Do crop sensor camera's lose sharpness in the same frame?

[Tim Deen]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

And if this is the case, do crop sensor images lose sharpness due to being enlarged within a same frame?

 

[Quarkcharmed]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen?

Crop sensor is called 'crop' against a 'full frame' because shooting with a crop sensor is equivalent to shooting with a FF sensor and then cropping the image, given the sensor pixels are the same size.

Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)?

To what frame exactly? You always have some display, computer monitor or a print of a certain size to display your image.

If your target frame is 3000:2000 pixels, and the image is 6000x4000 pixels, you'll need to scale it down.

And thus resulting in a more zoomed in image?

And if this is the case, do crop sensor images lose sharpness due to being enlarged within a same frame?

Generally you lose resolution due to increasing diffraction as you go from larger to smaller sensors. But the difference in sharpness between FF and crop sensors will depend on many other factors, including the lens/optics quality, aperture and sensor pixel size.

 

[alcelc]

IMHO it should be determined by the resolution.

Under a perfect focusing condition, and same lens, higher resolution (more pixel to form an object) will mean more detail and so should look sharper.

For an easy illustration, a FF camera has a 4000 x 4000 pixels resolution, vs a 2x crop factor camera which also has 4000 x 4000 pixels resolution. When an object will be covered by 1000 x 1000 pixels on FF camera, the same object on the 2x crop factor sensor will become 2000 x 2000 pixels by using the same lens at the same shooting distance. Obviously the crop sensor camera will have more detail and so should look sharper.

The zoom in effect of crop factor has played a role here. Therefore if you will use two lenses of similar IQ, a 50mm on full frame and 25mm on 2x crop factor camera, or shoot closer with a FF camera to eliminate the zoom in effect, they should be even.

However if the FF camera has higher resolution, says 8000 x 8000, the object in above will has a resolution of 2000 x 2000, becoming the same as it is on a 2x crop sensor camera. They will be the same again.

I think you could test it easily with your camera. Take a shot (best appearing small on the frame) with a YY focal length, another shot at 2x YY focal length (e.g. 50mm & 100mm), if sharpness of both lenses will be the same, compare them at 1:1, you should find the one from 2x YY focal length lens could have more fine detail and so should look sharper.

--
Albert
** Please forgive my typo error.
** Please feel free to download my image and edit it as you like **

 

[chrisfisheye]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

And if this is the case, do crop sensor images lose sharpness due to being enlarged within a same frame?

Yes and no ( if I understand correctly what you say).

Strictly speaking, what you say is correct. If a final image with a full frame has x effecfive resolution, then if you had used a m43 sensor with the same lens (and same settings) it willl have x/4 effective resolution. That said the effective resolution is not homogenous in the frame, so you loose much less but you loose... The logic is simple, you loose pixels by cropping. To simplify, I have considered you use the same lens but whatever..

Now let's say the max effective resolution of this lens is when you select f/5.6. This means between FF and m43 sensor with FF you have more max effective resolution when you select the best settings for this lens.

But..

Suppose you are a landscape photographer and you need f/16 for dof purpose. You will certainly need f/8 if you take the same image with m43 to have the same dof and f/8 has more effective resolution (per millimeter ) due to less diffraction. In fact, in theory, if you consider you have a perfect lens and an infinite number of pixels, the 2 images will be equivalent in terms of resolution. In this use case, the more effective resolution per millimeter compensates the smaller sensor size

Think about it. A small sensor in smartphones with crop factor 5× does not result in 25 times less effective resolution than FF !! Strangely, especially if you need dof, the images can be very similar even in terms of resolution.

Sometimes I rezd that a lens for a crop sensor is generally sharper(per mm) but the logic is biased.. the main cause is that you generally use it with lower f#.

Now the logic I described by using f/16, or more generally at high f# does not work with low f#, at least in practice. In pure theory when you can select equivalent f#, the resolution is the same.

There is a chance that the sweet spot of the lens corresponds for instance to f /5.6 and that the sweet spot for a FF lens is significantly sharper than with m43.. if you select f/5.6 for FF and f/2.8 with m43, f/2.8 is more challenging in practice.

At low f# and when you select the sweet spot of the lens, FF is generally significantly sharper.

