PDA

View Full Version : How Much of a Crop Part II



Doug Brown
07-06-2010, 07:24 AM
I found Fabs' thread on cropping to be quite interesting. Many BPNers weighed in on the subject. What wasn't clear to me was how people decide the percentage of an image that a crop represents. I'm trying to see if there's a common language among photographers about crop percentages. Here's a RAW file of mine with a crop outline. Without getting out a ruler and calculator, how much of a crop do you estimate this to be? If you saw this image on your camera's LCD, would you delete it because the subject doesn't occupy enough of the frame? Or would this be a potential keeper? Please also describe how you calculate a crop percentage.

Alfred Forns
07-06-2010, 07:46 AM
Doug that would be a keeper as far as the size in frame, no question about it.

Fabs was talking about large crops, like taking away half or more of the image. Have seen 75% crops posted and from a new camera can have very good quality (posted).

That image shown if it was taken with a 600? (couldn't read well) is one fine capture particularly if its the one I'm thinking with Jim. Same image made with a 300 would be a very large crop but would still give you a image you could post.... or the above taken with a 300 vs a 70-200 @ 70.

I understand some people will only look at the finish image but I also like to look at the skill involved in the capture. I don't find much joy in capturing a bird in 25% of the frame. With this I guess there is no right or wrong except if you want to enter a contest like the BBC.

... btw we could not come up with a good way of showing the crop percentage but is it needed?

Axel Hildebrandt
07-06-2010, 08:00 AM
I usually just take the longest side to determine the crop, for example if it is 3500 pixels with a 1D4 (4896 pixels horizontally), I would say that it is 71% of full frame.

Dave Blinder
07-06-2010, 08:17 AM
Doug - looks like a near perfect amount of pixels on the subject to me. Close enough for the sensor to resolve fine detail as well as show a bit of environment. I generally don't like filling too much more of the frame (unless going for a headshot) because of the danger of clipping limbs. I don't have the highest standards in the world, but I can't imagine anyone calling your image a discard because of proximity.

I'd guesstimate it as a ~30% crop, just from eyeballing it up.

Doug Brown
07-06-2010, 11:17 AM
57 views and only 3 comments?

Roger Clark
07-06-2010, 11:30 AM
Doug,
It looks like the left part of the image is cut off, so that makes it a little difficult to give the crop. It is probably about half the area.

But to avoid the different ways people call a crop, I say cropped to X megapixels.

In my view, the crop should be given as a percentage of the total area and the ratio of the cropped megapixels to the camera megapixels gives that ratio. But a 50% crop on a 21 megapixel camera is a lot different than a 50% crop on an 8 megapixel camera. Thus the % crop is misleading I feel, or at least incomplete. Then we get into the issue of some people counting from the top down and others the bottom up. For example consider an image from a 10 megapixel camera cropped to 8 megapixels. I've seen it described as a 20% crop and an 80% crop. Just saying cropped to 8 megapixels is more informative in my opinion.

Roger

William Malacarne
07-06-2010, 11:43 AM
Roger

I feel yours is the best answer I have heard on this subject...sure make sense to me.

Bill

Doug Brown
07-06-2010, 12:02 PM
Doug,
It looks like the left part of the image is cut off, so that makes it a little difficult to give the crop. It is probably about half the area.

It's actually the entire image Roger. Thanks for making some excellent points!

David Gancarz
07-06-2010, 12:34 PM
Doug: I can't recall ever seeing a definition of crop percentage, but I'd guess the appropriate definition would depend on how said definition is to be used. (btw, I didn't catch Fab's thread on cropping).

If the primary interest is resolution, then a straightforward measure should be based on number of pixels in the before and after images. So if the original was 2000x1500 pixels and it was cropped to 1000x750, then you are left with 1/4 the pixels. Would you then describe it as cropped to 25% of the original?

But that definition would be misleading if we were more interested in changes to height and width dimensions. In that case, going from 2000x1500 to 1000x750 is a 50% crop. What if you changed the aspect ratio and cropped, instead, to 800x960? In this particular case the diagonal of the cropped image is roughly the same for the 1000x750 and the 800x960, or 1250 = SQRT(H<sup>2</sup> + W<sup>2</sup>). So if the diagonal is the basis for your calculation, both would be cropped to 50% of the original.

