Why no digicam with large sensor...

[ngap]

It is due to the small aperature size of digi-cams. That is why compact film cameras have a large depth of field even though the negative is the same size as SLRs, therefore your digicam lenses will have to be a lot larger.

 

[artistx]

: )

It is due to the small aperature size of digi-cams. That is why
compact film cameras have a large depth of field even though the
negative is the same size as SLRs, therefore your digicam lenses
will have to be a lot larger.

 

[ngap]

I was putting a simplified answer but using the same digicam design and increasing the sensor size will not give you an increase in d.o.f.

D.O.F. is dependent on subject distance, focal length, aperature and circle of confusion.

Of these subject distance, focal length and aperature are independent of sensor size. The circles of confusion are dependent on sensor size but the effect on D.O.F is the least significant of the factors.

The easiest way to decrease the d.o.f. is to increase the focal length or change the aperature. For a digicam which people are talking about the effect of changing the sensor size will be minimal and as such the effectiveness.

 

[ngap]

These calculations using
http://dfleming.ameranet.com/dofjs.html

Focal length 100mm
100 feet

Aperature 2.8
35 mm
D.of.F 79.5 --> 134.7

APS

D.of.F 82.3 --> 127.4

Aperature 5.6

35 mm
D.of.F 73.3 --> 157.4

APS
D.O.F 76.7 --> 143.6

As this shows one step in Aperature has a much more significant difference than the format size. To increase the d.of.f it is much efficient to change the physical lens size via forcal length or Aperature than by changing sensor size. As people want the same physical dimensions the only practical way is to do this is to increase the aperature range.

 

[win39]

I don't see how cost can be pointed to. A mechanically complex 300D
SLR with lens and APS sized sensor sells for the same price as the
new 8MP digicams. Size may be a factor, but I think a digicam with
APS sensor could be made not much larger than one of the larger 8MP
digicams. Certainly there is a big range in size with those cameras
with 2/3" sensors. If Nikon can build a camera the size of the 8700
I have no doubt they could build an APS sized sensor digicam not
much bigger than an Olympus c-8080. I would love something like
that for travel.

A 2/3" sensor measures 8.8mm x 6.6mm. An APS sensor measures
23.7mm x 15.6mm. Stick an APS sensor into an 8700 or C8080, and
you would have to scale everything up (especially the lens) to
accomodate the larger sensor. The reason Nikon and Olympus can
make their 8700 and C8080 so small is because they're usi

g a

sensor that's smaller than your pinky finger.

I guess I did not make myself clear. An 8700 is about 2/3 the size of a C8080 and half the size of an 828. Of course Nikon would have to make a bigger camera and lens for a larger sensor, but their ability to make a capable small camera should carry over to a larger sensor camera that would probably not be larger than an 828.

 

[SomedayNever]

It's not aperture. Or format size.

It's focal length.

At a comparable focal length (not equivalents) and a given aperture, no matter the size of the sensor or film, depth of field is comparable. This holds true for everything from 35mm to digital to 4x5 view cameras.

For example: On a 35mm camera, a super wide angle lens, like 17mm, has a lot more depth of field than say a 105mm lens at the same distances. Small sensor digicams generally have lenses that start at around 5 or 7 mm, hence the large amount of depth of field.

My wish is that someone would make a small-ish rangefinder, similar to the new Leica, with an aps-sized sensor and a simple 50mm-equivalent f2 lens. It'll never happen, not without costing a fortune, but a guy can dream. Even a small sensor with a good optical manual focus would be great, as the autofocus on digicams is inaccurate and focusing electronically is near impossible with an EVF.

These calculations using
http://dfleming.ameranet.com/dofjs.html

As this shows one step in Aperature has a much more significant
difference than the format size. To increase the d.of.f it is much
efficient to change the physical lens size via forcal length or
Aperature than by changing sensor size. As people want the same
physical dimensions the only practical way is to do this is to
increase the aperature range.

--
Joe

---------------
http://www.somedaynever.com
click gallery for my photographic ramblings

omaha, nebraska, usa

 

[rennie12]

Stéphane wrote:
...so that you can achieve same shallow dof as with a dslr?

