Does FF body have advantage at low light?

I am asking this because I had been thinking about this for a while and reached a conclusion that it is not the body which has the advantage but the lens, so I invite people who have both FF and corp body to do a comparison shot.

To have the same angle of view and same DOF, it is required FF body have focal length of 1.6X and aperture of f#X1.6. Example 50mm/f5.6 on FF body and 31mm/f3.5 or closest one on crop body.

The theory is that for same angle of view and DOF(which leads to same aperture area), the total light entering lens is the same which is irrelevant to sensor size. Although FF allow more exposure time to reach maximum exposure, that is not available in low light, fast moving object shooting(high ISO setting). People mostly compare low light result at same ISO setting, but that will require more exposure time on FF sensor than a crop sensor(roughly 2.6 times for 1.6X crop factor) to achieve same DOF, at least in theory(because I do not have a FF body to verify it). In my mind, low light shooting should be conducted under same exposure time for the same lighting condition.

This is not to prove FF system does not have advantage, because the lens available today give larger maximum aperture area for the same angle of view on FF than a crop body. I am just trying to verify the system advantage comes from lens not bodies.

What happens when I shoot my FF 50mm at f1.2? Which is what most people would do for low light photography. To achieve the same DOF, one would have to employ a 31mm/0.75 lens on a crop according to your theory. I don't think take make one of those yet. Also, you are completely ignoring ISO and shutter speed which is part of the exposure calculations (aperture, shutter, sensitivity) and significantly contribute to capturing an image, especially in low light conditions.

You are correct. The alleged superiority of FF in low light is a By-Product of using the ISO definition os sensitivity.

That is, light flux per unit area.

DxOmark has low light sensitivity values for various cameras. If you square the crop factor, to get an Area Factor, and renormalize the published ISO sensitivities, you will be amazed to see little difference in these Renormalized Sensitivities.

Instead of maths formulae, take an example, two lenses with the sam field of view, one for APS-C and the other for FF.

If you set both at the same f-number, then the intensity on the sensor is the same. But the AMOUNT of light, the Total Flux falling on the FF is greater by the ratio of Area-Factors, as defined above ( by me ).

The Olympus E3 has a sensor with the same sensitivity to flus PER UNIT AREA as the renown Nikon D3.
By popular opinion says it has poor low light sensitivity.

But NO, the entrance pupil is larger for the FF lens, even thouigh the F-number is the same.

I omit a rigorous derivation, since previous explanations by me on this Forum have gone mostly ignored. I have lost patience, and now leave the proof as an exercise , to be applied to DxOmark data.

I feel most posters could do this HomeWork themselves, and learn the maths skills.
And gain some insight themselves.

FF is generally better in low light unless you need a certain DOF.

nt
--
Slainte (cheers)

You cannot depend on your eyes when your imagination is out of focus. :- Mark Twain

Just my opinion...in low light all you have to do is raise the ISO to 3200...between a 1dMark III and the 5dMark II....it's pretty much the same.
--
http://www.juliussantos.com

Sounds very interesting... Would be cool to see more test results to further explain this!

--
Camillo
http://www.cam-photo.com

well sometimes to freeze motion you need more than nay sort of 1/FL so that becomes irrelevant (andthe 1/FL depends upon sensor density anyway if you want all photosites resolved well and if you just want percentage of frame image sharp then FF does not hurt that at all)

if you are shooting like 24mm or 35mm and want to stop someone moving around that takes like 1/100th maybe regardless or for sports you might need 1/640th or more

sometimes you don't need the extra DOF

it depends, if you wanted P&S DOF you might need to stop FF down a LOT

in general practice FF generally works out better at high ISO in terms of SNR and DR, but it might not always

and with a 20D and a 5D2 the 5D2 can always take the same pic by cropping the center and can never do worse

Liwei Sheng said:

I am asking this because I had been thinking about this for a while and reached a conclusion that it is not the body which has the advantage but the lens, so I invite people who have both FF and corp body to do a comparison shot.

To have the same angle of view and same DOF, it is required FF body have focal length of 1.6X and aperture of f#X1.6. Example 50mm/f5.6 on FF body and 31mm/f3.5 or closest one on crop body.

The theory is that for same angle of view and DOF(which leads to same aperture area), the total light entering lens is the same which is irrelevant to sensor size. Although FF allow more exposure time to reach maximum exposure, that is not available in low light, fast moving object shooting(high ISO setting). People mostly compare low light result at same ISO setting, but that will require more exposure time on FF sensor than a crop sensor(roughly 2.6 times for 1.6X crop factor) to achieve same DOF, at least in theory(because I do not have a FF body to verify it). In my mind, low light shooting should be conducted under same exposure time for the same lighting condition.