Note that calculating the crop based on the cropped_diagonal/uncropped_diagonal is equal to SQRT(cropped_pixel_count/uncropped_pixel_count).

Either method seems valid to me, the important thing would be to identify which method is being used; i.e., "pixel crop" or "dimensional crop".

(sorry Artie, I couldn't help myself. I'm only in step 3 of my 10-step program ;)).

Daniel Cadieux
07-06-2010, 01:12 PM
Here's how I look at it:

If I cook a lasagna that is, say, 24 x 12 in area...then I eat until there is only 12 x 6 left in the pan, then I've only eaten half! 50%:).

Sabyasachi Patra
07-06-2010, 01:18 PM
The crop definition I use.

Original dimensions "X" pixels by "Y" pixels
Cropped dimension: "A" pixels by "B" pixels

So the percentage crop is: 100* {(X * Y) - (A * B)}/(X * Y)

So if Original dimensions are 3000 by 2000

and cropped dimensions are 2500 by 1500

Then the cropped percentage is : 100 * {(X * Y) - (A * B)}/(X * Y) = 100 * {(3000 * 2000) - (2500 * 1500)}/(3000 * 2000) = 0.375*100 = 37.5 %

That means 37.5 percent has been cropped or deleted or thrown off.

I feel this should be mentioned by people, so that the skill level involved comes to light. Skill interms of acquiring focus on a subject that is close and also skill in stalking or moving in close to the subject.

People may find it difficult to calculate it. One can just save this formula in microsoft excel and then input the figures to calculate.

Cheers,
Sabyasachi

Kaustubh Deshpande
07-06-2010, 01:29 PM
Doug, I prefer the percentage-of-area( 60% crop meaning 40% pixels thrown away) method. That is the most accurate mathematically. But if one is trying to make a pano crop by cropping from top and bottom, then it really doesnt matter that he/she has thrown away 30% of the pixels, right?

The idea is that the viewer should get a feel of how far the subject was. Helps newbies to learn. So words like 'big crop', 'cropped a bit on all sides for composition', 'cropped from top and bottom to make pano and little from sides for comp', 'cropped from horizontal to vertical' help a lot IMO. Without calculator, its not easy to figure out what percentage of pixels were thrown away.

Since we share the focal length and the crop factor, I think sharing the crop info is also important. goes hand-in-hand. This is becoming more important because the current cameras have a lot of pixels giving a lot of room to crop. Without the mention of crop info, its quite easy to get misled into thinking the photographer was very close to the bird. At times, the DOF does give it way....but not all the time. esp. flight-shots with sky BG.

Juan Carlos Vindas
07-06-2010, 01:43 PM
Quite interesting discussion!

Doug, I do it this way. My camera offers me 8.2 mp. so if I crop the image and get only 6.3mp then I had cropped about 23% off the original capture. Does it makes sense?
I am with Roger about every camera will offer different crop %, so of course is great to have more MP in case one wants to crop the image and still be able to make a fairly large print.

David Gancarz
07-06-2010, 01:56 PM
In observing the responses I notice a mix between what the output image is cropped TO (as in cropped to 46% of the original) versus how much the output image has been cropped AWAY (as in 54% of the original image was cropped away). It doesn't much matter whether you break your soft-boiled eggs from the big end or the little end, you just need to be specific. Personally, I'm a little-endian kind of guy.

Danny J Brown
07-06-2010, 03:16 PM
Here's how I look at it:

If I cook a lasagna that is, say, 24 x 12 in area...then I eat until there is only 12 x 6 left in the pan, then I've only eaten half! 50%:).

Man, I'm getting hungry.....what were we talking about?

