Reading mode: Light Dark

stupid 1.6 multipler question

Go to last post
You can use some of the properties of DoF equation to simplify things when doing things in the field. It turns out that things get really simple you think in terms of the hieght of the FoV in the viewfinder.

The tables presented below are approximations (they work as long at the DoF is not very large), but they work well enough for most practical work. If you don't believe me, try FCalc (for those that have FCalc -- the FCalc site seems to be down).

Photographers say that they have “more DoF control” when they can SHORTEN the depth of field. At a given focus distance AND focal length, then the DoF will be roughly proportional to the F-number. In other words the DoF will be about ½ as much at F1.4 as it would at F2.8. Often on portraits of people you want to isolate the person from the background, this both makes the subject stand out and keeps the background from being distractive.

The problem is that the formulas and tables you usually see require on the fly calculations or and it is not easy to figure out what to do to get the DoF you want. It turns out that it can be greatly simplified for most common situations. I will include simple tables below.

DoF is determined by 4 parameters, Focal Length, Focus distance, F-number, and “Circle of Confusion.” CoC takes some explaining, but take my word for it, about 0.016mm is a good number to use for typical D30/D60 use (the CoC is a function of the size of the image on the “negative/sensor” and the size of the output and viewing distance, but most tables assume a 8x10 size output).

DoF can get kind of messy to figure out in the field, but there are some rules and facts that can gets you “close enough” without a lot of computations.

To a first approximation for most “normal situations” DoF:

1. DoF is inversely proportional to the SQUARE of the Focal Length
2. DoF is proportional to the SQUARE of the focus distance
3. DoF linearly proportional to the F-number
4. DoF goes down linearly with the “Circle of Confusion”

What you see is that focal length does dramatically affect DoF due to the Square effect. NOTE THAT TERMS 1 and 2 tend to cancel each other out if you keep the subject the same size, or in other words, if you keep the F-number constant and move back from the subject AND then zoom in to compensate by the same amount so the Subject stays the same size, the DoF will be the SAME. Term 4, the CoC can be considered a constant for a given output size and cropping factor. THUS only term 3, the F-number gives control of the DoF IF you keep the subject the same size.

The Subject/Focus point SIZE in the viewfinder is controlled by the Focal Length and DoF. IF you hold one of these constant and change the SIZE of the subject by changing the other one, the DoF will change by the SQUARE of the size of the subject. Thus when you make a close up, the DoF gets very short very fast. On a wider shot, the DoF will be large almost regardless of the F-number (since F-number only affect DoF linearly where size affects it by roughly square law).

ONE CAN USE THE FACT THAT terms 1 and 2 pretty much cancel each other out (this approximation is true until the DoF gets very large or very short) to build some simple tables. You will NOTE THAT THERE IS NO Focal Length or Subject Distance in the tables as they cancel each other out (YES THIS IS TRUE), only the Height at the Subject/Focus distance. These tables, the first in Feet and the other Metric, factor in the D30/60 viewfinder size. All you have to do is approximate the height of the subject and use the corresponding F-number for the DoF you want. The grayed out areas are where the approximation starts falling apart.

http://www.fototime.com/ {D6DE395C-A0C9-4E3F-B2A6-CF0AD316A0AE} picture.JPG

http://www.fototime.com/ {B4332F02-F808-4EC9-A7A9-C03EE0FE8AD9} picture.JPG

NOTE: you want to have the peak focus on the EYEs of the subject. A lot of people put the NOSE in the center of a close up and end up focusing on it. With a close up, the DoF is relatively short. DoF is about 1/3 in front and 2/3rds behind the focus point. Humans tend to judge focus of a human or animal based on the eyes, so if the eyes are not sharp, the image is by definition out of focus.

Also note that if you are shooting a group of people and thus have a wider FoV at the focus point the DoF goes up dramatically. Generally if you have an arranged group of people (not a lot of depth from person to person), you can use a small F-number, but if you are making a close up, you have to be very careful.

