Pages: [1]   Bottom of Page
 Author Topic: Depth of Field for Canon  Digital Rebel  (Read 5504 times)
61Dynamic
Sr. Member

Offline

Posts: 1442

 « on: January 28, 2004, 07:04:22 PM » Reply

I'd suggest a prog, but don't know of any off hand. But I can tell you that you don't have to worry about the multiplier. It has no effect on DOF.
 Logged
Jonathan Wienke
Sr. Member

Offline

Posts: 5759

I have a depth-of-field calculator Excel spreadsheet that allows one to input the sensor dimensions and resolution, and calculates the appropriate circle of confusion for pixel-level crispness for your camera. Then input a distance and focal length, and DOF is displayed for f/1 to f/64. You can download it here. The spreadsheet has an English and metric version.
 Logged

Rick Hearn
Newbie

Offline

Posts: 25

Here is Michael's review of a PocketPC DOF calculator.
 Logged
61Dynamic
Sr. Member

Offline

Posts: 1442

Quote
Here we go again.

No need to be rude. :|

If COC concerns you in regards to print size then don't forget that viewing distance also has an effect on COC as well...

The question is, are you really concerned about what portions are in focus based on print size? If so, then sensor size will have an effect. But if your taking a pic of a statue and want a guy doing the hokey-pokey behind it to be blurred and that's all your concerned with, then you don't need to worry about sensor size.

It's really all a matter of personal prefference and what you are trying to achieve with your image. How many people will look at an image and say, "Gee, if only he had compensated the COC for the 1.6 crop factor..."

 Logged
Tim Gray
Sr. Member

Offline

Posts: 2002

ooops I meant flash is correct....
 Logged
BJL
Sr. Member

Offline

Posts: 5182

Quote
Don't take the lens length and multiply by the factor.  If you want to do that the mathematical calculation requires you take the square root of the factor times the lens length.
Why? If you want the same field of view with different formats, you multiply focal length by the linear dimension of the format: "crop factors" like 1.6x are length scales, not area scales.
Then the DoF calculators confirm that:

"To get the same FOV and DOF on prints of the same size viewed from the same distance (after whatever cropping you choose), one must scale CoC and hence aperture ratio roughly in proportion to the focal length used. In other words, the aperture diameter is roughly constant."
(Macro focusing might mess this up a bit.)

Typically the focal length is scaled in proportion to linear sensor size, but the above phrasing applies to other options like Jonathon Wienke's approach of scaling in proportion to sensor resolution, which means cropping to about the same pixel dimensions.

If this comes up one more time, I will be tempted to write out the formulas for DoF based on CoC as measured on prints, with print dimensions and viewing distance all there as input parameters, and the ultimate constant being a ratio of about 3000 between viewing distance and print CoC diameter. (A value of about 2000 is often used in traditional DoF scales, due to a historical underestimation of the human eye's capabilities.)
 Logged
Tim Gray
Sr. Member

Offline

Posts: 2002

Quote
Quote (Tim Gray @ Feb. 01 2004,11:22)
Don't take the lens length and multiply by the factor.  If you want to do that the mathematical calculation requires you take the square root of the factor times the lens length.

Why? If you want the same field of view with different formats, you multiply focal length by the linear dimension of the format: "crop factors" like 1.6x are length scales, not area scales.

Dof is proportional to the square of the focal length, and directly proportional to the COF.  If you don't take the sq root of the factor and multiply by the FL, you are overstating the DOF.

In fact it is related to area.  the factor refers to the ratio of diagonal corner to corner measurements compared to a 35mm frame  -  which is proportional to the area of the sensor, not the sum of the length and width.

Is there agreement that DOF is, in fact, dependent on sensor size?
 Logged
Tim Gray
Sr. Member

Offline

Posts: 2002

So - give me the DOF I'll experience with a 1D, 70-200mm ISL @ 200mm, F8, 100 feet to subject - and what COF do you use to get that?   If you need more data, say a print of 8x10 viewed at 18 inches.   I figure 24 feet using a COF of .019 and HFD of 863'.
 Logged
Tim Gray
Sr. Member

Offline

Posts: 2002

So if .019 is reasonable for a COF, the Dof in my example would be 24 ft, right?
 Logged
MAGIC9966
Newbie

Offline

Posts: 11

Need a Depth of Field pocketPC program for my Canon Digital Rebel with EF-S 18-55 f/3.5-5.6 USM and EF 55-200 f/4.5-5.6 II USM that converts with Lens Multiplier Factor (1.6).
 Logged
boku
Sr. Member

Offline

Posts: 1493

Here you go...

http://www.bep.co.uk/photo/Photodepth.htm

I have no idea how good this is, but just do a Google search on "depth of field chart" and you will find a ton of resources. Thing is, "hyperfocal distance" is so concrete when often you are more interested in just how you can throw something out of focus into a specific point in "bokeh-ville". DOF charts won't tell you that.

I like two other approaches (both have obvious drawbacks):

1) Depth of field preview in-viewfinder (dim at best).
2) Depth of field scale on lens barrel (I am showing my age here).

Good luck, let us know what you end up with.
 Logged

Bob Kulon

Oh, one more thing...
Play it Straight and Play it True, my Brother.
boku
Sr. Member

Offline

Posts: 1493

Quote
I have a depth-of-field calculator Excel spreadsheet that allows one to input the sensor dimensions and resolution, and calculates the appropriate circle of confusion for pixel-level crispness for your camera. Then input a distance and focal length, and DOF is displayed for f/1 to f/64. You can download it here. The spreadsheet has an English and metric version.
Jonathan,

I like it!

Thanks,
Bob
 Logged

Bob Kulon

Oh, one more thing...
Play it Straight and Play it True, my Brother.
flash
Full Member

Offline

Posts: 151

Quote
I'd suggest a prog, but don't know of any off hand. But I can tell you that you don't have to worry about the multiplier. It has no effect on DOF.
Here we go again.

The smaller sensor requires a different COC to acheive the same standard print size. The size of the sensor therefore does most definately have an effect of DOF. It's just like comparing like focal lengths between MF and 35mm.

Gordon
 Logged
flash
Full Member

Offline

Posts: 151

I did not mean to be rude and if you took it that way I appologise. A quick search will reveal that this has appeared many times and some of the threads are quite vigorous to say the least. However you did make a statement of fact and I don't believe that the fact is correct.

Do you make smaller prints or adjust your viewing distances because an image was taken on a larger or smaller sensor/ piece of film. I don't. I base print size on the subject not the sensor. When comparing different image formats you would need to keep other factors constant so as to be able to make a comparison of the criteria you are comparing. So, one could assume that both prints would be the same size and that the viewing distance would be the same. Depth of field calculations assume a standard print size (5x7 I think) and a standard viewing distance. The variable are focal length, aperture and COC.

In some specific areas like macro work where very precise DOF is required of for those of us unlucky enough to try and shoot portraits with slow zoom lenses a significant change in DOF is far from insignificant. The change in DOF can be up to 4x depending on the size of the sensor, especially in some digicams. If COC wasn't important we wouldn't need it as a variable in making DOF calculations.

For you, and the type of photography you do precise DOF may not be important, but for others it is, to the point where they will calculate the COC based on sensors size, print size and viewing distance.

Gordon
 Logged
Tim Gray
Sr. Member

Offline

Posts: 2002

61dynamic is correct.

Here is a "good" calculator

http://www.dudak.baka.com/dofcalc.html

I say "good" because it lets you choose the "sensor" size (actually the film format) - why?  Because absolutely the number you get out of the calculation is dependent on the  Circle of ConFusion which is in turn dependent on the size of the capture area.  Take COF (whatevery you want but as I recall .030mm is "standard" 35mm) and divide by 1.6 or 1.3 or whatever.  Sometimes this isn't apparent in the calculator (older ones) which assume 35mm format.  You need to be able to use the COF you want - not what the calculator assumes.

Don't take the lens length and multiply by the factor.  If you want to do that the mathematical calculation requires you take the square root of the factor times the lens length.

I think what causes this confusion is that folks tend to think in terms of 35mm either P&S or SLR and the debate whether its' a "multiplier" or "crop" factor.  Nobody says that the 100mm lens on a 35mm isn't really 100mm since the 35mm frame is just a crop of a 645 frame.

I also, confess to having experienced a touch of deja vu in response to this thread.
 Logged
Jonathan Wienke
Sr. Member

Offline

Posts: 5759

Quote
No, I just tried it and it worked. Try again...
 Logged

BJL
Sr. Member

Offline

Posts: 5182

Quote
Depth of field calculations assume a standard print size (5x7 I think) and a standard viewing distance.
I am not sure if the same print size has been used for all film formats, or if it has been assumed that larger formats are typically printed somewhat larger, like about 10x8 for MF; Michael has mentioned an assumption of a maximum possible enlargement of about 5x. This could correspond to Jonathan's ideas of scaling CoC according to sensor resolution.

