Answer this one question!

A FF sensor has 2.56 (1.6 ^ 2) times the area of a 1.6x sensor.
Correct.
Thus, it collects 2.56 times as much light for the same FOV
(framing) and f-ratio.
No. Sensor board on a FF camera measures 24mmx36mm. Sensor board on a 1.6x camera measures 16.7mmx25mm. With an EF 20mm lens infront of your FF camera, the light fall onto 24mmx36mm area. With a 1.6x camera, the light also fall onto 24mmx36mm area, but the sensor board is only able to see 16.7x25mm area, the rest of the light falls onto other part of the camera. Light density per area on both FF and 1.6x is the same. You cannot say there are more light per area density on FF compare to 1.6x.

For example, to get the same framing, if we
used a 100mm lens on a 1.6x camera, we'd use a 160mm lens on a FF
camera. If we used f / 4 on both cameras, the aperture on the 1.6x
camera would be 100mm / 4 = 25mm and the aperture on the FF camera
would be 160mm / 4 = 40mm. A circle with a diameter of 40mm has
2.56x the area of a circle with a 25mm diameter, which is why the
FF sensor collects more light for the equivalent FL and the same
f-ratio.

Now, since the intensity of the light on the sensor is the same,
it will meter the same, but since the total amount of light is
more, the noise will be less. However, by using the same f-ratio
for the same framing, you will also have 1.6x less DOF, and thus
the images will not be equivalent.

To get an equivalent image, we up the f-ratio by a factor of 1.6 (1
1/3 stops) to get the same aperture, anc consequently, same DOF,
and same total light. However, that light is spread over 2.56
times as much area, so the intensity of that light is 2.56 times
less. Thus, you must up the ISO by a factor of 2.56 to get the
same exposure (or, alternatively, reduce the shutter speed by a
factor of 1.6, or some equivalent comination of ISO and shutter
speed).

So, for example, if I take a pic with a 1.6x camera at 100mm, f /
2.8, 1/100, ISO 100, I will get an equivalent image with a FF
camera taken of the same scene from the same location at 160mm, f /
4.5, 1/100, ISO 250. The two pics will have the same FOV, same
DOF, same exposure, and same noise.

Now, if the total number of pixels is the same , then the noise
will be the same on a per-pixel basis as well. However, FF sensors
have more pixels, thus the per-pixel noise is higher, but the
total image noise is the same. So, the FF sensor gives a more
noisy, but more detailed, picture. However, if you then downsample
the FF image to the same dimensions as the 1.6x image, then you
will now regain noise parity, but will also have lost the detail
advantage with it. Well, not entirely -- since the lens on the FF
camera was stopped down, it will have a sharper image to being
with, regardless of pixel count, so the downsampled FF image will
still have slightly more detail.

See this thread for pics that illustrate this point:

http://forums.dpreview.com/forums/read.asp?forum=1029&message=23287156

I hope this answers your question.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

Please feel free to criticize, make suggestions, and edit my
photos. If you wish to use any of my photos for any purpose other
than editing in these forums, please ask.
 
Please read to the end before you answer.
I don't understand your statement:
So, to get the same exposure, you either have to reduce the shutter
speed by a factor of 1.6 or multiply the ISO by a factor of 2.56, or
some equivalent inbetween combination.
Please explain.
A FF sensor has 2.56 (1.6 ^ 2) times the area of a 1.6x sensor.
Thus, it collects 2.56 times as much light for the same FOV
(framing) and f-ratio. For example, to get the same framing, if we
used a 100mm lens on a 1.6x camera, we'd use a 160mm lens on a FF
camera. If we used f / 4 on both cameras, the aperture on the 1.6x
camera would be 100mm / 4 = 25mm and the aperture on the FF camera
would be 160mm / 4 = 40mm. A circle with a diameter of 40mm has
2.56x the area of a circle with a 25mm diameter, which is why the
FF sensor collects more light for the equivalent FL and the same
f-ratio.
True so far.
Now, since the intensity of the light on the sensor is the same,
it will meter the same, but since the total amount of light is
more, the noise will be less.
Not true. 1Ds has the same light gathering capability as 5D and 1Ds2 and more than 30D, 400D yet it has more noise than any of these cameras.