WIlliam Maroldo
07-06-2010, 05:29 PM
What about working directly with the file size of an image? For example a 50% crop of a 10MB image would be 5MB. Simple mathematics and no dimensions are necessary to deal with. Just wondering why not. regards~Bill

John Chardine
07-06-2010, 05:59 PM
I'm formulating a reply!!!

arash_hazeghi
07-06-2010, 05:59 PM
Crop is an area factor so a 1000X1000 image (1Mpixels) is 1/4 or 25% of a 2000X2000 (4 Mpixel) image, in digital images quanta of information is a pixel so it is the number of pixels that matters and quantifies the information in the file. Of course all pixels are not created equal in terms of signal to noise ratio and whether they contain valid information etc. but that is a separate topic.

Stating the crop in percent doesn't mean much in terms of the quality that you would expect, for example a 50% crop of a 5D2 image will be superior to a full frame image of a 40D camera so it is pointless IMO, many people cannot do the math or judge the crop percent by looking at their image either. I think the best is just to state how many Mpixels you have after cropping for example if you have a 3000X2000 image after copping then say" crop is 6 Mpixels. So that is the best way IMO as well.

The main reason that cropping doesn't yield good results is lack of sharpness in the original file, many people struggle to get a photo that is tack sharp when viewed at 100% size, especially if is BIF. photographs that have sub-par sharpness or are noisy at full size cannot be cropped much without noticeable loss in IQ. If a photo is pin sharp and clean you can crop the heck out of it and it will still look good, on the web and printed. I have printed 2 mpixel crops from 5D2 at 12"X8" and the quality is just awesome :)

Doug Brown
07-06-2010, 06:16 PM
I like your math Daniel!

Doug Brown
07-06-2010, 06:20 PM
Hi Bill. A potential issue with your idea is that some of us do our cropping in LR on the RAW file itself. Thus it's a 'virtual' crop. LR tells you the new pixel dimensions but does not alter or provide you with the new file size.

Doug Brown
07-06-2010, 06:34 PM
I'll just throw in that the crop in my first post represents 50% of the pixels in the overall image (8 MP out of 16 MP).

Ed Cordes
07-06-2010, 06:34 PM
I look at the cropped image in relation to the original and judge the area of the crop as a percentage of the total original. With this in mind I would judge your image to be about a 50% crop. In my 7/d and 1D4 this is an easy crop still retaining a lot of detail.

Doug Brown
07-06-2010, 06:39 PM
The problem that I'm beginning to see is that we all have different definitions of crop percentage. :confused: Very interesting! And not entirely surprising. Keep the comments coming!

WIlliam Maroldo
07-06-2010, 07:33 PM
I was looking at ACR and the measurement (at the bottom) is in MP(megapixels), which is updated with the crop. Whether it is MP or MB doesn't really matter .Therefore MP(of the crop)/MP(of the base image)=X /100, where X is the percentage. Simple algebra, divide the original MP into the crop MP, multiply by 100, and you have your crop factor. Is Lightroom different, I thought they used the same RAW converter? regards~Bill

Doug Brown
07-06-2010, 08:28 PM
LR only gives you the pixel dimensions of the cropped image, not file size. It doesn't give you total pixels either, although that's easy enough to figure out.

Doug Brown
07-06-2010, 08:34 PM
Here is the range of crops based on peoples' methods of calculating them so far. I'm talking about the percentage of pixels remaining after the crop.

Me: 50% based on total pixels
Axel 1: 70% based on percentage of image width
Axel 2: 75% based on percentage of image height

Of course you could also call these 50%, 30%, and 25% crops. Kind of confusing!

Desmond Chan
07-06-2010, 08:35 PM
The problem that I'm beginning to see is that we all have different definitions of crop percentage. :confused: Very interesting! And not entirely surprising. Keep the comments coming!

Well, if we are to establish some standard for posting here, all we have to do is to agree on one method regardless of what we use somewhere else.

Dan Brown
07-06-2010, 08:53 PM
I like to crop in ACR, get the size that I want at a 5 by 7 ratio (which seems to me to be about the same ratio as a full frame) and just before opening the image into PS, I move the crop box to the upper left corner and then estimate the % of the full frame I will be opening, based on the width dimension. I think this is close enough. I know it doesn't reflect the % of pixels but it makes more sense to me to make this comparison to the full frame, IMHO!!