DOF is a product of the focal length of the lens, not the sensor, actually. (That is, a 50 mm focal length lens will have the same depth of field no matter what the image is projected on - film, a piece of paper, or a sensor of large or small size).

I understand what you mean, tho - a larger sensor posits a longer focal length lens, which would result in greater dof....

However, it would also result in a much larger, heavier, and more expensive digicam - and with the D70 and 300D so cheap, there is a limit on what people will pay for digicams.

In addition there is another factor, which is how the sensor works for a digicam versus a DSLR - the digicam sensor permits a live view of the image, while the dlsr does not - I don't remember the details, frankly, but if your'e curious Mr Askey has an excellent description you can find thru his menus.

I hope this is helpful. Kindest regards - rennie12

 

[ngap]

Agree the most significant thing is focal length. However the other three things come into the D.O.F equation.
d = s/[1 ± ac(s-f) f²]

If you want to keep the digicam compact it is hard to increase the focal length.

 

[Stéphane205018]

Stéphane wrote:
...so that you can achieve same shallow dof as with a dslr?

DOF is a product of the focal length of the lens, not the sensor,
actually. (That is, a 50 mm focal length lens will have the same
depth of field no matter what the image is projected on - film, a
piece of paper, or a sensor of large or small size).

I understand what you mean, tho - a larger sensor posits a longer
focal length lens, which would result in greater dof....

However, it would also result in a much larger, heavier, and more
expensive digicam - and with the D70 and 300D so cheap, there is a
limit on what people will pay for digicams.

In addition there is another factor, which is how the sensor works
for a digicam versus a DSLR - the digicam sensor permits a live
view of the image, while the dlsr does not - I don't remember the
details, frankly, but if your'e curious Mr Askey has an excellent
description you can find thru his menus.

I hope this is helpful. Kindest regards - rennie12

 

[BobTrips]

The lenses still require a specific lens-to-sensor distance to
properly focus throughout their range. So removing the mirror
would not reduce the space.

No, but removing the pentaprism would. Yeah, what I want is a 3/4
sensor size on a camera with interchangeable lenses - you could use
the E1 lens system - but with an EVF. Ideally, the EVF would be
modular so that different types of EVFs could be plugged in.

You're still going to want some kind of eyepiece to look through.
Holding a camera a foot or two in front of you while looking at the
LCD to frame the picture might be fine for a little point-n-shoot
digicam, but that's not going to fly for a larger camera with
larger lenses. Serious photographers want to be able to look
through a viewfinder to frame the picture. It also helps
stabilize the camera, too, for better handholdability. Removing
the pentaprism, but replacing it with an EVF, doesn't do anything
to the size of the camera.

--
bob
Latest offering - 'Dusk on the Buriganga'
http://www.pbase.com/bobtrips
Shots from a bunch of places (esp. SEA and Nepal).
Pictures for friends, not necessarily my best.

http://www.trekearth.com/members/BobTrips/photos/
My better 'attempts'.

 

[T3]

Bob, read what he was saying. He was proposing that replacing the pentaprism viewfinder with an EVF viewfinder would result in a reduction in size and space of the camera. Either way, you're still going to have a viewfinder hump. He meant chopping off the viewfinder hump altogether (the hump that houses either the pentaprism or the EVF) in order to reduce size and space, then all you're left with is the external LCD to compose with.

The lenses still require a specific lens-to-sensor distance to
properly focus throughout their range. So removing the mirror
would not reduce the space.

No, but removing the pentaprism would. Yeah, what I want is a 3/4
sensor size on a camera with interchangeable lenses - you could use
the E1 lens system - but with an EVF. Ideally, the EVF would be
modular so that different types of EVFs could be plugged in.

You're still going to want some kind of eyepiece to look through.
Holding a camera a foot or two in front of you while looking at the
LCD to frame the picture might be fine for a little point-n-shoot
digicam, but that's not going to fly for a larger camera with
larger lenses. Serious photographers want to be able to look
through a viewfinder to frame the picture. It also helps
stabilize the camera, too, for better handholdability. Removing
the pentaprism, but replacing it with an EVF, doesn't do anything
to the size of the camera.