This is not to prove FF system does not have advantage, because the lens available today give larger maximum aperture area for the same angle of view on FF than a crop body. I am just trying to verify the system advantage comes from lens not bodies.

Click to expand...

If you have a FF camera and a cropped camera, the photon noise will be less in a given image because with equal settings, the larger sensor will capture more light proportional to the larger size of the sensor. Whether or not this impacts a final picture or print depends on the picture content, size, viewing distance, processing, etc.

Being able to gather more light tends to be more valuable as light decreases but, again, the devil is in the details.
--
Leon
http://web.me.com/leonwittwer/landscapes.htm

OP is right, it's not about the sensor, but the lens. It is all determined by the amount of light coming in, which is simplistically-speaking proportional to the area of the front element of the lens (or smaller if you stop down the aperture).

If you require specific DOF, you have to stop down the lens more for FF compared to Crop. If you go through the math, turns out the physical lens opening is about the same on FF and crop in this case, if you keep DOF requirement constant!

This is actually a very good rule of thumb: for similar FOV and DOF, the amount of light is simply determined by the DIAMETER or the lens (I did the math, got a PhD in Optical Physics), and will give you similar noise level on any camera. Alternatively, larger lens diameters give you smaller DOF. This is why cell phone cameras have huge DOF, P&S medium, Crop cameras smaller, FF could be thin, and large format can give you razor-thin DOF, but requires big lenses.

Look at it from a very practical angle. If you have FF camera with 24-105 F/4 lens, it's a waste of money. Just get a crop camera with 17-55 F/2.8 lens, same result. It's only worth going to FF if you get 24-70 F/2.8 lens (though it's not stabilized). You can shoot in lower light with this lens simply because it's got larger front element (this lens is huge!). Or get 50/1.2 and be prepared for super-thin DOF. You simply cannot get equivalent lens for crop sensor, as earler posted pointed out.

of course lots of sport and PJ work is done shooting wide open say sports with a 300 2.8 or 70-200 2.8 or 400 2.8 and DOF is ok on any DSLR sensor size and the shutter speed is already at 1/640th or more so there you do gain by going up in sensor size for sure

EDIT: actually for the 70-200 work where you are not reach limited yes

for the longer work, maybe not since you may be cropping in more with the FF and ending up with an identical result, it depends

Yes...but....the 1dmk3 is better in low light than the old 5D......
--
Slainte (cheers)

You cannot depend on your eyes when your imagination is out of focus. :- Mark Twain

Liwei Sheng said:

Liwei Sheng wrote:

I am asking this because I had been thinking about this for a while and reached a conclusion that it is not the body which has the advantage but the lens, so I invite people who have both FF and corp body to do a comparison shot.

Click to expand...

It's both the lens and the sensor. The lenses for FF often have larger max apertures for a given AOV, so they can put more light on the sensor for a given shutter speed. Since the sensor is larger, it can absorb more light before oversaturating.

Liwei Sheng said:

To have the same angle of view and same DOF, it is required FF body have focal length of 1.6X and aperture of f#X1.6. Example 50mm/f5.6 on FF body and 31mm/f3.5 or closest one on crop body.

Click to expand...

Yes.

Liwei Sheng said:

The theory is that for same angle of view and DOF(which leads to same aperture area), the total light entering lens is the same which is irrelevant to sensor size. Although FF allow more exposure time to reach maximum exposure, that is not available in low light, fast moving object shooting(high ISO setting). People mostly compare low light result at same ISO setting, but that will require more exposure time on FF sensor than a crop sensor(roughly 2.6 times for 1.6X crop factor) to achieve same DOF, at least in theory(because I do not have a FF body to verify it). In my mind, low light shooting should be conducted under same exposure time for the same lighting condition.

Click to expand...

For the same perspective, framing, DOF, and shutter speed, FF has very little advantage over crop, and is sometimes at a disadvantage.

Liwei Sheng said:

This is not to prove FF system does not have advantage, because the lens available today give larger maximum aperture area for the same angle of view on FF than a crop body. I am just trying to verify the system advantage comes from lens not bodies.

Click to expand...

Like I said, it comes from both. The lenses usually have larger maximum apertures for a given AOV, and the larger sensor not only can absorb more light, but, for a given generation camera, has more pixels as well. In addition, the larger sensor will stress the lenses less than a smaller sensor.