Karl
 
John Mason said:
the focal length is by far the largest factor in depth of field

the formulas show that
Click to expand...
Subject Distance is about the same factor as Focal Length in the DoF equation.

This is sort of like saying that speed is the biggest factor in how long it will take to get some place without including the distance. In the case of DoF, Focal Length, Subject Distance, F-number, and CoC are all factors.
Member said:
CoC calcs also get into pixel size and grain size of film not just
the sensor size.
Click to expand...
No it does not. CoC is a based on the opticals and the amount of magnification of the resultant image. The pixels/grain size is a DIFFERENT issue and has nothing to do with CoC.
Member said:
There are effects there but no where close to the
obvious effects of the smaller focal length needed to achieve any
given perspective with a smaller sensor as compared to a larger
sensor.
Click to expand...
To a first approximation it is the difference in DoF of about 1-Fstop. For a Photographer making portraits who want a shallow DoF, that extra F-stop may mean a very expensive lens, if it can be bought at all.
Member said:
Look at the forumlas, I already have fcalc. Use fcalc and play
with the variables and you'll see this is true.
Click to expand...
I have FCalc and understand the formulas.

--
Karl
 
Doug Walker said:

Murphy said:
AaronT said:
Dumb question, but the 1.6 multipler for the focal-length that the
D30/D60 experiances, does that only effect the field of view (ie
narrows it) or does it actually make images bigger (zoom)? I keep
reading about people saying their big zoom lens is even a longer
zoom lens (like the 70-200mm becomes a 112-320mm) but I can't
honestly see how thats a good thing (unless you're taking a picture
of something small).

Thanks,
Aaron
Click to expand...
When you put a lens on a camera, the lens projects a circular image
onto the imaging plane. (where the film is). This image is of a
"standard" size and is made to be just a bit larger than a
rectangular film negative.
When you put that same lens on a D60, the circular image it creates
where the film is supposed to be is the same size as a normal film
camera.
The problem is that the digital sensor in the D60 (a CMOS) is
slightly smaller than a film negative so its capturing less of the
circle then a film camera captures.
If you multiply the area the CMOS sensor captures by a factor of
1.6, you will end up with a standard 35mm negative.

Now put this into your mind and all your other questions should be
answered by just visualy constucting what is going on in your own
head.

:)) This is how I have come to understand it.

I hope this helps.
Murphy
Click to expand...
--
Doug Walker
And this was supposed to be hobby!
D30 w/BG-ED, Elan 7e, Sigma 15-30, Tokina 17mm ATX PRO, Canon
28/2.8, Canon 28-135 IS, Canon 28-70L, Canon 50mm F1.4, Canon
70-200 F2.8L IS, Canon 100-400L IS, Canon 100 USM macro, Canon 1.4x
teleconverter, Kenko Pro Extension tubes, Arca Swiss B1, Wimberley
Sidekick, Hakuba carbon fiber, 3 550ex, ST-E2, 3 Alien Bees B800
Click to expand...
 
David Kieltyka said:
Ya know, someone with both a D60 and an EOS film SLR could easily
put this to bed by taking photos of the same subject with the same
lens on both cameras. Take the digital photo at various apertures,
then move the camera forward 'til the field-of-view with the film
camera is identical (which you can verify by looking at the digital
photos on the D60's LCD screen), then take the film shots. Get the
film developed, scan the frames and compare the files.
Click to expand...
You've just taken two pictures which look different because they're taken from different distances from the subject.

In order to produce identically framed photos with a D60 and 35mm film SLR, you need (1) the same camera to subject distance, (2) the same field of view. The only way to achieve both (1) and (2) is to use different focal length lenses on each camera (hence the 50mm on D60 vs. 85mm on film comparison - a 50mm on a D60 gives the same FOV as a 80mm on film, which is pretty close to 85mm).