However, by using the same f-ratio
for the same framing, you will also have 1.6x less DOF, and thus
the images will not be equivalent.
True.
To get an equivalent image, we up the f-ratio by a factor of 1.6 (1
1/3 stops) to get the same aperture, anc consequently, same DOF,
and same total light. However, that light is spread over 2.56
times as much area, so the intensity of that light is 2.56 times
less.
True.

Thus, you must up the ISO by a factor of 2.56 to get the
same exposure (or, alternatively, reduce the shutter speed by a
factor of 1.6, or some equivalent comination of ISO and shutter
speed).
Not true. There is a small sensor inside the camera that determines the exposure in the camera not the size of the main sensor. Your statement above contradicts your answer to me about the gray card.
So, for example, if I take a pic with a 1.6x camera at 100mm, f /
2.8, 1/100, ISO 100, I will get an equivalent image with a FF
camera taken of the same scene from the same location at 160mm, f /
4.5, 1/100, ISO 250. The two pics will have the same FOV, same
DOF, same exposure, and same noise.
True, but you also changed f/stop. I think I know at this point what you are trying to say you just need to rewrite your statement above. To include that 2 things need to be changed not just one.

For example putting camera in program mode and changing F stop from 2.8 to 4.5 will give different shutter speed but DOF will be the same as on the 1.6x crop camera.
Now, if the total number of pixels is the same , then the noise
will be the same on a per-pixel basis as well. However, FF sensors
have more pixels, thus the per-pixel noise is higher, but the
total image noise is the same. So, the FF sensor gives a more
noisy, but more detailed, picture.
Not true. It depends on the camera.

However, if you then downsample
the FF image to the same dimensions as the 1.6x image, then you
will now regain noise parity, but will also have lost the detail
advantage with it. Well, not entirely -- since the lens on the FF
camera was stopped down, it will have a sharper image to being
with, regardless of pixel count, so the downsampled FF image will
still have slightly more detail.
True for every camera. But only regarding DOF.
See this thread for pics that illustrate this point:

http://forums.dpreview.com/forums/read.asp?forum=1029&message=23287156

I hope this answers your question.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

Please feel free to criticize, make suggestions, and edit my
photos. If you wish to use any of my photos for any purpose other
than editing in these forums, please ask.
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

 
that with FF camera you can achieve maximum shallow DOF (that is why I like FF) but with crop factor camera you can achieve the maximum DOF with the same lens and sometimes I missed that.

--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

 
The light gathering ability of a camera for a given scene depends
only on the size of the aperture and how much of that light is
captured by the sensor. The aperture is determined by the FL
divided by the f-ratio.

For example, let's say we're at 24mm, f / 2.8 on a 1.6x DSLR. Most
of the light from the image circle from the lens falls outside the
sensor. The amount of light that does fall on the sensor is the
same amount of light as would fall on a FF sensor with a FL of 38mm
and an f-ratio of f / 4.5.
Please allow me to make an analogy using pieces of white card, not sensors, for the time being to simplify the situation and remove the need for calculation

One is 100cm square

The other is 50cm square

If you have a light source that covers 100cm square, the first card is touched by all the light

The second card captures only half as much light though - with the rest of the light falling outside of its area

I agree with this, and believe this to be the same as what you state above
Thus, both sensors collect the same amount of light.
I don't see how this is possible.

Surely one card is collecting 100cm square amount of light, and the other is collecting 50cm square with the rest being wasted (i.e. falling on darkness)

Surely the 50cm square card therefore collects half as much light???
However, the
FF sensor has 2.56 times the area as the 1.6x sensor, thus the
intensity of that light is 2.56 times less.
I also don't see how this is true

If you have one light source, the intensity of the light does not change.