With time, the lenses will get better and better so maybe even at low f# (but equivalent f#l the difference will not be important. The sweet spot maybe will be at f/2.8 where FF will keep its advantage.

Hope this makes sense,

 

[bjn70]

A lot of things are interconnected here. For 2 cameras taking a photo of the same subject at the same distance the FF camera would use a longer focal length lens than the APS-C camera. Another factor is what is the image resolution of each one? I have a 24MP full frame and a 24MP APS-C. That means the pixels of the APS-C sensor are more densely packed than the pixels of the FF and this requires that the lens does a better job of projecting the detail onto the smaller sensor. It is not hard to find lenses that are capable of this but there are probably lenses that would not do as well so you have to consider the quality of your lenses.

Some years ago I had a 24MP APS-C camera, I took a photo of the city skyline from the roof of the building where I work. Then I lost that camera in an accident and replaced it with a 24MP FF camera. I ended up with a photo from it of the city skyline from the roof, almost the exact same image as I took a year before from the other camera. So I compared them and pixel peeping the APS-C actually looked sharper. The main reason is because its sensor did not have an AA filter but the FF camera did, so that can be another factor.

 

[MarshallG]

Tim, a full frame sensor is only 24x36mm, so everything you shoot with a full frame or crop sensor is enlarged. You always “lose sharpness” when you enlarge, but the pixel counts are so high these days that you can print quite large without really “scaling” at all. For instance, if your display medium is a 60” 4K TV, you only have 8 megapixels, and any APS-C camera these days has a lot more resolution than that. Same with printing.

Actual side-by-side comparisons need to take more into account than the sensor size, because sensor technology continually improves, and different sensors have different advantages.



It is very easy to get confused by the physics of it all, so my advice to you, is start with an APS-C setup, because they’re less expensive and lighter. It used to be that you could only get premium features in full frame cameras, but that’s not the case anymore. All things being equal, full frame cameras are better at delivering ultra thin depth of field and they’re a little better in low light conditions. In sports and wildlife photography, crop sensors can offer significant advantages. In Canon, for instance, their crop sensor R7 has the highest pixel density of all their cameras. So if you’re shooting wildlife and using your telephoto’s entire range, the R7 will put a lot more pixels on the subject than the 45 mp sensor of their full frame R5 camera.

 

[ThrillaMozilla]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

You tell me. That's probably what you're going to do, isn't it?

It really is that simple. If you enlarge the image, everything in the image gets enlarged. You have it figured out. Think it through before you read all the confusing and probably poorly worded explanations.

 

[meow]

An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

I hope I understand you right. If I do you have misunderstood this a bit.

Nothing gets enlarged. Simply put It's just that due to the smaller sensor a lens with a certain focal length makes the subject fill a larger part of the sensor.

Say you want to photograph birds. You have 200mm lens. On a FF you'd need to get very close to small bird to make it fill the sensor. Put the same lens on a MFT camera and it would become a lot more likely that you'd manage to do that, because it corresponds to a 400 mm lens on a FF - and that's because the sensor is smaller, nothing else.

The so called cropping happens at the sensor. Not that it's possible, but imagine you shoot that 200mm lens on a FF camera, but you have cut its sensor down to the size of an MFT sensor. That would give you the same result when it comes to how large a part of the sensor the subject fills..

 

[Tim Deen]

An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

I hope I understand you right. If I do you have misunderstood this a bit.

Nothing gets enlarged. Simply put It's just that due to the smaller sensor a lens with a certain focal length makes the subject fill a larger part of the sensor.

Say you want to photograph birds. You have 200mm lens. On a FF you'd need to get very close to small bird to make it fill the sensor. Put the same lens on a MFT camera and it would become a lot more likely that you'd manage to do that, because it corresponds to a 400 mm lens on a FF - and that's because the sensor is smaller, nothing else.

The so called cropping happens at the sensor. Not that it's possible, but imagine you shoot that 200mm lens on a FF camera, but you have cut its sensor down to the size of an MFT sensor. That would give you the same result when it comes to how large a part of the sensor the subject fills..