--
bob
Latest offering - 'Dusk on the Buriganga'
http://www.pbase.com/bobtrips
Shots from a bunch of places (esp. SEA and Nepal).
Pictures for friends, not necessarily my best.

http://www.trekearth.com/members/BobTrips/photos/
My better 'attempts'.

 

[DRG]

It is due to the small aperature size of digi-cams. That is why
compact film cameras have a large depth of field even though the
negative is the same size as SLRs, therefore your digicam lenses
will have to be a lot larger.

In making any comparison, you need to state what is being held fixed as other variables are compared.

The most useful parameter to hold fixed in exploring the impact of sensor size is the angle of view. After all, it is the most logical assumption that the various cameras involved in the comparison are trying to take a similarly-framed shot. This also implies that the camera-to-subject distance is being kept fixed, keeping perspective (the relative size of foreground and background objects) the same.

Once you make this assumption, the importance of sensor size becomes clear. Let camera #1 have a sensor size that is 1/2 that of camera #2. Then, the focal length of lens #1 must be half that of lens #2 in order to maintain the same angle of view.

Now, given these facts, obtaining identical DoF between the two cameras requires that the F-number of camera/lens combo #1 be 1/2 that of camera/lens combo #2.

That is:

50mm @ F/5.0 on camera #1 = 100mm @ F/10.0 on camera #2, for identical AoV and DoF.

From this, you can see that if you move the aperture of camera #1 to F/10.0 to match camera #2, you'll be increasing the DoF significantly. You can say that the larger DoF is ultimately due to the smaller FL of this camera, but such a focal length was mandated by the desire for constant AoV. It wasn't really a choice. Sensor size forced this change, which, in turn, forced the aperture change to maintain constant DoF.

David

 

[Jay Turberville]

Bob, read what he was saying. He was proposing that replacing the
pentaprism viewfinder with an EVF viewfinder would result in a
reduction in size and space of the camera.

Yes, that was part of what I said. The other part that I said was that the EVF would be modular and that different types could be plugged in. Ideally, it would be non-proprietary - but that is a severe pipe-dream.

Either way, you're
still going to have a viewfinder hump. He meant chopping off the
viewfinder hump altogether (the hump that houses either the
pentaprism or the EVF) in order to reduce size and space, then all
you're left with is the external LCD to compose with.

I guess you never used a TLR YashicaMat or similar. They had these nifty little little flip up hoods and when you popped the front panel a thin little magnifier would be popped up and you now had a finder that you could put your eye to. It all folded into a neat package.

--
Jay Turberville
http://www.jayandwanda.com

 

[Just Looking]

DRG and ngap, you're both right on; I've tried to make the same points myself on many occasions.

As DRG points out, any such comparisons need to be based on clarity about what's fixed and what's being compared. As ngap points out, the sensor size is irrelevant to DoF if you compare using aperture diameter, instead of the more conventional relative aperture, or f-number.

That makes it easy to appreciate the original poster's point: it's hard to get shallow DoF with a small sensor, because the required large aperture diameter can become an unrealizable low f-number less than 1.

j

It is due to the small aperature size of digi-cams. That is why
compact film cameras have a large depth of field even though the
negative is the same size as SLRs, therefore your digicam lenses
will have to be a lot larger.

In making any comparison, you need to state what is being held
fixed as other variables are compared.

The most useful parameter to hold fixed in exploring the impact of
sensor size is the angle of view. After all, it is the most logical
assumption that the various cameras involved in the comparison are
trying to take a similarly-framed shot. This also implies that the
camera-to-subject distance is being kept fixed, keeping perspective
(the relative size of foreground and background objects) the same.

Once you make this assumption, the importance of sensor size
becomes clear. Let camera #1 have a sensor size that is 1/2 that of
camera #2. Then, the focal length of lens #1 must be half that of
lens #2 in order to maintain the same angle of view.

Now, given these facts, obtaining identical DoF between the two
cameras requires that the F-number of camera/lens combo #1 be 1/2
that of camera/lens combo #2.

That is:

50mm @ F/5.0 on camera #1 = 100mm @ F/10.0 on camera #2, for
identical AoV and DoF.

From this, you can see that if you move the aperture of camera #1
to F/10.0 to match camera #2, you'll be increasing the DoF
significantly. You can say that the larger DoF is ultimately due to
the smaller FL of this camera, but such a focal length was
mandated by the desire for constant AoV. It wasn't really a
choice. Sensor size forced this change, which, in turn, forced the
aperture change to maintain constant DoF.

David

 

[Just Looking]

DOF is a product of the focal length of the lens, not the sensor,
actually. (That is, a 50 mm focal length lens will have the same
depth of field no matter what the image is projected on - film, a
piece of paper, or a sensor of large or small size).

that is incorrect.

DoF depends very much on using a sensible circle-of-confusion criterion for what is "in focus", and the sensible criterion is conventionally taken as some fraction of the image size. So the format matters.

j

 

[Medic Laviña]

DOF is a product of the focal length of the lens, not the sensor,
actually. (That is, a 50 mm focal length lens will have the same
depth of field no matter what the image is projected on - film, a
piece of paper, or a sensor of large or small size).

that is incorrect.

DoF depends very much on using a sensible circle-of-confusion
criterion for what is "in focus", and the sensible criterion is
conventionally taken as some fraction of the image size. So the
format matters.

j

...let me give you something to think about:

Imagine replacing a Canon 10D sensor with a smaller or larger one (same pixel pitch and pixels per inch). Which do you think will be affected?
(a) Depth of field; or (b) Field of view

If your answer is "b", you're right. And this is the reason why Canon EOS-1D and 10D produces the same DOF when using the same lens. Only the field of view is different. The smaller the sensor, the narrower the field of view. While EOS-1D has 1.3x focal length multiplier, EOS-10D has 1.6x, because of the smaller sensor.
Now you understand it better. ;-)

--
Medic

 

[Michael Long]

If your answer is "b", you're right. And this is the reason why
Canon EOS-1D and 10D produces the same DOF when using the same
lens. Only the field of view is different. The smaller the sensor,
the narrower the field of view. While EOS-1D has 1.3x focal length
multiplier, EOS-10D has 1.6x, because of the smaller sensor.

On the one hand, any given lens of a given focal length at the same aperture will obvously produce the same image characteristics at the focal plane, no matter what the media size.

On the other, a 85mm lens on a 1Ds gives roughly the same FOV as a 50mm lens on a 10D.

Given the same subject distance of 10ft, and as such, the same scene, @ f/8, and the same CoC, the 10D will get a DOF of 7.56-14.7, while the 1Ds will get 9-11.23.

So to get the same FOV at the same distance, and hence the same image frame, the 10D needs a shorter focal length.

And since the 10D is using shorter focal lengths to get the "same" results, the smaller format seems to have more DOF.

 

[Just Looking]

DOF is a product of the focal length of the lens, not the sensor,
actually. (That is, a 50 mm focal length lens will have the same
depth of field no matter what the image is projected on - film, a
piece of paper, or a sensor of large or small size).

that is incorrect.

DoF depends very much on using a sensible circle-of-confusion
criterion for what is "in focus", and the sensible criterion is
conventionally taken as some fraction of the image size. So the
format matters.

j

...let me give you something to think about:
Imagine replacing a Canon 10D sensor with a smaller or larger one
(same pixel pitch and pixels per inch). Which do you think will be
affected?
(a) Depth of field; or (b) Field of view

or (c) both of the above.

If your answer is "b", you're right.

No, "c" is the corrent answer.

Unless when you say "same pixels per inch" you really mean to step outside of the usual practice of taking Circle of Confusion to be a fixed fraction of the image size, and instead make it an absolute or fixed to the pixel size. If that's what you mean, you have to so state, since it's not the usual assumption about Depth of Field. Than (b) would be correct. But as my message stated, that was not the case being considered.

... And this is the reason why
Canon EOS-1D and 10D produces the same DOF when using the same lens.

That doesn't even match the situation you proposed. The EOS-1D has 4 MP in a bigger area than the 10D's 6 MP, so has much bigger pixels. Whether you tie CoC to format size or pixel size, the DoF will be greater on the EOS-1D when the focal lengths and f-numbers are the same, and greater on the 10D when the fields of view and f-numbers are the same.