Just because the 50cm square card is smaller, it does not mean the light intensifies ???
So, to get the same
exposure, you either have to reduce the shutter speed by a factor
of 1.6 or multiply the ISO by a factor of 2.56, or some equivalent
inbetween combination.
Because i don't understand the above, I do not understand how this works
But since both sensors receive the same amount of light, the noise
will be the same for both images.
As before... I don't see how a smaller sensor receives the same amount of light

I also believe 'noise' is not just related to the sensor size, but also the size of the pixels on the sensor

The only way I can see how a smaller sensor receives the same amount of light is if the light is concentrated onto a smaller area

So with the card example, the 50cm card would only receive the same amount of light as the 100cm card if there was a modifying device in front of the card that channels the light source into half the space

If this is true, then I agree that both cards receive the same amount of light, and that the larger card therefore has "less intense" light

However, what is it in a smaller sensor camera that does this?
The reason you have those options is that the FF sensor has more
pixels than the 1.6x sensor. If the number of pixels were the
same, you'd always get the same detail, just with 1 1/3 stops less
noise. But, since the 5D has 50% more pixels, you only get a 2/3
stop noise improvement for the same ISO. When you up the f-ratio
and ISO to get the same DOF, shutter speed, and exposure, the 5D
image will actually have 2/3 stops more noise, along with its 50%
more pixels. However, if you then resample that image to the same
dimensions as the 1.6x image, you will have lost the detail
advantage, but regained the noise equivalency.

Does that make sense?
Do you have any images to show the above in a practical sense? i.e. the same scene with the 5D and a smaller sensor camera

I know all about the framing issue / 1.6x magnification factor or whatever you want to call it, it's just the light-capturing theory that I wasn't aware of

--
-----
Neil C
http://www.homelands.me.uk/gallery/
 
For example aperture of F2.8 gathers twice as much light as F4 even though
4 divided by 2.8 is not 2. But 4 square divided by 2.8 square equals to 2.
FF sensor is 2.56 times larger than 1.6x sensor. 1.6 square is 2.56.
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

 
--
Eugene

The only time a smaller sensor with the same pixel count is superior to a larger sensor (aka higher pixel density) is when you are focal-length limited.

Lee Jay

 
Sorry, I just don't get it especially when he said FF body gather 2.56 times more light than a 1.6x equivalent, hence a 2.56 stops advantage over 1.6x crop. I still struggle to understand where the 2.56 stop advantage comes from.
For example aperture of F2.8 gathers twice as much light as F4 even
though
4 divided by 2.8 is not 2. But 4 square divided by 2.8 square
equals to 2.
FF sensor is 2.56 times larger than 1.6x sensor. 1.6 square is 2.56.
--
Eugene

The only time a smaller sensor with the same pixel count is
superior to a larger sensor (aka higher pixel density) is when you
are focal-length limited.

Lee Jay

 
Yeap, I think Neil has the same understanding as me. Just that he is better at explaining himself than me. I still struggle to understand how a FF body has 2.56 stop advantage over an equilvalent 1.6x cropped camera.
 
Yeap, I think Neil has the same understanding as me. Just that he
is better at explaining himself than me. I still struggle to
understand how a FF body has 2.56 stop advantage over an
equilvalent 1.6x cropped camera.
2.56x advantage = 1 1/3 stop advantage.

Because the sensor is 2.56x larger and at the same f-stop, it's getting the same light per unit area. 2.56x the area, times the same light per unit area = 2.56x more light.

--
Lee Jay
(see profile for equipment)
 
He can take a pic of a scene at 38mm, f / 4.5 with his 5D, and then
take a pic of the exact same scene from the exact same position
with his 20D at 24mm, f / 2.8, and post the pics. They will be
identical in terms of perspective, FOV, and DOF.
I never argued with that. You conviniently "forgot" to mention the
fact that 5D will require either longer exposure or higher ISO and
that's my point: 24/2.8 on 1.6x is equivalent of 38/2.8 on FF in
everything except DOF.
And image quality. It'll be noisier because it's on a smaller sensor at the same f-stop, and therefore it's gathering only 39% (1/2.56) as much light.
You are trying to mislead the people by
saying all the time that 135/2 on FF is equivalent not 85/2 on
1.6x, but 85/1.2. No, it is not, this equivalence is in DOF, but
not in exposure.
The equivalence is in DOF, image quality, diffraction effects, and exposure which includes ISO.

