Crop sensor and lens aperture

[robbins.photo]

Can you explain it a bit better ? If there is the same intensity of light per every square millimetre of the sensor how comes bigger sensor should sense more light ?

Well I can.. problem is doing so in such a fashion that the post doesn't become longer than Tolstoy's War and Peace.. lol. Ok, I'll give it a quick stab, and no this is by no means complete but hopefully it will cover the basics.

A sensor in a digital camera has millions of spots that are sensetive to light, called photosites. Each photosite records information about what is being "seen" through the lens, so the more photosites you have and the larger they are naturally the more information that is being collected. A larger sensor allows for larger photosites - and a larger photosite will collect more from a ray of light than a smaller one can. If it helps think of it in terms of water. If you put a coffee cup on the floor and pour water into it, once the cup is full any more water will flow out of the top and simply be wasted. Replace the cup with a bucket, and now you can pour the same amount of water and because of the buckets larger size, it will collect a lot more of the water.

The amount of water your pouring doesn't change (much like the amount of light coming into the camera doesn't change when your using the same aperture) however because you now have row after row of buckets instead of row after row of coffee cups, the amount of "water" or in this case the amount of information from the light captured that is retained is a whole lot more.

I know that explanation, it only explains, why larger sensors produce less noise giving cleaner images (that's why people buy them). You try to convince me, that larger sensors because of larger photosites records more light then it is there ? Maybe they have this ability, but shouldn't they keep ISO standards ? What you are saying ISO 100 on FF sensor equals ISO ... 250 on APS sensor ?

What I am saying is that the larger photosites on a full frame sensor will capture more information when exposed to the same amount of light than the smaller photosites of an APS-C sensor, and as a result you will have better dynamic range and less noise at the same aperture settings on a full frame sensor as you will on a APS-C sensor when the two are exposed to the exact same amount of light.

If you'd like you can translate that into an ISO equivalent or an Fstop equivalent to describe how the full frame sensor is more efficient and captures more information using the same amount of light than it's smaller APS-C cousin, if you prefer to think of it in those terms of you certainly can. The thing to note here is that at the same aperture and the same ISO setting and shutter speed the Full Frame is going to have a lot more information to work with to create the final image, which is going to result in a lot less noise and much better dynamic range. Using the exact same settings on an APS-C will result in a lot more noise, so the net result is that the full frame is going to give you much better results in low light conditions.

 

[timor]

What I am saying is that the larger photosites on a full frame sensor will capture more information when exposed to the same amount of light than the smaller photosites of an APS-C sensor, and as a result you will have better dynamic range and less noise at the same aperture settings on a full frame sensor as you will on a APS-C sensor when the two are exposed to the exact same amount of light.

If you'd like you can translate that into an ISO equivalent or an Fstop equivalent to describe how the full frame sensor is more efficient and captures more information using the same amount of light than it's smaller APS-C cousin, if you prefer to think of it in those terms of you certainly can. The thing to note here is that at the same aperture and the same ISO setting and shutter speed the Full Frame is going to have a lot more information to work with to create the final image, which is going to result in a lot less noise and much better dynamic range. Using the exact same settings on an APS-C will result in a lot more noise, so the net result is that the full frame is going to give you much better results in low light conditions.

With this explanation I can agree. Larger format always is able to catch more info in the sense, that it can better resolve between subtle intensities of light. Be it sensor or film. In low light good size sensor (FF or higher) will have a big advantage over the film.
But when you said, that FF sensor in relation to APS is: " your looking at roughly a 1.4 Fstop difference " for me it relates directly to sensitivity (actually for the sensor it will be a lever of amplification).

 

[robbins.photo]

What I am saying is that the larger photosites on a full frame sensor will capture more information when exposed to the same amount of light than the smaller photosites of an APS-C sensor, and as a result you will have better dynamic range and less noise at the same aperture settings on a full frame sensor as you will on a APS-C sensor when the two are exposed to the exact same amount of light.

If you'd like you can translate that into an ISO equivalent or an Fstop equivalent to describe how the full frame sensor is more efficient and captures more information using the same amount of light than it's smaller APS-C cousin, if you prefer to think of it in those terms of you certainly can. The thing to note here is that at the same aperture and the same ISO setting and shutter speed the Full Frame is going to have a lot more information to work with to create the final image, which is going to result in a lot less noise and much better dynamic range. Using the exact same settings on an APS-C will result in a lot more noise, so the net result is that the full frame is going to give you much better results in low light conditions.

With this explanation I can agree. Larger format always is able to catch more info in the sense, that it can better resolve between subtle intensities of light. Be it sensor or film. In low light good size sensor (FF or higher) will have a big advantage over the film.
But when you said, that FF sensor in relation to APS is: "your looking at roughly a 1.4 Fstop difference " for me it relates directly to sensitivity (actually for the sensor it will be a lever of amplification).

