How does one calculate the percentage of a crop? I've seen references to "100 percent crop," "50 percent crop," and the like.
Suppose the original frame is 4000 pixels by 2000 pixels. I crop the frame so it is 2000 pixels by 1000 pixels. This would logically be a 50 percent crop to me (half the linear dimensions of the original), but if we multiply the pixels of the original and the cropped version, we get 8,000,000 pixels and 2,000,000 pixels. That looks like a 75 percent crop to me.
And linearly a 100 percent crop would yield 0 pixels!
Thanks for any insight into this. My head is hurting from trying to figure this out. I am just not that smart.
I dare to open my mouth... I'm by far not an expert.
Think you've got a P&S camera with 800x600 pixels.
your screen resolution is 800x600. Then the picture taken is matching exactly the dimension of your screen. So the OS doesn't have to resize a big picture to a small display. this picture is rendered at 100% resolution
now think you crop an area of 200x300 of the bigger picture. That is a crop of an 100% image.
In fact, any crop with 200x300 or any other dimension that fits in the screen can be called 100% crop. Cropping is not resizing
Generally "N% crop" is shorthand for "an area (of any size) cropped from an image that has been resized to N% of its original size". The percentage is referring to any resizing algorithm used, not to the amount of the image contained within the crop.
For example, in this post, I first show a resized image. Below that I show what I call "100% crops", meaning I didn't perform any resizing on them. Each cropped area shows less than 1% of the area of the original image (less than 15% in each dimension), but the part shown is identical to the original; nothing was magnified or shrunk. (It's useful information for that post, since it's discussing pixel-level detail.)
As for the resizing, it does refer to the linear dimensions, not the actual area: a 50% resize of a 4000x2000 image results in a 2000x1000 image. You can crop an area of any size you like from that image and call it a 50% crop.
Also, the only crop percentage you normally see anyone talk about is 100%. It means what people say - a crop of *any* sie at all, as long as you don't resize it also. A 50% crop would mean a crop of any size at all that you then resize to 50%, but hardly anyone ever talks abut that, because no one would care. We care about 100% crops because it's the only way to see (a part of) the image at its full size and have it still fit on the screen.
I recognize that it's a very common term, but it's a very poor one and wish that people would stop using it. Saying "cropped 1:1 view" or "pixel-level detail" would be so much more informative. That way "50% crop" can go on meaning what it makes sense to mean: I cropped out half the image.
I recognize that it's a very common term, but it's a very poor one and wish that people would stop using it. Saying "cropped 1:1 view" or "pixel-level detail" would be so much more informative. That way "50% crop" can go on meaning what it makes sense to mean: I cropped out half the image.
I agree, Matt. When I first saw the term, it took me several days of reading threads to put it into context and figure out what it meant.
I recognize that it's a very common term, but it's a very poor one and wish that people would stop using it. Saying "cropped 1:1 view" or "pixel-level detail" would be so much more informative. That way "50% crop" can go on meaning what it makes sense to mean: I cropped out half the image.
That makes a lot of sense to me. If the original is resized to say 1024x768 pixels to display on the web and then a section is cropped from the original FULL size image and displayed, one can immediately see the detail in the image and its perspective
Actually, the term %100 crop does make sense--we just need to think of the term the way it was conceived. It means you are seeing a 1:1 (or %100) version of the original shot, without "resampling" (resizing). This is useful because it allows us to see a section of what the original full-blown shot looked like, without having to download the whole image. You are seeing a small section of the original, at %100--in other words, at full size resolution.
If our monitors were extremely large, we could view the entire shot, all at once, on the screen. But our monitors are not nearly large enough, nor are our processors nearly fast enough, to make such viewing practical.
So the %100 crop plays a key standardizing, role. It enables different images, from different cameras, to be viewed on a level playing field. There is more discussion of this, here:
Actually, the term %100 crop does make sense--we just need to think of the term the way it was conceived. It means you are seeing a 1:1 (or %100) version of the original shot, without "resampling" (resizing). This is useful because it allows us to see a section of what the original full-blown shot looked like, without having to download the whole image. You are seeing a small section of the original, at %100--in other words, at full size resolution.
If our monitors were extremely large, we could view the entire shot, all at once, on the screen. But our monitors are not nearly large enough, nor are our processors nearly fast enough, to make such viewing practical.
So the %100 crop plays a key standardizing, role. It enables different images, from different cameras, to be viewed on a level playing field. There is more discussion of this, here:
That's the way I've always used "100% crop". If a "50% crop" means you took away half the image than wouldn't a "100% crop" mean you took away all the image? If instead "100% crop" means it's the full image than it's not a crop it's "un-cropped." That being said, I do agree that "1:1 crop" might require less explanation. Something I need to think about...
I recognize that it's a very common term, but it's a very poor one and wish that people would stop using it. Saying "cropped 1:1 view" or "pixel-level detail" would be so much more informative. That way "50% crop" can go on meaning what it makes sense to mean: I cropped out half the image.
Originally Posted by Sean Nelson
In an ideal world, everyone would use two different terms to refer to the two concepts being bandied about here:
Crop - the portion of the original image that has been cut out and presented for viewing.
Scale - the degree of enlargement or reduction from the original pixels to the pixels being viewed.
For example, "this picture is a crop of the central 10% of the original image at a scale of 100%" (or, "at a scale of 1:1").
Matt and Sean: You always seem to cut to the logical heart of an issue. I like "pixel-level detail" to express the intent of "100% crop." Now we'll see if I remember to use it next time . . .
Matt and Sean: You always seem to cut to the logical heart of an issue. I like "pixel-level detail" to express the intent of "100% crop." Now we'll see if I remember to use it next time . . .
+1. It would eliminate any confusion. Maybe we could adopt a standard abbreviation like "PPV". (Pixel Peep View)
N% crop is shorthand for area cropped and resized to N%
Originally Posted by Quension
Generally "N% crop" is shorthand for "an area (of any size) cropped from an image that has been resized to N% of its original size".
+1
Personally, I have already used the terms 100% crop, 200% crop, 50% crop, 27% crop (full size K20D @ 1200 px) and 16% crop (full size K20D @ 700 px). Note how the latter two aren't crops at all
But I only do so in technical discussions to help interpret the technical aspects in an image.