligned exactly with the pixel grid. If we were to shift the image half a pixel, then the edge transition of brightness would go from 0% to 50% to 100%. That's consistent with the Shannon/Nyquist principle, that it takes more than 2 pixels to reliably reconstruct 1 cycle (~line-pair). Therefore the brightness of pixels on a slanted edge would transition from 0% to 100% depending on the contribution ratio of both pixels across the edge.

Natural images (taken with lenses, area averaging sensels, and demosaicing) will not be able to make such a steep transition, there will be a very slightly gradual transition. That transition (AKA Edge Spread Function or ESF) is at best equal to a Gaussian blur with radius 0.323, as plotted in the following chart:
 

Now, when we know that the theoretically best possible artifact free sharpness corresponds to a Gaussian blur of 0.323, then we can determine the maximum possible contrast between any pixel and its neighbor. For that we could take a single white pixel, say RGB[255,255,255], on a black background RGB[0,0,0], and blur (convolve) that with a 0.323 Gaussian blur (Point Spread Function, PSF).

That will set the central pixel to 196.7591, and its brightest neighbors to 13.6177, or a ratio of 14.45:1 for a horizontal / vertical neighbor. That is therefore the highest possible per pixel contrast, anything more is a signal that it is over-sharpened. That is for a perfectly aligned per pixel contrast, which will be halved when the edge falls exactly halfway two pixels. Therefore the optimal sharpness-zone is between a per pixel contrast of 7.23:1 and 14.45:1, but never more than the latter.

That should make it relatively easy for a modern software producer to create a signal zone in their sharpening dialog that indicates optimal/over sharpening. Unfortunately they instead seem to be hibernating (or worse).

For printed output, one might want to over-sharpen a bit at the native printing resolution, to pre-compensate for diffusion losses in the output medium and/or due to the specific printing technology.

Cheers,
Bart
« Last Edit: April 13, 2013, 07:12:37 PM by BartvanderWolf » Logged
louoates
Sr. Member
****
Offline Offline

Posts: 781



WWW
« Reply #8 on: April 13, 2013, 07:06:00 PM »
ReplyReply

I agree that most folks use the wrong tools for sharpening or have no idea how to use the tools they have. Lots of art show photographers have been pushed into bigger and bigger canvas prints to keep up with the competition and thus are using way too much enlarging for the pixels available. Rather than learning how to up-res properly they try the sharpening route with predictably poor results.

You can't sharpen correctly for your output printing by using your monitor. Here's the foolproof way: Nik Sharpening tools. Use Nik Presharpener for screen viewing. That's where you can use your monitor to judge what it will look like on someone else's monitor. But if you are going to print it, use Nik's Output sharpener. Use their defaults for whatever paper or canvas you will be using. I guarantee it'll look like crap on your monitor. Print it anyway. If you're like me, you'll throw away your previous prints. Yes, it's that good. Yes, it's that easy. No, I don't work for Nik or Google. Yes, there are other sharpening tools out there but I see no need for me to bother with testing them.
  It took me a few months to get the courage to print my first Nik Output sharpened image despite how lousy it looked on my Eizo monitor.
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8917


« Reply #9 on: April 13, 2013, 07:54:22 PM »
ReplyReply

Mrs Smith,

The question can't be satisfactorily answered without reference to specific images. People post images for different reasons. Some post images of themselves to declare that they are, or were, at a particular location. "Look at me! Look at me! Aren't I cool!"

For such images, 'pop' is more desired than appropriate sharpening, and the occasional blown highlight and/or blocked shadow is of no concern.

Others post images to demonstrate how sharp their new lens or camera can be,  but without a comparison of the identical scene shot with another lens or camera with an equally appropriate amount of sharpening applied, showing 100% crops, no firm conclusions can be reached about the resolving merits of the new lens or camera. An image from a P&S camera can look as as sharp as the same scene shot with 40mp MFDB, at a particular print size.

If one is exploring the potential of one particular aspect of lens performance, such as resolving power, then it should be no surprise that one would attempt to extract the maximum amount of detail from any test images, which of course requires serious sharpening. If it so happens that other parts of the image which contain little or no detail, and which should be smooth, are adversely affected by the sharpening, it is of no consequence in relation to the purpose of the exercise, but it would be of consequence if such images had been processed for sale and exhibition. In such cases one would be advised to selectively sharpen certain parts of the image that contain detail whilst protecting other areas that have no detail.

Likewise, if one were exploring the potential of a new design of sensor to produce clean shadows, and comparing that quality with another design of sensor, one would probably shoot images of the same very high contrast scene with both cameras, then raise the shadows till the noise became apparent in both images, so that one could clearly see the noise, not because one likes to see noise, but in order to compare the degree of noise. However, if one were processing such images for sale or to hang on one's wall, one would certainly not want to see noisy shadows.

Hope I've clarified this for you.  Wink
Logged
hjulenissen
Sr. Member
****
Offline Offline

Posts: 1683


« Reply #10 on: April 14, 2013, 01:38:51 AM »
ReplyReply

Sharpening, like saturation, contrast and other parameters seems to be ones where the public wants "larger than life", while some purists wants "realistic" values. Is this not a general social mechanism in which the "masses" want to make sure that their friends get the message, while the elites wants to distinguish themselves from the masses by using more subtle cues that takes another member of the elite to recognize?

-h
Logged
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7703


WWW
« Reply #11 on: April 14, 2013, 02:12:35 AM »
ReplyReply

Downscaling to web size images is a bit tricky. I guess LR makes a good job of it.