Well in terms of noise reduction at high iso you are looking at approximately between 1.2 to 1.4 fstop difference between say a D5100/D5200 crop sensor and a D600 full frame. Meaning that if you were to leave both shutter speed and ISO constant you'd need to shoot at more than a full fstop wider on the lens on the APS-C sensor to get approximately the same level of noise you'd see using the D600 under the same conditions. That's a pretty significant difference in lowlight capabilities and it's easier at least for me to relate it to differences in Fstop.

 

[hamlet]

I mean if the image on my d3200 is less shallow, then it must mean that my 50mm isn't really shooting at f1.4 when i use it at f1.4 on my aps-c camera. I'm a bit confused here about this.

The D600 in that video is sitting closer to the subject to frame it similarly to the D5200, therefore the DOF has been reduced.

Had both cameras been sitting in the same physical location, the DOF would have looked exactly the same.


The amount of light gathered has nothing to do with it--that would change the exposure--it's a factor of the rate of light spread (the distance to the subject, the focal length, and the aperture).

50mm @ f/1.4 is a 35.7mm apertaure opening regardless if it's on a DX or FX sensor.

What about the the objects to the sides? In the d600 they are much more blurry, is that also attributed to being closer or does it have to do with something else?

 

[WayneF]

What about the the objects to the sides? In the d600 they are much more blurry, is that also attributed to being closer or does it have to do with something else?

Your video intentionally did what you had to to show the subject at same size. With the same lens, that meant that DX had to stand back 1.5x farther than FX, so the DX crop would still see the same full view as FX. This is NOT the same thing, because depth of field also depends on distance, and DOF is greater at larger distances, and is less at small distances. You ask why FX shows less DOF, and this is why. It is NOT about exposure.

Or possibly you cropped FX and enlarged it more, to show it same size as DX. Same thing... greater enlargement also reduces apparent DOF.

Repeat your test, with camera on same tripod at same distance for both cameras. Then the same lens at same distance will project the same image. But still not quite the same for both, since DX will be cropped smaller (FX shows a wider view). So to show the images the same size, DX has to be be enlarged more. So don't do that either.

If you do show the frames (same lens at same distance) at their actual size (same enlargment, cropped DX frame is only 2/3 size of FX frame), everything will necessarily look exactly the same. The same lens at same distance cannot do anything different.

 

[KmH]

the shallower your dof becomes because it is letting in more light.

The shallower DoF does not result because of more light. It results from the difference in the point of focus distance.
Note that online DoF calculators have to be told what image sensor size to use making DoF calculations. Most calculators ask for the camera make/model knowing few using the calculator will know the size of the image sensor in their camera.

It isn't often mentioned here on TPF but a full size 135 format (35 mm/full frame/FX/EF) image sensor is a small sensor when compared to medium and large formats.
Consequently DoF can be significantly different using medium and large formats when compared to a FF 35 mm sensor.

 

[timor]

Well in terms of noise reduction at high iso you are looking at approximately between 1.2 to 1.4 fstop difference between say a D5100/D5200 crop sensor and a D600 full frame. Meaning that if you were to leave both shutter speed and ISO constant you'd need to shoot at more than a full fstop wider on the lens on the APS-C sensor to get approximately the same level of noise you'd see using the D600 under the same conditions. That's a pretty significant difference in lowlight capabilities and it's easier at least for me to relate it to differences in Fstop.

In this terms yes. But that refers only to the IQ, not DoF or actual exposure.

 

[timor]

What about the the objects to the sides? In the d600 they are much more blurry, is that also attributed to being closer or does it have to do with something else?

Your video intentionally did what you had to to show the subject at same size. With the same lens, that meant that DX had to stand back 1.5x farther than FX, so the DX crop would still see the same full view as FX. This is NOT the same thing, because depth of field also depends on distance, and DOF is greater at larger distances, and is less at small distances. You ask why FX shows less DOF, and this is why. It is NOT about exposure.

Or possibly you cropped FX and enlarged it more, to show it same size as DX. Same thing... greater enlargement also reduces apparent DOF.

Repeat your test, with camera on same tripod at same distance for both cameras. Then the same lens at same distance will project the same image. But still not quite the same for both, since DX will be cropped smaller (FX shows a wider view). So to show the images the same size, DX has to be be enlarged more. So don't do that either.

If you do show the frames (same lens at same distance) at their actual size (same enlargment, cropped DX frame is only 2/3 size of FX frame), everything will necessarily look exactly the same. The same lens at same distance cannot do anything different.

You nailed it.