I understand that everything, when the photo is taken remains the same size. The smaller sensor just takes a smaller portion of the subject your are making a picture of.
Where it went wrong for me is that I misunderstood that the image will look larger in a same frame, when compared to a FF sensor. To me it feels like the image gets enlarged in that case to match a similar frame of the FF image (taken it's not being cropped).

Thanks for your reply.

 

[Tim Deen]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

You tell me. That's probably what you're going to do, isn't it?

It really is that simple. If you enlarge the image, everything in the image gets enlarged. You have it figured out. Think it through before you read all the confusing and probably poorly worded explanations.

Thanks, good to hear that my way of thought isn't wrong.

I do have a remaining question though. Let's say the pixel density on both a FF sensor and a crop sensor is the same. I take a picture of a subject with both a FF camera and a crop sensor camera. I then project them within the same frame. The crop sensor image would get enlarged to fit the frame. Would I then lose sharpness due to enlarging the same pixel size compared to the FF sensor?

 

[Tim Deen]

Thanks for your explanation. I have owned a crop sensor camera for quite a while. I am making a step towards a Full frame camera because I focus on landscape alot. I will problably end up getting a crop sensor camera again later though for things like wild life.

I am happy I understand the differences and what happens to the image of a crop sensor camera when you compare it to the same frame.

 

[alcelc]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

You tell me. That's probably what you're going to do, isn't it?

It really is that simple. If you enlarge the image, everything in the image gets enlarged. You have it figured out. Think it through before you read all the confusing and probably poorly worded explanations.

Thanks, good to hear that my way of thought isn't wrong.

I do have a remaining question though. Let's say the pixel density on both a FF sensor and a crop sensor is the same. I take a picture of a subject with both a FF camera and a crop sensor camera. I then project them within the same frame. The crop sensor image would get enlarged to fit the frame. Would I then lose sharpness due to enlarging the same pixel size compared to the FF sensor?

It seems that the ENLARGEMENT could be the issue here.

1st, the most common standard resolution of monitor/TV is 4K nowadays. The max resolution a 4K monitor can display is 8Mp. Therefore, an image of 8Mp resolution would display 100% (1:1) and fill the monitor. Any image has a lower resolution will have to enlarge (by intepolation of the display card of the computer or by the TV) to fill the screen.

As even the very old cameras of last 10+ years would produce images not smaller than 10Mp or 12Mp, not to mention the 24Mp is now a starting point for most APSC or FF, when you display an image on Full Screen basis, you are actually seeing a scale down image.

Generally softness, slightly out of focus or blurry, noise, color transition etc of an image could be hidden up during scaling down e.g. a 24Mp image has to be shrinked by 67% when display on a 4K monitor.

Therefore, no matter it will be an image from FF, APSC, M43, 1" or from smaller size sensor, on Full Screen basis you might not easy to judge which one is really sharper.

Sharpness should be determined by the condition of the edge formation (on pixel basis) of the image. Less concern on resolution (subject to no enlargement), not to mention sensor size.

When 8K (33Mp) monitor would become a standard in future, those <33Mp image would start to show their shortage as they would look smaller on 100% display, or have to be enlarged on Full Screen basis which might degrade the displayed image. Again, FF or small sensor is not the determine factor here.

I shall take adventages of FF system on noise performance and shallower DoF vs a crop sensor system. Not sharpness...

 

[Michael Fryd]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

And if this is the case, do crop sensor images lose sharpness due to being enlarged within a same frame?

Let's compare a full frame 50 megapixel camera to a 30 megapixel 1.6X crop body. Both are shooting the same subject, under the same lighting, with the same actual focal length, f/stop, shutter speed and ISO.

The image from the full frame will cover a wider field of view. I believe that your question is how the results compare if we crop that image to match the field of view of the crop body.

The answer is that they will be the same, except that the crop body image will have a higher resolution. This is because it has more pixels covering that area of the frame.

The 30 megapixel 1.6X crop body produces a 30 megapixel image.

A 50 megapixel full frame cropped to match has only about 19 megapixels.

.

Let's suppose you are shooting a subject, and you don't have a lens long enough for your subject to fill the frame. Perhaps you are shooting a full moon, and you only have a 400mm lens. If you shoot with a 30 megapixel crop body, you will have more pixels covering the moon, than if you had shot with a 50 megapixel full frame. If pixel count is your limiting factor, then you can make a larger print from the crop body image.