Without saying what you're keeping constant, including possibly CoC or the ratio of CoC to format size (but not both), you can't make a meaningful comparison across formats.

By the way, the 8 MP Mark II also has bigger pixels than the 10D, but not by such a large margin.

... Only the field of view is different. The smaller the sensor,
the narrower the field of view. While EOS-1D has 1.3x focal length
multiplier, EOS-10D has 1.6x, because of the smaller sensor.
Now you understand it better. ;-)

Yes, I do; thanks you for that explanation. But see my comments for DoF and CoC issues that you missed.

j

Medic

 

[Just Looking]

If your answer is "b", you're right. And this is the reason why
Canon EOS-1D and 10D produces the same DOF when using the same
lens. Only the field of view is different. The smaller the sensor,
the narrower the field of view. While EOS-1D has 1.3x focal length
multiplier, EOS-10D has 1.6x, because of the smaller sensor.

On the one hand, any given lens of a given focal length at the same
aperture will obvously produce the same image characteristics at
the focal plane, no matter what the media size.

On the other, a 85mm lens on a 1Ds gives roughly the same FOV as a
50mm lens on a 10D.

Given the same subject distance of 10ft, and as such, the same
scene, @ f/8, and the same CoC, the 10D will get a DOF of
7.56-14.7, while the 1Ds will get 9-11.23.

Same CoC is a rather unusual condition to compare. If you allow the 1Ds a CoC that is the same percentage of image size (a conventional, but not uniquely correct approach), then the 1Ds DoF will still be less than the 10D's, but not by as much.

j

So to get the same FOV at the same distance, and hence the same
image frame, the 10D needs a shorter focal length.

And since the 10D is using shorter focal lengths to get the "same"
results, the smaller format seems to have more DOF.

 

[Stéphane205018]

that a "simple" question can result in complicated answers....or let's say, "elaborated" replies

So basically, forgetting all the tech stuff, is it a possible option for the future: digicam with big sensor: more dof (am I wrong? ), LESS NOISE, HIGHER ISO

Or should I switch to dslr?

DOF is a product of the focal length of the lens, not the sensor,
actually. (That is, a 50 mm focal length lens will have the same
depth of field no matter what the image is projected on - film, a
piece of paper, or a sensor of large or small size).

that is incorrect.

DoF depends very much on using a sensible circle-of-confusion
criterion for what is "in focus", and the sensible criterion is
conventionally taken as some fraction of the image size. So the
format matters.

j

...let me give you something to think about:
Imagine replacing a Canon 10D sensor with a smaller or larger one
(same pixel pitch and pixels per inch). Which do you think will be
affected?
(a) Depth of field; or (b) Field of view

or (c) both of the above.

If your answer is "b", you're right.

No, "c" is the corrent answer.

Unless when you say "same pixels per inch" you really mean to step
outside of the usual practice of taking Circle of Confusion to be a
fixed fraction of the image size, and instead make it an absolute
or fixed to the pixel size. If that's what you mean, you have to
so state, since it's not the usual assumption about Depth of Field.
Than (b) would be correct. But as my message stated, that was not
the case being considered.

... And this is the reason why
Canon EOS-1D and 10D produces the same DOF when using the same lens.

That doesn't even match the situation you proposed. The EOS-1D has
4 MP in a bigger area than the 10D's 6 MP, so has much bigger
pixels. Whether you tie CoC to format size or pixel size, the DoF
will be greater on the EOS-1D when the focal lengths and f-numbers
are the same, and greater on the 10D when the fields of view and
f-numbers are the same.

Without saying what you're keeping constant, including possibly CoC
or the ratio of CoC to format size (but not both), you can't make a
meaningful comparison across formats.

By the way, the 8 MP Mark II also has bigger pixels than the 10D,
but not by such a large margin.

... Only the field of view is different. The smaller the sensor,
the narrower the field of view. While EOS-1D has 1.3x focal length
multiplier, EOS-10D has 1.6x, because of the smaller sensor.
Now you understand it better. ;-)

Yes, I do; thanks you for that explanation. But see my comments
for DoF and CoC issues that you missed.

j

Medic