--
Lee Jay
(see profile for equipment)
 
Yeap, I think Neil has the same understanding as me. Just that he
is better at explaining himself than me. I still struggle to
understand how a FF body has 2.56 stop advantage over an
equilvalent 1.6x cropped camera.
2.56x advantage = 1 1/3 stop advantage.
yes, ok, but...
Because the sensor is 2.56x larger and at the same f-stop, it's
getting the same light per unit area. 2.56x the area, times the
same light per unit area = 2.56x more light.
yes, overall it gets more light - but that's only because it's bigger

as i said before, how can a smaller sensor get more intense light, which would equate to a difference in exposure

imagine you have a white card. shine a torch at it. if the torch covers the whole card, then the the whole card has (for arguments sake) 1unit of light power at every point

now cut the card in half

the card is now 50% in size, and the torch hasn't changed

the card is still receiving 1unit of light power at every point, but there is also 'wasted' light where the card area has been cut away

the light is not more intense

the only way the light can be more intense is if there is some kind of modifier that focuses all the light that was on the 100% card onto the card when it is half the size

therefore a card half the size would receive 2units of light power at every point

surely just because the sensor is bigger doesn't mean the image will be brighter. it just means that more of the image will be seen (i.e. no "crop factor" on focal length)

personally i suspect it has something to do with the pixels being larger, which makes them better 'capturers' of light. however are the pixels proportionately the same size compared to the sensor on each camera?

e.g. on a piece of card 100cm square the pixels are 1cm square

on a 50cm square piece of card are they 0.5cm square?

because if the pixels are larger, they are able to capture more light - therefore it has nothing to do with sensor size per se, just pixel size

although this is just a whim - i may be wrong

--
-----
Neil C
http://www.homelands.me.uk/gallery/
 
Yeap, I think Neil has the same understanding as me. Just that he
is better at explaining himself than me. I still struggle to
understand how a FF body has 2.56 stop advantage over an
equilvalent 1.6x cropped camera.
2.56x advantage = 1 1/3 stop advantage.
yes, ok, but...
Because the sensor is 2.56x larger and at the same f-stop, it's
getting the same light per unit area. 2.56x the area, times the
same light per unit area = 2.56x more light.
yes, overall it gets more light - but that's only because it's bigger

as i said before, how can a smaller sensor get more intense
light, which would equate to a difference in exposure
Actually Joe meant the bigger sensor get more intense light (2.56 stop advantage, less exposure required) on the FF. That is what I still struglle to understand.
imagine you have a white card. shine a torch at it. if the torch
covers the whole card, then the the whole card has (for arguments
sake) 1unit of light power at every point

now cut the card in half

the card is now 50% in size, and the torch hasn't changed

the card is still receiving 1unit of light power at every point,
but there is also 'wasted' light where the card area has been cut
away

the light is not more intense

the only way the light can be more intense is if there is some kind
of modifier that focuses all the light that was on the 100% card
onto the card when it is half the size

therefore a card half the size would receive 2units of light power
at every point
Actually that will mean overexposure on the smaller cardboard. But I think he meant there are more intensity on the bigger board. Which I couldn't understand too.
surely just because the sensor is bigger doesn't mean the image
will be brighter. it just means that more of the image will be
seen (i.e. no "crop factor" on focal length)

personally i suspect it has something to do with the pixels being
larger, which makes them better 'capturers' of light. however are
the pixels proportionately the same size compared to the sensor on
each camera?

e.g. on a piece of card 100cm square the pixels are 1cm square

on a 50cm square piece of card are they 0.5cm square?

because if the pixels are larger, they are able to capture more
light - therefore it has nothing to do with sensor size per se,
just pixel size

although this is just a whim - i may be wrong

--
-----
Neil C
http://www.homelands.me.uk/gallery/
 
right, i believe i have found the answer

it is kind of to do with the size of the sensor, but not really. the way it was being explained (i.e. larger sensor = more light = brighter) is rubbish

based on these comments:

"As to aperture ratings - a given f/stop ("f/2" or whatever) delivers the same light intensity to the film/sensor surface in all cases.