So I have a question for those familiar with 35mm and MF lenses and their DoF scales. Comparing 35mm and MF lenses of the same focal length, do the MF lenses have the same relationship of DoF to f-stop, or do they give more DoF at the same stop, or less?

If they give about the same DoF, they seem to be working with a bigger print size and the same viewing distance, perhaps based on a common magnification from the negative.

If the MF lens scales give more DOF, there are probably using a common print size/distance standard, as assumed here.

If they give less DoF, I am wrong somewhere!
 Logged
BJL
Sr. Member

Offline

Posts: 5182

Quote
Dof is proportional to the square of the focal length
You have to adjust focal length in proportion to crop factor to be considering the same field of view; changing it by square root of the linear scaling factor is considering DoF for a different framing of the subject.

It is time to look at some actual equations; a nice reference is
http://dfleming.ameranet.com/
and in particular the simpler, approximate DOF formulas at
http://dfleming.ameranet.com/equations2.html
These become inaccurate only for extreme close-ups.

What they say is

a) the DoF is approximately determined by the subject distance and the hyperfocal distance H, where the latter is the distance at which to focus in order to have everything out to infinity in focus.

The nearest and farthest in-focus objects are at approximate distances S.H/(H+S) and S.H/(H-S),
S = subject distance.

c) The hyperfocal distance is in turn given exactly by
H = F^2/(N.C)  [first form]
H = F.A/C [second form]
F=focal length
N = F/A = aperture ratio
A = aperture diameter
C = maximum allowable circle of confusion on the image formed in the camera in order for something to be considered in-focus.

So yes, when working with a fixed format, and so with fixed circle of confusion, then for a given aperture ratio and subject distance, DOF varies inversely with the square of focal length.

But when you compare different formats, it is more useful to think of it this way:
i) adjust focal length in proportion to linear format size [or in-camera image size after cropping, if any] to get the same field of view for the image.
ii) adjust allowable circle of confusion C in proportion to focal length, to get the same maximum "print CoC" size for in-focus objects, on prints in which subjects appear at the same size;

then,

A) if you keep the same aperture ratio N, the first form shows that the hyperfocal distance increases in proportion with focal length and so with with format size: DOF decreases with increased format size at fixed f-stop, and

if you keep the same aperture diameter A (by changing aperture ratio in proportion to focal length), the second form shows that you get (approximately) the same depth of field.
 Logged
BJL
Sr. Member

Offline

Posts: 5182

Quote
So - give me the DOF I'll experience with a 1D, 70-200mm ISL @ 200mm, F8, 100 feet to subject - and what COF do you use to get that?   If you need more data, say a print of 8x10 viewed at 18 inches.   I figure 24 feet using a COF of .019 and HFD of 863'.
There are several plausable schools of thought on choosing the allowable CoC. (Once CoC is chosen, you can use the calculator at the site I cited!)

One idea is to use the limits of human visual acuity which suggest print CoC = (viewing distance)/3000, so in your example, 0.006in = 0.15mm. Since to get an 8x10 from a 1D with frame height of about 19mm, the magnification is about 10, use sensor CoC of .15mm/10 = 0.015mm. But demanding visually perfect sharpness at the near and far limits of the field of view, away from the critical parts of the subject, is perhaps too strict, and it seems that most DoF scales are based on a denominator of 2000 instead, which would give .023, on the other side of your value of .019. So close enough!

Another idea is to set the limit to match the sensor's resolution;  very ambiguous with film but deceptively simple with digital: resolution of a standard sensor using standard Bayer style interpolation algorithms is roughly the spacing of the green pixels, or pixel pitch times sqrt(2), and the 1D has a pixel pitch of about 11microns, so that gives CoC = 16microns = 0.016mm.

What a coincidence: the two values are almost equal, so it seems that the 1D is designed for exactly the sort of print size and viewing condition combination that you envisioned! [The Mark II allows you to view that 8"x10" from about 13", which is roughly the print's diagonal length, the traditional "normal field of view" perspective, so the step up to 8MP might reach a sweet spot for completely adequate resolution for "normal viewing".]

If the resolution-based CoC were larger, there would have been no point using the value based on visual acuity, since by that standard, nothing would be sharp enough to be called "in focus".
 Logged
 Pages: [1]   Top of Page