Beginner question about depth of field

fgianni

fgianni

Trainee Amateur
Local time
1:20 PM
Joined
Aug 9, 2005
Messages
1,401
I assume that with the 1.5 crop factor on the RD-1 the circles of confusion must be quite different from the ones on a 35mm full frame camera, so the depth of field scale found on lenses most likely does not apply to the RD-1.

What I have done so far is the fiollowing:
The 1.5 crop factor is very close to the factor obtained using a 1.4 multiplier, which causes the loss of 1 f-stop.
If the effect on COC is similar (Very big assumption I admit) then I can use the DOF scale for the next f-stop up, i.e. if I am shooting at f8 I use the DOP scale for f5.6.

So far I have obtained good results that way, but I'd like to know if there is a more scientific way to determine the DOP for a camera with a 1.5 crop factor given the DOP for full frame 35mm.

Thanks
 
Sorry I forgot to mention the rationale for my assumption, above:
The COC should be the same size on APS sized and full frame, assuming the APS sized 1.4 times smaller, the COC related to the image area must be 1.4 times bigger, so DOP should be close to the one for 1 stop faster. (Clear as mud)

I know the crop factor is 1.5 instead of 1.4 but I think things must be fairly similar.
 
As a starting point your assumption is not bad, especially if you get good results, but you should be aware that DOF is not an absolute value. It depends on the distance of the subject to the lens, the aperture of the lens and on the magnification and viewing distance of the final print. The "circle of confusion" as used by camera-manufacturers is just a mathematical value valid for the film and lenses of the time (the 1920ies) which is no more than a convention to enable comparisons. To top it all off, DOF in front of the subject is different from DOF to the rear of the subject, the ratio being dependent on the film to subject distance but about 1/3-2/3 at normal shooting distances. As user of a digital camera you are in the unique position to develop your intuitive "feel" for DOF far quicker than film users, as you can see your result far quicker. So it all boils down to learning to use your tools in your own artistic way, as it should be.....
 
jaapv,

Well said. I was going to make much the same point. DOF scales are guestimates (for various reasons, many of which you listed above) and so actual individual experience and experimentation are more useful ways of determining what works for "acceptable" DOF.

Cheers,

Sean
 
Fransesco

With a cropfactor you have MORE depth of field than without cropfactor (ceteris paribus).
So using the distancescale for f 5.6 when shooting at f 8 seems too conservative .. would not it be the other way around giving your reasoning ... using the distancescale for f8 when shooting at 5.6. ???

I myself just ignore the cropfactor and use the distancescale as is ...... my assumption is i probably do not have the fastest shutter this way but my Dof will never be too short!

Han
 
J. Borger said:
With a cropfactor you have MORE depth of field than without cropfactor (ceteris paribus).Han
Click to expand...

Sorry Han but i struggle with your statement: an image cropped by a factor of 1.5 is like taking the central part (roughly 1/2 the area) and blowing it up, a detail that looks in acceptable focus in the original image, might look a bit too fuzzy when blown up, not the other way round.
Essentially if I have a 50mm lens, with a crop factor of 1.5 it is like having a 75mm lens, but the DOF scale on the barrel is still the one for a 50mm glass, please let me know where I am wrong.

Cheers
 
Hi Francesco,

Yep you are right ....... i forgot the effect of the change in focal length you mention :bang: .
With a cropfactor you have MORE depth of field (about 1 stop) because of the smaller sensor (the APS size COC you refer too) .... but that is only true if the focal length is equal.
So there are 2 effects in opposite directions (smaller sensorsize: more dof; longer focal length: less dof )

Han
 
The DOF of a lens will be the same on any camera, if the images are enlarged by the same amount. As Han said, there are two things going on here...

(a) you would choose a shorter focal length for a given subject/distance (=>more DOF)
(b) you will probably enlarge more than a full frame image (=>less DOF)

If you do the math, it turns out that (a) has a bit more effect than (b), so you get a net increase in DOF, however it's not a lot, and probably a good starting point is just use the scale on the lens.

Phil
 
fgianni said:
...if I have a 50mm lens, with a crop factor of 1.5 it is like having a 75mm lens, but the DOF scale on the barrel is still the one for a 50mm glass...
Click to expand...