But if you compare at the same perspective, framing, DOF, and shutter speed, FF won't have much, if any, advantage.

Worth a read:

http://www.josephjamesphotography.com/equivalence/

Obviously, you do better with lower fstops/wider apertures. Given the same f number and shutter speed in two cameras, the one with the larger sensor will capture more light. A FF body will capture about 1.3 stops more light than a 1.6 cropped body with other things the same.
--
Leon
http://web.me.com/leonwittwer/landscapes.htm

Cheburashka said:

But if you compare at the same perspective, framing, DOF, and shutter speed, FF won't have much, if any, advantage.

Click to expand...

There can be a difference when you can not get a lens for a cropped body that has the smaller f number. For example, if I shoot with my 24 mm f1.4 lens, a 1.6 cropped body would have to have an f0.875 lense to match DoF, diffraction, perspective, etc. The main difference relates to equipment availability at thin DoF.

--
Leon
http://web.me.com/leonwittwer/landscapes.htm

Not in theory.

The whole light gathering ability is not down to total area of the sensor and all down to the light gathering ability of the individual pixels. The only reason that the current FF models are better in low light is that their individual pixels are larger and can gather more light.

If you take the 50d sensor (pixel size and density) and expand it to the area of FF sensor you will not have any better light gathering capability.

Robbie
--
Canon Can...Can you??

nothsafoto said:

If you take the 50d sensor (pixel size and density) and expand it to the area of FF sensor you will not have any better light gathering capability.

Click to expand...

But of course I will. The 50D has 15.1 MP. A FF with the same pixel size would have 38.9 MP. For any given print size the FF pixels would be scaled much smaller, resulting in less visible noise.

And, as another posters have said: there is nothing in the crop world that would match FF lenses like 24/1.4 (equiv to 15mm f/0.9), 35/1.4 (22mm f/0.9), 50/1.2 (31mm f/0.75) or 85/1.2 (53mm f/0.75). The whole reason for me to buy into the DSLR world four years ago was the northern light opportunities offered by the combination of an affordable FF camera (the 5D) and the 24/1.4L.

Kind regards,

• Henrik

--
And if a million more agree there ain't no great society
My obligatory gallery: http://www.iki.fi/leopold/Photo/Galleria/

Depends on pixel density and image
processor, not just the size of sensor.
However given all other things being
equal, the larger sensor will have
less noise and allow higher usable
ISO in low light.

--



Matt Cham
http://www.mattcham.com

...but compared to a cropped body with the same pixel count and same technology, pixels can be larger on the 35mm-format sensor, resulting in higher signal-to-noise ratio (i.e., cleaner images) at all ISO sensitivities.

Liwei Sheng said:

I am asking this because I had been thinking about this for a while and reached a conclusion that it is not the body which has the advantage but the lens, so I invite people who have both FF and corp body to do a comparison shot.

To have the same angle of view and same DOF, it is required FF body have focal length of 1.6X and aperture of f#X1.6. Example 50mm/f5.6 on FF body and 31mm/f3.5 or closest one on crop body.

The theory is that for same angle of view and DOF(which leads to same aperture area), the total light entering lens is the same which is irrelevant to sensor size. Although FF allow more exposure time to reach maximum exposure, that is not available in low light, fast moving object shooting(high ISO setting). People mostly compare low light result at same ISO setting, but that will require more exposure time on FF sensor than a crop sensor(roughly 2.6 times for 1.6X crop factor) to achieve same DOF, at least in theory(because I do not have a FF body to verify it). In my mind, low light shooting should be conducted under same exposure time for the same lighting condition.

This is not to prove FF system does not have advantage, because the lens available today give larger maximum aperture area for the same angle of view on FF than a crop body. I am just trying to verify the system advantage comes from lens not bodies.

Click to expand...

--
- -
Kabe Luna

http://www.garlandcary.com

Cheburashka said:

Worth a read:

http://www.josephjamesphotography.com/equivalence/

Click to expand...

Thanks for the link, the links explain a lot of things, everybody who is interested in the topic should read it and it make discussion more efficient. And I was wondering why the lens do not indicate angle of view(AOV) and absolute Aperture size, that way no mater what kind of camera you use, the picture taken always looks alike at the same AOV and physical aperture size. Only bad is that exposure time will change when using different format, but there always be a meter to do the measurement. Comparing lens will be more straight forward, just angle of view and aperture size for same equivalent picture effect one want to achieve.