Best regards
Erik


I have noticed this trend recently of over sharpening to the point of nasty artifacts and smooth areas taking on an orange peel effect. Why do people do this? Particularly amateurs with images that only get viewed on the web.
Can they not see what they are doing to their images?

I do a small amount of sharpening at output* depending on lens used then leave it up to the client to sharpen for their intended use as sharpening for print will mean the image is too gritty for small web use.

*usually just the normal/soft look from the old capture one in C1-7
Logged

marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #12 on: April 14, 2013, 11:22:40 PM »
ReplyReply

I think MrSmith is referring to an image that Ray posted under "Zooms or Primes" I like Ray and he is a good photographer My guess is he oversharpened w/o viewing the image downscaled for the web. We are all human and I'm the first to admit my mistakes, I make quite a few! LuLa is usually a civilized forum so Mr Smith if you have a personal issue with someone please take it off line. I'm glad you did bring up the topic because of it I will try the Nik sharpening software! And I'd like to thank Ray for his post I learned how good the Nikkor 85mm 1.8 lens really is.
Marc
Logged

Marc McCalmont
MrSmith
Sr. Member
****
Offline Offline

Posts: 917



WWW
« Reply #13 on: April 15, 2013, 02:57:11 AM »
ReplyReply

Absolutely nothing personal against any LuLa member and the posts/thread you refer to was a cordial one without venom. And those images were crops to show the sharpness differences in 2 particular lenses not finished portfolio pieces.

My opening post refers to a trend I have noticed growing over time which people seem happy to comment on, I expect they can think of many instances where they have noticed this trend.

As you were.
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8917


« Reply #14 on: April 15, 2013, 05:22:41 AM »
ReplyReply

I think MrSmith is referring to an image that Ray posted under "Zooms or Primes" I like Ray and he is a good photographer My guess is he oversharpened w/o viewing the image downscaled for the web. We are all human and I'm the first to admit my mistakes, I make quite a few! LuLa is usually a civilized forum so Mr Smith if you have a personal issue with someone please take it off line. I'm glad you did bring up the topic because of it I will try the Nik sharpening software! And I'd like to thank Ray for his post I learned how good the Nikkor 85mm 1.8 lens really is.
Marc

Not a problem, Marc, but thanks for your support. However, I  did also wonder if Mr Smith was directing his question mainly at me for oversharpening those recent lens-comparison images, but apparently he wasn't. The fact is, when doing such comparisons I'll try various degrees of sharpening in order to get a better practical perspective on the significance of such differences.

Whilst the resolution differences in the centre of many of my test shots are not that significant with D800E, which probably motivated me to oversharpen them in search of visible differences, I'm now seeing a noticeably greater difference in the centre when comparing the two lenses on the D7100, but not so great a difference at the edges of course.

Cheers!
« Last Edit: April 15, 2013, 05:24:15 AM by Ray » Logged
kers
Sr. Member
****
Offline Offline

Posts: 773


WWW
« Reply #15 on: April 15, 2013, 07:58:10 AM »
ReplyReply

Sharpening is one reason why i like to control my output; I have a large format printer so can do some trail and error searching for the best possible sharpening - i Like.
I have always liked to control the photographic process from start to finnish.
If i send images to magazines i have a standard procedure and check the outcome when i get my copy. Accordingly i make adjustments...
Logged

Pieter Kers
www.beeld.nu
Les Sparks
Full Member
***
Offline Offline

Posts: 207


WWW
« Reply #16 on: April 15, 2013, 10:48:41 AM »
ReplyReply

Optimum sharpening is hard. There is no good soft proof for sharpening. As Bart points out the sharpening tools don't say do more do less or whoa.
Part of the problem is defining optimum sharpening what does an optimum sharpened image look like--especially a optimum sharpened print. Back 100 years or so ago when I printed color negatives I never got good results until I finally saw one of my negatives printed by a great printer and found out what a good print was. From then on it was easy to get good results. I had a target. I think the same holds with sharpening. Once you see one of your images printed with optimum sharpening, the light goes on, and you know what the target is.
Logged

Ellis Vener
Sr. Member
****
Offline Offline

Posts: 1831



WWW
« Reply #17 on: April 16, 2013, 10:18:43 AM »
ReplyReply

Why do people oversharpen?
The simplest and truest answer: ignorance.
Logged

Ellis Vener
http://www.ellisvener.com
Creating photographs for advertising, corporate and industrial clients since 1984.
Ben Rubinstein
Sr. Member
****
Offline Offline

Posts: 1733


« Reply #18 on: April 16, 2013, 11:24:49 AM »
ReplyReply

I think we would also need to define what 'over sharpening' is, what is over sharp for some is good sharpening for another. I remember when canon came out with what they considered was optimum capture sharpening for the 1DsII at the time, USM 300,0.3,0 and the forums were full of people saying just how OTT that amount was. Once we can define what is correct sharpening we can work out who to point a finger at. Until that point and given that like colour it's a matter of taste anyway, perhaps we can just carry on with our lives?
Logged

MrSmith
Sr. Member
****
Offline Offline

Posts: 917



WWW
« Reply #19 on: April 16, 2013, 01:34:38 PM »
ReplyReply

I'm talking about creating detail that isn't there like the aforementioned orange peel effect that is artefacts from over sharpening.
 I don't worry about other peoples images so I can carry on with my life without getting anxious about them  Roll Eyes but wondered what was the thinking behind the phenomenon.
Logged
Pages: [1] 2 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad