lens digital equivalents

[abaazov]

can anyone please explain to me, in layman's terms, the connection between the focal length of a lens and the size of the digital sensor on the camera. am i correct in assuming that the same lens, at the same length, will show two "different" pictures on, say, the 20d and the 1ds?

[jani]

so a wide-angle lens on one can will not be wide angle on another. but why does the size of the sensor determine how much a camera sees?

Here's an over-simplification:

That's because a given lens projects an image of a fixed size.

If your lens is made for 35mm cameras, it's projecting an image circle of at least 43mm diameter at the plane of focus.

If your sensor doesn't cover all of that area, it isn't covering all of the image, either.

DPReview has a nice article explaining some of the basics, while Digital Outback Photo has a more technical article.

[abaazov]

ok, i am beginning to understand. i appreciate the info guys. so if most of the lenses out there are made for 35mm cameras, and let's say i use a 20d, which has a diagonal sensor size of 27mm, any lens i buy (other than the ones specifically made for the 20d) i will have to use the conversion ratio of 1.6 to get the real focal lens i will effectively be using. so the 70-200 is actually a 112-320 on the 20d??

[dazzajl]

When you look in the view finder of your 10D, the image will "appear" the same as the scene viewed with a 112-320mm lens on a regular 35mm camera.

This can be a really tricky subject, there are pitfalls in the statement above too. Not trying to be picky Howard as I know you were talking about the magnification factor and what appears in the frame but it goes deeper.

Yes the 70-200 when used on a 10/20D or other similar camera with fill the frame like a 112-320 would on 35mm film body BUT....  the effect of compression and the depth of field remain true to the 70-200 focal length.

One analogy I like to use is the simple window one. If you move back or forwards looking out of a window the angle of view (or the amount you can see side to side, top to bottom) will increase or decrease. This corresponds to changing a lens on your camera.

If that window was the rear window of your car, then imagine the effect of making the rear view mirror bigger or smaller. As you increase it you will see a larger view (or wider angle) if you were to make it smaller, you would see less. By seeing less you are croping into the view and this is the same as the 10/20D's are doing.

I hope that makes some sense to someone.  

[abaazov]

is it fair to say that you are "losing" data when you use a lens on a camera with a smaller than 35mm camera sensor? in a way is it a waste?

[boku]

I was thinking of magnification where the 20D and 35mm camera have the same size view finder image.

I know, I was just trying to add on to what you were saying.    

Many people talk about 200mm lenses becomming a 320 on some digi SLR and so on but as we know that's not the case. There are many times we might choose to move back from a subect and use a longer lens to shrink the depth of field to get a desired look. Changing from a full size sensor to a smaller one will of course not change the DoF at all.

Oh dear! - You've got hold of the wrong end of the stick I'm afraid.

The DOF equation has 4 variables:
1. Sensor/film size
2. Focal length of lens
3. Aperture
4. Distance to subject

If you keep 2,3,4 constant, the DOF increases with a decrease in sensor size - by precisely the value that is usually quoted as the crop factor.

So that 70-200mm lens will give 1.6 time greater DOF on the 20D as it would on the 1DsMkII.

Here's a link to a handy calculator so that you can experiment with the variables.

http://www.dofmaster.com/dofjs.html

Just an observation...

Your #1 is not correct as stated. You really mean circle of confusion. That is empirically related to sensor size, but if the circle of confusion remains the same for different sensor sizes being compared, then your #1 factor is no longer relevant. Look at your link.

This has been discussed at great length here. You might want to do a search on discussions, but they are very arduous. For the practical photographer - and I am one - your #1 factor is not a strong consideration.

I just thought I'd bring this up. You are not entirely right or wrong, but you have failed to explain the rationale. I have as well since it involves final size of the image. I just need to set the record straight. #1 should correctly be stated as circle of confusion.

Over.

[howard smith]

abaazov, it is not a pun.  Check out the article "Understanding Depth of Field."  Ansel Adams described it thus:

"The image of a 'point" in the subject should be a 'point' on the film.  If the subject is not exactly in the plane of critical focus, however, this image becomes a small blurred disc, called a circle of confusion.  The size of any circle of confusion becomes smaller as the aperture is reduced, making the image appear sharper.  We define limits of the size of these circles of confusion that we consider to be acceptably sharp focus, even though they are not quite as sharp as the critical focus plane."

The depth of field is the distance between the nearest and farthest point that appear acceptably sharp (or not unacceptably out of focus).

According to Michael's article, you cannot understand depth of field until you understand circle of confusion.  That is for sure true.

[boku]

but as an empirical matter as you say COC is determined in the calculator by a lookup table, which in the DSLR world correlates generally very well with the sensor size, print size and viewing conditions are of course assumed to remain constant for the calculation too.

Yup, I noticed that.

I guess I was just side-stepping starting another DOF-fest!  

[Ray]

To say that DoF is reliant on sensor size is like saying you can change DoF by cropping an image.

Here we go again!  

Actually, you can change DoF by cropping an image. Shoot a 2-dimensional object, say a free standing brick wall, set against a distant background which is out of focus. The brick wall is tack sharp. The background is slightly fuzzy because you used f5.6 and were fairly close to the wall. The impression you will get from almost any size print is a sense of a rather shallow DoF.

Crop the image so that all that remains is the brick wall. How do you then describe the DoF? Is it greater or less? Would it make any difference if you used F2.8, F5.6 or f11? Is there anything in the cropped image that would give a clue?

If you think this example is absurd. then I would maintain that making DoF decisions and calculations without regard to format is equally absurd.

[Ray]

Actually, I'm really surprised at the deep seated confusion on this issue. It's true that you can calculate a CoC on the negative or sensor knowing just the focal length of the lens, the f stop and the distance to subject. But without knowledge of both the format of the negative and the size of print that will result from a specific degree of enlargement from that negative or sensor, the CoC has an unknown effect on DoF. It is therefore not a DoF calculation, but a CoC calculation and incomplete as regards DoF considerations.

To calculate the true DoF that will be apparent on a given size print it is necessary to know 5 essential ingredients.

1. The format of the film or sensor, and included in that is the amended format resulting from post shooting cropping.

2. The degree of enlargement from the film format or amended format (2x, 4x 8x etc).

3. The focal length of the lens.

4.  The f stop.

5. The distance to subject.

If any one of these five ingredients is missing, then accurate DoF calculation is not possible (with regard to a complete composition, of course).

It's possible to use a couple of the variables as a constant to create a formula that only applies to a specific format and degree of enlargement, thus giving the impression that format has no bearing on the matter. But this is pure illusion.

If y = 2x so that I can substitute 2x for all instances of y in the equation, should I then pretend that y doesn't exist? That would be foolish. If y doubles in size then so does 2x.

[drh681]

you will see "more" picture from a 1Ds.

the 1Ds sensor is roughly 24mmx36mm or 864 sq.mm. with a diagonal of roughly 43mm.

the 20D sensor is 15mmx22.5mm or 337.5 sq.mm. with a diagonal of 27mm

lenses with less focal length than the diagonal of the sensor(or film) are considered wide angle.

lenses with focal lengh longer than the diagonal of the sensor are considered to be telephoto.

an "ultra" wide angle might have a focal length of half or less of the sensor diagonal.

a nice "portrait" telephoto might have a focal length double the sensor diagonal.

[dazzajl]

I was thinking of magnification where the 20D and 35mm camera have the same size view finder image.

I know, I was just trying to add on to what you were saying.    

Many people talk about 200mm lenses becomming a 320 on some digi SLR and so on but as we know that's not the case. There are many times we might choose to move back from a subect and use a longer lens to shrink the depth of field to get a desired look. Changing from a full size sensor to a smaller one will of course not change the DoF at all.

[howard smith]

#1 should be circle of confusion.  In addition, read the article "Understanding Depth of Field" on this site.

peripatetic, if you go to the link you provided, schroll down to "equation" (in red) and click there, you will see the way DoF is calculated and that nowhere is there a "format" term.  As Boku said, it is wraped up in the circle of confusion term along with some other things, like print size and viewing conditions.

[Ray]

Thank you!  

[abaazov]

so a wide-angle lens on one can will not be wide angle on another. but why does the size of the sensor determine how much a camera sees?

[howard smith]

dazzajl,

Your comments are correct.  A 70-200mm lens is a 70-200mm lens regardless of the camera's format.  I was thinking of magnification where the 20D and 35mm camera have the same size view finder image.

[dazzajl]

Oh dear! - You've got hold of the wrong end of the stick I'm afraid

If you're failing to find me at the end of the stick then that would be because you have the wrong stick

There seems to be a common theme on this board of people ignoring common sense in order to (mis)quote from the text books.

To say that DoF is reliant on sensor size is like saying you can change DoF by cropping an image.

[abaazov]

isnt it amazing they call this circle of confusion! i never knew photographers had such a well-developed, dark, ironic, sense of humor!