So Who Believes that Full Frame Camera's Gather More Light Then APSC

[n614cd]

I always thought it was about pixels.
If you a 35mm with the 20 mega pixel crop sensor and compare it to a 50mm with a 20 mega pixel full frame. The images and resolutions will be identified.

Or am I missing something?

Tim

Sent from my SM-J737T using Tapatalk

 

[greybeard]

I always thought it was about pixels.
If you a 35mm with the 20 mega pixel crop sensor and compare it to a 50mm with a 20 mega pixel full frame. The images and resolutions will be identified.

Or am I missing something?

Tim

Sent from my SM-J737T using Tapatalk

Yes and no. A 20 mp FF file and a 20mp APS-C file should be roughly the same size and contain the same amount of information. Because the APS-C is sampling a smaller image size the lens won't usually be able to resolve as many lines as the FX. If you look at DxO lens testing, you will notice that FX lenses can resolve 25 to 30 MP where as APS-C lenses rarely resolve more than 14mp and in most cases it is like 6-10. And, if you compare the resolution of the same FX lens on APS-C, it will usually show much lower lower resolution numbers. There are darn few lenses on APS-C that can take advantage of the 21-24MP sensors on even entry level dSLR's

 

[n614cd]

I always thought it was about pixels.
If you a 35mm with the 20 mega pixel crop sensor and compare it to a 50mm with a 20 mega pixel full frame. The images and resolutions will be identified.

Or am I missing something?

Tim

Sent from my SM-J737T using Tapatalk

Yes and know. A 20 mp FF file and a 20mp APS-C file should be roughly the same size and contain the same amount of information. Because the APS-C is sampling a smaller image size the lens won't usually be able to resolve as many lines as the FX. If you look at DxO lens testing, you will notice that FX lenses can resolve 25 to 30 MP where as APS-C lenses rarely resolve more than 16mp. And, if you compare the resolution of the same FX lens on APS-C, it will usually show much lower lower resolution numbers. There are darn few lenses on APS-C that can take advantage of the 21-24MP sensors on even entry level dSLR's

Yes. If comparing like lenses. You will notice I compensated for lens in my example.

Tim

Sent from my SM-J737T using Tapatalk

 

[JBPhotog]

Actually it's just the opposite..
every lens crops out a circle of light, if you use a full frame lens on a crop sensor, the crop sensor is using the center of the full frame lens, which is always the best part of that lens,

kind of cutting the crust off of your bread.

Full frame sensors are always using more of that circle of projection,
every lens drops a circle, the bigger the sensor the more of that circle it's using.

The further out it goes the worst it gets,
and it doesn't matter what lens from a cheap lens

or super expensive , like, leica or zeiss,

the further you go out of the center of the circle of projection the worst the image gets,

A rather simplistic deduction that contradicts many lens designs. Do you actually know the image circle of a lens? Hint, it’s not the same for every focal length.

 

[Trever1t]

sounds to me like a classic case of sensor envy.

 

[greybeard]

If you think about it, the lens is what gathers the light and projects it onto a piece of film or sensor. Film through the chemical process of development creates a analog negative of the image. The sensor through multi-millions of pixels and the microprocessors they are connected too, creates a digital description of the image.

 

[Solarflare]

Yes and no. A 20 mp FF file and a 20mp APS-C file should be roughly the same size and contain the same amount of information. Because the APS-C is sampling a smaller image size the lens won't usually be able to resolve as many lines as the FX. If you look at DxO lens testing, you will notice that FX lenses can resolve 25 to 30 MP where as APS-C lenses rarely resolve more than 14mp and in most cases it is like 6-10. And, if you compare the resolution of the same FX lens on APS-C, it will usually show much lower lower resolution numbers. There are darn few lenses on APS-C that can take advantage of the 21-24MP sensors on even entry level dSLR's

Nope.

DxO is not worth discussing. Bunch of ignorant technocrats who only look at their equipment and measurement results but never look at actual images. For example bitingly sharp lenses get called soft, soft lenses get called sharp, all because they are unable to even test something trivial as lens sharpness without doing every n00b error in the book. And thats sharpness, which is relatively easy to test.

And yes DX lenses are usually pretty poor. Thats because of how they are made - cheaply - not because theres something magic about full frame lenses thats suddenly gone with APS-C. Besides one can always use full frame lenses on APS-C sensors, anyway.

And no an APS-C sensor will not record the same information as a full frame sensor. Otherwise why would we bother buying bigger sensors ?

If you offer me a D4s with 16 megapixels and a D7200 with 24 megapixels as a constant free loan (so I cant sell either) I'll pick the D4s. Unless I need the smaller pixels.

 

[Overread]

Bunch of ignorant technocrats who only look at their equipment and measurement results but never look at actual images.

Actually that is a very GOOD thing. If they are testing and comparing multiple lenses against each other then having really boring dull test photos that are the same photo for every lens is far superior for comparison work. You don't need a pretty subject or interesting scene to compare sharpness, edge performance, curvature, edge softness, vignetting etc... Far far far easier to compare it on test shots taken of a test card.

Sample variation can be an issue, some lenses like the Canon 100-400mm original were famous for being quite wide in their variation. Some could be sharper through their entire range, some sharper at the short, some at the long and some soft all through - and all that on the same camera body (to say nothing of variation between camera bodies due to variations in calibration tolerances). So yes sometimes lens they test might under or over perform, but that often gets flagged up and noted. Often good compare their results to viewpoints from places like Lens Rentals - Lens Rentals being a rental company and thus sees a bigger selection of the same lens pass through their hands far more so than many review sites can afford to purchase for testing purposes.

 

[greybeard]

Yes and no. A 20 mp FF file and a 20mp APS-C file should be roughly the same size and contain the same amount of information. Because the APS-C is sampling a smaller image size the lens won't usually be able to resolve as many lines as the FX. If you look at DxO lens testing, you will notice that FX lenses can resolve 25 to 30 MP where as APS-C lenses rarely resolve more than 14mp and in most cases it is like 6-10. And, if you compare the resolution of the same FX lens on APS-C, it will usually show much lower lower resolution numbers. There are darn few lenses on APS-C that can take advantage of the 21-24MP sensors on even entry level dSLR's

Nope.

DxO is not worth discussing. Bunch of ignorant technocrats who only look at their equipment and measurement results but never look at actual images. For example bitingly sharp lenses get called soft, soft lenses get called sharp, all because they are unable to even test something trivial as lens sharpness without doing every n00b error in the book. And thats sharpness, which is relatively easy to test.

And yes DX lenses are usually pretty poor. Thats because of how they are made - cheaply - not because theres something magic about full frame lenses thats suddenly gone with APS-C. Besides one can always use full frame lenses on APS-C sensors, anyway.

And no an APS-C sensor will not record the same information as a full frame sensor. Otherwise why would we bother buying bigger sensors ?

If you offer me a D4s with 16 megapixels and a D7200 with 24 megapixels as a constant free loan (so I cant sell either) I'll pick the D4s. Unless I need the smaller pixels.

Unlike a lot of people, I actually have and use a DX (D7000 and now D7500) and a FF FX (D750) and I use FX lenses with my DX body all the time. I also have and use several upper shelf DX lenses and their build quality is excellent. I like DxO because they test and compare lenses on several different bodies and give us a read out of how a lens technically performs. A FX lens will always have about half its' FF resolution on APS-C and the reason is simple, the sensor is only recording the center portion of its' FF image. Now, a high quality DX lens will only produce an image as large as the APS-C sensor and, its' resolution will be similar to the FX on APS-C but will be smaller and less expensive.

 

[TreeofLifeStairs]

Bunch of ignorant technocrats who only look at their equipment and measurement results but never look at actual images.

Actually that is a very GOOD thing. If they are testing and comparing multiple lenses against each other then having really boring dull test photos that are the same photo for every lens is far superior for comparison work. You don't need a pretty subject or interesting scene to compare sharpness, edge performance, curvature, edge softness, vignetting etc... Far far far easier to compare it on test shots taken of a test card.

Sample variation can be an issue, some lenses like the Canon 100-400mm original were famous for being quite wide in their variation. Some could be sharper through their entire range, some sharper at the short, some at the long and some soft all through - and all that on the same camera body (to say nothing of variation between camera bodies due to variations in calibration tolerances). So yes sometimes lens they test might under or over perform, but that often gets flagged up and noted. Often good compare their results to viewpoints from places like Lens Rentals - Lens Rentals being a rental company and thus sees a bigger selection of the same lens pass through their hands far more so than many review sites can afford to purchase for testing purposes.

I agree. DxO, IMO, does a good job of taking as much subjectivity out of their tests as possible. This is good for technical comparisons, but there are many subjective nuances that lenses and cameras have that are not testable. These may be desirable or not. You just need to know what DxO is good for and what it’s not.


Sent from my iPhone using ThePhotoForum.com mobile app