A 10mm f/2 lens delivers the same brightness per square mm as a 50mm f/2 lens. But it has to cover a much smaller area (fewer square mm), so it can use a smaller absolute opening (10mm/2 = 5mm vs. 50mm/2 = 25mm).

Which is a long way of saying - if the meter calls for 1/250th sec. @ f/2 with an ISO100-speed film/sensor - you'll get the same exposure @ 1/250th with a 10mm f/2 lens on a chip, or a 50mm f/2 lens on 35mm film, or a 100mm f/2 lens on 6x9 film."

from http://photo.net/bboard/q-and-a-fetch-msg?msg_id=007FMI

and

"T-stop.

Some lenses, mainly those used in motion cinematography and not still photography, have apertures rated in T-stops. T-stops indicate the actual or absolute amount of light being passed through (transmitted) the aperture diaphragm. They differ therefore from f-stops, which indicate the aperture size relative to the focal length. (in a sense, T-stops are absolute and f-stops are relative)

T-stops can be an important concept with certain zoom lenses in particular, since zooms tend to result in some light loss, though this is not a huge issue for most modern multicoated lenses. Lenses with T-stop markings usually have the f-stops marked in white and the T-stops in red."

http://photonotes.org/cgi-bin/entry.pl?id=Tstop

i get the gist that...

It does not matter how big the sensor is per se. What matters is the size of the aperture in relation to the size of the sensor.

Quite frankly, i'm going to bed and am not going to look into it further right now, but I now start to understand why sensor size makes a difference...

it's because sensor size changes the field of view with a given lens, which in turn will require a different focal length of the lens, which in turn will require a different aperture to maintain the same exposure (i.e. amount of absolute light hitting the same absolute size on the sensor)

so it most certainly isn't "bigger sensor = more light".

i think joe would have been able to explain had he have seen the message as i now understand what he was trying to say about intensity... it's all because "f-stop" is NOT an absolute value for the size of the aperture

--
-----
Neil C
http://www.homelands.me.uk/gallery/
 
A FF sensor will collect 2.56 times more light at 38mm (24mm x 1.6)
f / 2.8 than a 1.6x sensor will collect at 24mm f / 2.8.
I thinkl you are having problem to understand the basics of exposure.
F/2.8 any focal length lens create the same light intensity on any
size sensor.
He did not say intensity.
He said "more", ie the amount not the amount per unit area.

Andrew
 
It does not matter how big the sensor is per se. What matters is
the size of the aperture in relation to the size of the sensor.
Exactly.
Quite frankly, i'm going to bed and am not going to look into it
further right now, but I now start to understand why sensor size
makes a difference...
Excellent!
it's because sensor size changes the field of view with a given
lens, which in turn will require a different focal length of the
lens, which in turn will require a different aperture to maintain
the same exposure (i.e. amount of absolute light hitting the same
absolute size on the sensor)
Yes, yes, and yes. : )
so it most certainly isn't "bigger sensor = more light".
Well, you have to put it into context, and I most certainly did:

http://forums.dpreview.com/forums/read.asp?forum=1029&message=23300792

If I may quote the opening sentence from the above link:
A FF sensor has 2.56 (1.6 ^ 2) times the area of a 1.6x sensor. Thus,
it collects 2.56 times as much light for the same FOV (framing) and f-
ratio.
it's all because "f-stop" is NOT an absolute value for the size of the
aperture
That is the critical point that so many seem to miss. However, intensity is important in terms of the function of the AF sensors. Thus, while f / 4.5 on FF is equivalent to f / 2.8 on 1.6x for equivalent FLs (that is, 1.6 times the FL on FF as on 1.6x), an f / 2.8 lens on 1.6x will deliver a greater intensity of light for the AF sensors than an f / 4 lens on FF, even though the overall images would collect the same light.