Not exactly because the DoF markings of your lens are made for 35mm film, hence a different CoC.
You'll get a good approximation if you use the DoF markings of the nearest faster f stop of your lens i.e. f/5.6 when you choose f/8 for instance.
Best,
LCT
 
In this post forget all about the fake crop-factored focal length of a 35mm format lens used on the R-D1. The physical properties of the lens itself remain exactly the same on the R-D1 as they would be on a 35mm camera, including things like the distance of the lens elements from the film or sensor. The focal length f will always designate the nominal focal length of the lens designed for the 35mm format.

The hyperfocal distance for a 35mm format lens with focal length f, aperture A, and CoC diameter c is:

h = f^2/(Ac)

Since the diagonal of the image on the R-D1 sensor is only about 28.4mm, it must be enlarged about 1.5x more than a 35mm image in order to make an 8"x10" enlargement. Therefore if the CoC diameter is c for a 35mm image, the equivalent CoC for the R-D1 image would be c/1.5. The hyperfocal distance for this new CoC would be

h = f^2/(Ac/1.5) = 1.5f^2/(Ac) = (1.22f)^2/(Ac)

I.e., the new hyperfocal distance is the same as for a lens having a focal length of 1.22f. This would also be the case for the DoF, since (approximately):

D(near) = hs/(h+s)

D(far) = hs/(h-s)

where s = subject distance.

Therefore if f is the actual focal length of a 35mm format lens used on the R-D1, then the DoF would be about the same as that of a lens (35mm format) having a focal length of 1.22f.
 
Last edited:
There are many, many variables at play here. The only way for an individual photographer to answer this correctly (for him or herself) is to do tests with the actual camera/lens combination(s) in question at various distances with various lens openings. Those test pictures should then be printed at the size the photographer is aiming for. The resolution of many printers, for example, will have an affect on the final, perceived, DOF. The slight bleeding of ink dots will have an effect, etc., etc. - so many possible factors. Experimentation and observation is the only real way to answer this.


Sean
 
Sean Reid said:
There are many, many variables at play here.
Click to expand...
My goodness yes! That's true of everything in photography. My computations were based on the assumption that everything remains constant except the crop factor. Without simplifying assumptions nothing in photography (or physics for that matter) could be computed.
 
Suffice it to compare on a DoF converter or to apply the formula H = f^2/Nc+f where f is the focal length, N the f number and c the circle of confusion (CoC).
For instance, to get an hyperfocal distance of roughly 15 meter through a 50mm lens, you must open at f/8 on the R-D1 (CoC = 0.02mm) and at f/5.6 on a 35mm camera (CoC = 0.03mm).
Best,
LCT
 
So it seems that using the DOF markings of the nearest faster f stop is a safe bet, actually the nearest faster half stop could be enough.
 
LCT said:
Suffice it to compare on a DoF converter or to apply the formula H = f^2/Nc+f where f is the focal length, N the f number and c the circle of confusion (CoC).
For instance, to get an hyperfocal distance of roughly 15 meter through a 50mm lens, you must open at f/8 on the R-D1 (CoC = 0.02mm) and at f/5.6 on a 35mm camera (CoC = 0.03mm).
Best,
LCT
Click to expand...
Exactly. On the R-D1, if the aperture is set at f-number N, the DoF would be approximately that for the f-number N/1.5 using the DoF scale.
 
Then the practical result is what i said above.
You'll get a good approximation if you use the DoF markings of the nearest faster f stop of your lens i.e. f/5.6 when you choose f/8 for metering.
It works for me.
Best,
LCT
 
LCT said:
Then the practical result is what i said above.
You'll get a good approximation if you use the DoF markings of the nearest faster f stop of your lens i.e. f/5.6 when you choose f/8 for metering.
Click to expand...
Yes, you were correct.
 
Just a mini-rant. It can be misleading to talk about effective focal length when a 35mm format lens is used on the R-D1. A 50mm lens on a Leica M7 is still a 50mm lens on the R-D1. The field of view (and only the field of view, nothing else) would be approximately that of a 75mm lens on a 35mm camera. The DoF would be reduced to approximately that of a 60mm lens only because the smaller image size on the R-D1 sensor must be enlarged more to make "normal" size prints. Both the change in field of view and the change in DoF are due only to the smaller size of the image, nothing else. Once a 50mm lens, always a 50mm lens.

Addendum. That was preaching to the choir, since probably everybody contributing to this thread knew that already.
 
Last edited:
He he! Yes a lens is a lens is a lens... :)
It explains why a big nose remains a big nose through a 28mm lens whatever cam we use.
Quite a problem in practice if we don't like playing with the 'Perspective' tool of P'shop... which cannot do much for the big nose anyway :)
 
You must log in or register to reply here.
Back
Top Bottom