 

[MarshallG]

Thanks for your explanation. I have owned a crop sensor camera for quite a while. I am making a step towards a Full frame camera because I focus on landscape alot. I will problably end up getting a crop sensor camera again later though for things like wild life.

I am happy I understand the differences and what happens to the image of a crop sensor camera when you compare it to the same frame.

Tim, to be honest, I think that there are a lot of ways to apply terms like “sharpness” and at a certain point, if you want to be very precise in what you’re trying to accomplish, it takes a lot of effort and expense.



The best answer to your question, I think, is that for something as casual as onscreen display, Instagram, or basic 8x10 prints, you will not be able to find any difference.



But if you’re trying to make 3x4 foot metal or paper prints, and you want a very specific look, then you can get into a lot of expensive trial and error perfecting the look you’re after. To be clear, people do that, they definitely do!

 

[ThrillaMozilla]

An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

I hope I understand you right. If I do you have misunderstood this a bit.

Nothing gets enlarged.

The image gets enlarged when it's viewed. It's maybe 24 mm x 36 mm on the sensor, but you're going to view it much bigger than that. That's enlarged.

 

[ThrillaMozilla]

I understand the basics of a crop sensor vs. a full frame sensor. The crop sensor catches a narrower field of view at a same focal length. The crop factor is the factor between the size of a full frame sensor and a crop sensor. All clear.

What I fail to understand though is when you start comparing images. An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

You tell me. That's probably what you're going to do, isn't it?

It really is that simple. If you enlarge the image, everything in the image gets enlarged. You have it figured out. Think it through before you read all the confusing and probably poorly worded explanations.

Thanks, good to hear that my way of thought isn't wrong.

I do have a remaining question though. Let's say the pixel density on both a FF sensor and a crop sensor is the same. I take a picture of a subject with both a FF camera and a crop sensor camera. I then project them within the same frame. The crop sensor image would get enlarged to fit the frame. Would I then lose sharpness due to enlarging the same pixel size compared to the FF sensor?

I'm not sure what you mean by the same pixel density -- whether that's the same number of pixels per millimeter (i.e., the same pixel pitch) or the same number of pixels per image. I suppose I should know, but I'm confused. Obviously, the answer will be different in each case.

I also don't know if you are using the same lens or a different lens. Normally you would use a different lens, but sometimes, especially for wildlife photography, people will want to compare the cameras with the same lens.

And then there's the possibility of cropping the full frame image, so now I'm triply confused about the question.

So without knowing exactly what the problem is, I can't be sure of answering the right question. You have already received three different answers -- and all of them correct and helpful.

I'll give you a couple of answers, though. If both cameras give the same number of pixels, a 32 mm lens on APS-C can give you essentially the same image as a 50 mm lens on FF. You would use different settings on each camera. That also assumes equal quality of lenses, etc., but it's very reasonable to expect approximately equal results from each camera under many conditions. (The FF camera, however, has a somewhat wider range of capabilities for low light, however, and the APS-C camera also has its advantages, so the cameras are not completely equal.)

The second answer is that if both cameras have the same number of pixels and you use the same lens to photograph a duck, the APS-C camera will give you more pixels per duck, and that's good.

I'm not sure either of those answers corresponds to your question, however. Sorry.

 

[ThrillaMozilla]

I shall take adventages of FF system on noise performance and shallower DoF vs a crop sensor system. Not sharpness...

Oops. Those are other subjects. Your answer is oversimplified and correct only under some conditions. It's best not to discuss them here. Everything else in your post is correct and reasonable, however.

 

[chrisfisheye]

An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

I hope I understand you right. If I do you have misunderstood this a bit.

Nothing gets enlarged.

The image gets enlarged when it's viewed. It's maybe 24 mm x 36 mm on the sensor, but you're going to view it much bigger than that. That's enlarged.

An image has no dimension..

The sensor is just a tool to record an image but the result is dimensionless.

You can imagine taking the exact same image, one with FF and the other with APS-C with the same resolution. You would have 2 identical image files. What would it mean to say that the image with the crop sensor needs more enlargment ??

Also, it is important to note that the resolution is present in the image from start. An image has a given resolution (calculated as LW/PH, H is height of the image). So the resolution is understood as resolution per image.