Don Railton
07-06-2010, 08:57 PM
Hi All

I'm going to vote for the (cropped megapixel count) / (as shot megapixel count) expressed either as a percentage or as the two numbers... We all know or can find out the sensor aspect ratio of the camera involved and from the posted image we can determine the aspect ratio of the final image, ie we can figure if it was cropped from the sides or top and bottom (if it matters..) What wever we do I think it should adhere to the "KISS' principle.

DON

Jim Neiger
07-06-2010, 10:11 PM
If you multiply the long by short number of pixels or use the file size for the original tif vs cropped tif, the resulting percentage should be the same. I think that either of these methods are the correct way to measure crop percent. Crop percent is not a very useful number, IMO. Filesize or number of pixels are more useful.

Juan Carlos Vindas
07-06-2010, 10:12 PM
It's funny to find that some fellow photographers crop in ACR. I never do this even if I am cropping a copy of the original file. I do prefer to crop later in PS and maintain my original file untouched.

Doug Brown
07-06-2010, 10:20 PM
It's funny to find that some fellow photographers crop in ACR. I never do this even if I am cropping a copy of the original file. I do prefer to crop later in PS and maintain my original file untouched.

RAW processors such as ACR and LR do not crop the actual RAW file. The original RAW file remains untouched. I find the LR crop tool to be much easier to use than the one in PS.

Juan Carlos Vindas
07-07-2010, 10:19 PM
RAW processors such as ACR and LR do not crop the actual RAW file. The original RAW file remains untouched. I find the LR crop tool to be much easier to use than the one in PS.
Good to know that Doug! I was not very sure if I could crop in ACR and don't make a mistake to be worried later. Thanks for the clarification.

WIlliam Maroldo
07-07-2010, 10:43 PM
Juan: indeed the RAW file is never touched; ACR creates a .xmp sidecar file (with the same name as the RAW file) with any changes made. Want to start over? Just delete the .xmp file. regards~Bill

Daniel Cadieux
07-08-2010, 06:55 AM
Stating the crop in percent doesn't mean much in terms of the quality that you would expect, for example a 50% crop of a 5D2 image will be superior to a full frame image of a 40D camera so it is pointless IMO,

Arash, I agree 100% that stating the crop doesn't matter in terms of image quality as posted....BUT!!!!....it matters in terms of fieldcraft, or moment of capture in the field - this is important for beginners who think we were all within arms reach of everybird we post and then get dissapointed that all they get is a small dot in their viewfinder (especially with small songbirds), even with "long" lenses.

Chris Baker
07-08-2010, 07:55 AM
Hi Doug,

To answer your questions in order;

Like many have already stated if I lose a quarter of the original I'm left with a 75% crop, the image I would deffo keep. i'm a firm believer in, it's the end result that counts. We take pictures and if we produce a picture that's sharp and pleasing to the eye, great. Fieldcraft does come into it but take the LBJ's, even a 500 with a 1.4 giving approx 12 feet min focussing distance they're not that big so cropping is required. Where's the fieldcraft (other than picking the spot) in setting up a blind, turning on your tape and waiting for a show. From what i've seen you get great results but there's no fieldcraft involved and i don't hear complaints about beginners and fieldcraft in this regard. I saw a guy in Spain once, pull in to a layby, wind his window down, fire up the tape, crack a beer and wait. When I returned a couple of hours later he showed me some nice shots, no fieldcraft, no achievement as far as I'm concerned but the end result is a nice, uncropped pic. If the technology is there people will use it and why not, I guess you'll probably still end up with a similar percentage of people like me and probably most of the members here who like to go out and connect with nature and take great joy with a capture taken in the wild with no assists other than fieldcraft.

Regards

CB

Rob Miner
07-08-2010, 12:01 PM
Here's where a beginner steps into the CROP!

What is the difference between cropped percentages? B.Moose Peterson states, I have a picture 20X24 hanging on my wall in the office that was shot with a 3 mega pixel camera that customers think it was shot with 12.4 mega pixels.

We are shooting pictures with all types of equipment. We use "cropping" to finalize our presentation. Should it make a great deal of difference to shoot a scene from 20" or from 30'? Presenting a 8X10" or a 20X24" picture from that capture? A quality presentation is a quality presentation be it 3 mega pixels or 27.555555 mega pixels. Or is the reason to have an equipment minimum (lens size, How big is yours club?.)