However, as long as both lenses have the same min f-ratio, both will AF the same regardless of the f-ratio used for the actual pic, since lenses AF wide open.

Thus, while a 24-105 / 4L IS on FF is equivalent to a 15-65 / 2.5 IS on 1.6x, the faster 1.6x lens will have better AF in low light.

And, yes, before the "do not get it" crowd come in, again, yes, the above equivalency is valid. On occasion, you will have to up the ISO on the FF camera to get the same shutter, but even then, the FF image will have the same noise as the 1.6x image even with the increased ISO after it is resampled to the same dimensions as the 1.6x image. For the times you do not need to up the ISO to get the same shutter speed, the FF image will have more detail and less noise.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

Please feel free to criticize, make suggestions, and edit my photos. If you wish to use any of my photos for any purpose other than editing in these forums, please ask.
 
Please read to the end before you answer.
Is this a test? : )
Now, since the intensity of the light on the sensor is the same,
it will meter the same, but since the total amount of light is
more, the noise will be less.
Not true. 1Ds has the same light gathering capability as 5D and
1Ds2 and more than 30D, 400D yet it has more noise than any of
these cameras.
Eugene, obviously, I am talking about sensors of the same design and generation. I mean, the new 1DIII has less noise than the 5D sensor as well. It's a new generation of sensor.

Likewise, FF has even more than a 1 1/3 stop noise advantage compared to Nikon sensors, and more than a 2 stop advantage compared to Olympus sensors. That's a whole other argument.
Thus, you must up the ISO by a factor of 2.56 to get the
same exposure (or, alternatively, reduce the shutter speed by a
factor of 1.6, or some equivalent comination of ISO and shutter
speed).
Not true. There is a small sensor inside the camera that determines
the exposure in the camera not the size of the main sensor. Your
statement above contradicts your answer to me about the gray card.
Not "not true". : ) If you shoot f / 2.8, ISO 100 on 1.6x and I shoot f / 4.5, ISO 250 on FF, we will get the same shutter speed and exposure.
So, for example, if I take a pic with a 1.6x camera at 100mm, f /
2.8, 1/100, ISO 100, I will get an equivalent image with a FF
camera taken of the same scene from the same location at 160mm, f /
4.5, 1/100, ISO 250. The two pics will have the same FOV, same
DOF, same exposure, and same noise.
True, but you also changed f/stop. I think I know at this point
what you are trying to say you just need to rewrite your statement
above. To include that 2 things need to be changed not just one.
For example putting camera in program mode and changing F stop from
2.8 to 4.5 will give different shutter speed but DOF will be the
same as on the 1.6x crop camera.
Again, you must up the ISO 1 1/3 stops as well. Then the shutter speeds will be the same.
Now, if the total number of pixels is the same , then the noise
will be the same on a per-pixel basis as well. However, FF sensors
have more pixels, thus the per-pixel noise is higher, but the
total image noise is the same. So, the FF sensor gives a more
noisy, but more detailed, picture.
Not true. It depends on the camera.
Again, I just thought that was kind of obvious that I was talking about sensors of the same design and generation.

--
--joe

http://www.josephjamesphotography.com
http://www.pbase.com/joemama/

Please feel free to criticize, make suggestions, and edit my photos. If you wish to use any of my photos for any purpose other than editing in these forums, please ask.
 
If you are new to photogrpahy don't even read this stuff - most of it is just crazy ~
 
Am I correct in assuming, that a 20D, due to it's 1.6x crop factor, actually uses "less" of the lens' glass and therefore does not "see" deteriorating edge IQ?
--
Flemming
 
Good answer, but. . . the subject line (42) makes even more sense to the question.

LOL!

--
...Bob, NYC

http://www.pbase.com/btullis

You'll have to ignore the gallery's collection of bad compositions, improper exposures, and amateurish post processing. ;)

 

Keyboard shortcuts

Back
Top