For me there is no enlargment. It is much better to say that an image has been recorded with a sensor of a given size.

 

[ThrillaMozilla]

An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

I hope I understand you right. If I do you have misunderstood this a bit.

Nothing gets enlarged.

The image gets enlarged when it's viewed. It's maybe 24 mm x 36 mm on the sensor, but you're going to view it much bigger than that. That's enlarged.

An image has no dimension..

Really? The real image is 24 mm x 36 mm in a full frame camera. A 24 MB digital image is typically 4000 pixels x 6000 pixels. Those are the dimensions.

The sensor is just a tool to record an image but the result is dimensionless.

Really? A two-dimensional array is dimensionless? And the display is dimensionless?

You better look up the meaning of "dimension".

You can imagine taking the exact same image, one with FF and the other with APS-C with the same resolution. You would have 2 identical image files. What would it mean to say that the image with the crop sensor needs more enlargment ??

A 24 mm x 36 mm real image is digitized and viewed at 24 cm x 36 cm. (Hint: it's bigger.) Could you possibly do any more to obfuscate the discussion?

Also, it is important to note that the resolution is present in the image from start. An image has a given resolution (calculated as LW/PH, H is height of the image). So the resolution is understood as resolution per image.

And the resolution of the real image is measured in micrometers. In related news, all green frogs are green and frogs. (Thanks to Dem Bell for that observation. )

For me there is no enlargment. It is much better to say that an image has been recorded with a sensor of a given size.

The image was small, then it was digitized and viewed at a larger size. To use the same word as the OP in his perfectly adequate colloquial manner, it is enlarged.

The OP is just trying to understand the subject. Mindless, pedantic obfuscation really doesn't help anything.

 

[chrisfisheye]

An often praised feat of crop sensor camera's is the fact that you can enlarge your subject. But what does actually happen? Does the image get enlarged to the same frame as an full frame image (if you compare FF to crop)? And thus resulting in a more zoomed in image?

I hope I understand you right. If I do you have misunderstood this a bit.

Nothing gets enlarged.

The image gets enlarged when it's viewed. It's maybe 24 mm x 36 mm on the sensor, but you're going to view it much bigger than that. That's enlarged.

An image has no dimension..

Really? The real image is 24 mm x 36 mm in a full frame camera.

A 24 MB digital image is typically 4000 pixels x 6000 pixels.

Those are the dimensions.

Nope. A pixel is not a unit.

When you say 6000 pixels it is a pure quantity..

The sensor is just a tool to record an image but the result is dimensionless.

Really? A two-dimensional array is dimensionless? And the display is dimensionless?

You better look up the meaning of "dimension".

https://en.wikipedia.org/wiki/Pixel...dered in another way, a,a pixel density (ppi).

"Considered in another way, a pixel has no inherent size or unit (a pixel is actually a sample), but when it is printed, displayed, or scanned, then the pixel has both a physical size (dimension) and a pixel density (ppi).[1]"

But we can go to the scientific forum (PST forum) , no problems !!!

You can imagine taking the exact same image, one with FF and the other with APS-C with the same resolution. You would have 2 identical image files. What would it mean to say that the image with the crop sensor needs more enlargment ??

A 24 mm x 36 mm real image is digitized and viewed at 24 cm x 36 cm. (Hint: it's bigger.) Could you possibly do any more to obfuscate the discussion?

Also, it is important to note that the resolution is present in the image from start. An image has a given resolution (calculated as LW/PH, H is height of the image). So the resolution is understood as resolution per image.

And the resolution of the real image is measured in micrometers. In related news, all green frogs are green and frogs. (Thanks to Dem Bell for that observation. )

For me there is no enlargment. It is much better to say that an image has been recorded with a sensor of a given size.

The image was small, then it was digitized and viewed at a larger size. To use the same word as the OP in his perfectly adequate colloquial manner, it is enlarged.

The OP is just trying to understand the subject. Mindless, pedantic obfuscation really doesn't help anything.

You are trying to discredit other people's answers and you continue unfortunately..

Here is what you said, not me..;

"Think it through before you read all the confusing and probably poorly worded explanations."

And sorry, you should learn about what is a dimension, just saying...