Apologies to the offended!

Rob................

arash_hazeghi
07-08-2010, 12:58 PM
Arash, I agree 100% that stating the crop doesn't matter in terms of image quality as posted....BUT!!!!....it matters in terms of fieldcraft, or moment of capture in the field - this is important for beginners who think we were all within arms reach of everybird we post and then get dissapointed that all they get is a small dot in their viewfinder (especially with small songbirds), even with "long" lenses.

Dan,
Most people cannot visualize or estimate distance to the bird correctly just from a given crop percentile, again we have different sensor sizes etc. in this case it is best to indicate approximate subject distance in ETL forums.

Roger Clark
07-08-2010, 09:44 PM
What is the difference between cropped percentages? B.Moose Peterson states, I have a picture 20X24 hanging on my wall in the office that was shot with a 3 mega pixel camera that customers think it was shot with 12.4 mega pixels.


I agree. It's all about the image, not what crop it is. I have a 3 megapixel image that was printed full page in Natures Best photo magazine in their annual contest and has sold in galleries as 16x18 inch prints.

But having said that, the image would have been that much more stunning if it were on a modern camera with 16 to 20+ megapixels.

Roger

Roger Clark
07-08-2010, 09:49 PM
Like many have already stated if I lose a quarter of the original I'm left with a 75% crop,


Chris,
This gets back to my first response in this thread. I would say that is a 25% crop with 75% of the image left. You cropped away 25%, not 75%. It this kind of confusion why I think it more informative to simply give the megapixels of the cropped image.

Roger

Doug Brown
07-08-2010, 10:07 PM
I'm casting another vote for 'it's all about the image.' But I do agree with Al that I don't take a whole lot of pride in taking a bird photo in which the subject occupies a very small portion of the frame. Having said all of that, it's important to remember that quite a few avian subjects will not allow you to get close to them. Flying raptors don't typically buzz you; they tend to keep their distance. You don't often know which direction they're going to fly, and if you get too close you'll start clipping wings in your images. If I can keep 50% of the pixels in a flight image from a high-megapixel camera, I'm usually pretty satisfied. Maybe we should think about adopting a standard BPN cropping vocabulary; if you talk about your crop, this is how we prefer that you calculate it. We could add a sentence to the posting guidelines. What do you all think?

For anyone who's interested, here's a 100% crop of the RAW file processed in LR only (with LR default sharpening).

Markus Jais
07-08-2010, 11:19 PM
This is a very interesting thread.
When I had my 10D I did not crop that much but I did more with the 40D.
Now that I own the 7D which delivers excellent results up to ISO 800 (at least when not shooting black crows against a dark BG) I crop down to 3 MP if I have too and this still delivers great results for posting on BPN or even for printing.
So I crop down to about 15% if I have too.

Markus

Eric Virkler
07-10-2010, 09:55 PM
I've never given any thought to crop pecentages. My only considerations are if the final outcome I have in mind for the image will be acceptable to me. I shoot with the 50D. I will often keep images with subject matter smaller in the frame than this owl, knowing I can crop in on it. Of course, it all depends on my plans for the image. If I'm going to sell it as a framed 8x10 or 11x14, I can crop much more than if I plan to see it as a larger image. And if my only plan is to use in a slideshow for family and friends, I can obviously crop much much more

I like the way Dan described the crop percentage, and I'll take the other half of the lasagna.

Eric Virkler
Faces of Nature Photography
www.ericjvirkler.com

Tom Redd
07-10-2010, 11:56 PM
Interesting thread, as the newbie here, I almost asked this question a couple of days ago and later I got lucky to stumble upon this. Thanks to all for the comments.

Sabyasachi Patra
07-12-2010, 02:45 AM
if it is difficult to calculate the amount of image cropped and discarded, then one can atleast mention the number of pixels in the horizontal and vertical side after cropping and the camera model. Others can know how much crop it is.

I feel at the moment there is no level playing field, so to speak. People crop a lot and when a newbie comes and sees the images he/she would wonder how others are getting such frame filling shots. By cropping, we are paying no attention to field craft. I think if a person deals with the distractions in the field itself and composes an image in the camera, then it is worth more than cropping in the computer.

Ray Rozema
07-13-2010, 02:49 PM
Thanks to everyone for helping us to learn. Very interesting and slightly confusing. I hope a standard guideline can be reached

Ray Rozema

Ákos Lumnitzer
07-13-2010, 08:23 PM
I firmly believe and stand by the fact that the real skill lies in being able to capture an image as big as possible on the sensor whether 4mp or 40mp. Cropping is the lazy and unskillful way of nature photography and holds no true value to me personally. However, that is most likely a minority opinion.

:D

I know people got pooey with me in the past for telling them how I would calculate ratio of crop when it proved that they cropped much more than what they disclosed. I have attached a sample from my simple excel worksheet that can tell me if I really want to know. I usually round up or down to the nearest 5.

John Chardine
07-13-2010, 09:23 PM
My 2 cents worth, and with the greatest of respect to responders- it matters little if you crop in camera or in processing so long as you consider the process of making an image a continuous one from planning the image to pressing the shutter to the final output, whatever that might be. This is what photographers have done for as long as photography has been around. If you have worked in a darkroom you cannot imagine not cropping an image if it called for it- it's just another step in the creative process. The idea that somehow the creative process stops after you press the shutter really misses the mark as far as I am concerned, and as I have said, ignores the long history and standard methodologies of producing a photographic image.

Nature photography is not a video game where you either hit the target or you don't.

Roman Kurywczak
07-14-2010, 10:32 PM
I've tried to refrain....reallly:2eyes2:!!! I am with John in one respect.....cropping has always been a part of photography.....can you say slide mounts!!! But Akos will be happy to hear my in the field motto: "Shoot tight or go home!".......but that is to my more experienced friends who have the ability......I think people need to grow into that one.:cheers:
OK......I just wanted to use the new icons.....can you guess which one I proposed? ;)

Desmond Chan
07-14-2010, 11:01 PM
I firmly believe and stand by the fact that the real skill lies in being able to capture an image as big as possible on the sensor whether 4mp or 40mp.

Well, I just happen to have a beautiful vase sitting just a few feet away from me. And I just also happen to have a lens that allows me to take a close up of that vase. How much skill is involved here really? :p


Cropping is the lazy and unskillful way of nature photography ...If only nature is what you want it to be all the time.

I know, I know. I shouldn't call myself a photographer. Now you know what I am not.

Daniel Cadieux
07-15-2010, 08:41 AM
Not sure if this has been mentioned before, but another good reason to mention cropping along with the exif is that it helps with the critiques comp-wise. We can better judge if the image has been cropped too much, not enough, if there is enough room to crop even more than already, etc...

P.S. Hmmm, not sure which new smily you would have proposed, Roman:cheers:

John Chardine
07-15-2010, 09:41 AM
I've tried to refrain....reallly:2eyes2:!!! I am with John in one respect.....cropping has always been a part of photography.....can you say slide mounts!!! But Akos will be happy to hear my in the field motto: "Shoot tight or go home!".......but that is to my more experienced friends who have the ability......I think people need to grow into that one.:cheers:
OK......I just wanted to use the new icons.....can you guess which one I proposed? ;)

You bring up a good point about slides Roman, but you know I think slide film is at the root of this idea that everything has to be done in the camera. Before the digital era broke, slide film was very popular with serious photographers- the quality was superb and some brands at least were archival. Nothing matched showing a high quality slide on a good projector. The problem with slide film was that you pressed the shutter button and that was virtually it- nothing more could be done to the image (with the exception of those corny cropped slide mounts). The slide film was typically sent away for processing (you HAD to for Kodachromes) and came back mounted or unmounted. Sometimes large-scale dupes were made of the slides and these were printed using an internegative or directly using products like Cibachrome but the majority of non-professionals simply stored their slides and occasionally hauled them out for a slide show. This is where many people lost the common thread of photography that creating a final product was a continuum from pressing the shutter button through processing, including film development and printing.