As for the DOF, just look at any portrait taken with a consumer digicam at f/2.8 and 100mm equivalent or so. It's blindingly obvious that you get more DOF with a smaller imager for an identically framed shot. The most common reason people misunderstand this is because for some reason they think of DOF in terms of the image the lens projects onto the sensor/film. DOF is defined in terms of the final print, not the initial image the lens creates.

Over the years, there have been a couple posts of identical pictures taken with a D1/D30/D60 and a 35mm SLR, shot from the same distance to subject, same FOV, and same aperture (hence requiring different focal lengths on the two cameras - usually done with a zoom). I used to think the DOF didn't change either, but I saw one of those comparison pics and realized I was wrong.
 
John Mason said:
So the CoC does change based on film/sensor dimensions, but the
huge overiding factor in all this as shown above is the focal
length of the lens.
Click to expand...
Well, I wouldn't go so far as to call it a "huge overriding factor." DOF decreases as the square of the focal length. DOF increases in proportion to the imager size. Combine these two and for an identically framed shot, DOF decreases in proportion to the imager size (in other words, cut the imager size in half and you double the DOF).
Member said:
As you go from 50mm to 24mm you see a 460% improvement in DOF
solely related to focal length for APS film. You see 455%
improvement in DOF with 35mm film.

Leaving the focal length the same but changing the sensor size from
35mm to APS sized you get an improvement of 20% improvment instead
of a 460% improvment or 1/23rd as much effect as changing the focal
length.
Click to expand...
First, the increase in DOF going from 35mm to APS according to your numbers is 26%-27%, not 20%.

Second, you're comparing a 2:1 reduction in focal length, to a 3:2 reduction in imager size. If you compared a 2:1 reduction of both (or a 3:2 reduction of both), you'd see that the change in DOF goes as the square of the change in focal length, but is proportional to the change in imager size, as I noted above.

Third, you must've made a math error someplace, or are using different DOF formulae for the 35mm and APS cases. The only effect film size has on the DOF equations is in the circle of confusion, which affects DOF linearly. Since you're reducing film size by 3:2, that will change DOF by 3:2, or 50%.

Fourth, something is really screwy with your numbers. A smaller film format (APS) will have less DOF for a given focal length. You need to enlarge a smaller film negative more, which makes the blurry bits blurrier, which means less DOF. Your numbers show greater DOF for APS, which is opposite of what happens if you compare to the same focal length on 35mm.

If you do a 2:1 comparison, going from a 50mm lens to a 25mm lens will (to a first approximation) increase the DOF by 4x. Going from 35mm film to 17.5mm film will decrease the DOF by 2x. Combine these two (the 50mm on 35mm and 25mm on 17.5mm will produce identical FOV), and you get a net 2x increase in DOF.
 
Member said:
The most common reason people
misunderstand this is because for some reason they think of DOF in
terms of the image the lens projects onto the sensor/film. DOF is
defined in terms of the final print, not the initial image the lens
creates.
Click to expand...
Another reason is that many people still think that DOF is some well defined area that is written on their older lenses: "if Leica writes that on their lenses, it must be true!" And as you said they believe that the output size and viewing distance doesn't have any influence on this.

Severi
 
John Kim said:
John Mason said:
So the CoC does change based on film/sensor dimensions, but the
huge overiding factor in all this as shown above is the focal
length of the lens.
Click to expand...
Well, I wouldn't go so far as to call it a "huge overriding
factor." DOF decreases as the square of the focal length. DOF
increases in proportion to the imager size. Combine these two and
for an identically framed shot, DOF decreases in proportion to the
imager size (in other words, cut the imager size in half and you
double the DOF).
Click to expand...
Very true for most practical uses (I believe if varies at very close and far focus a bit). Furthermore DoF goes up with the square of the distance to the focus point and up linearly with the F-number.
Member said:
Fourth, something is really screwy with your numbers. A smaller
film format (APS) will have less DOF for a given focal length.
You need to enlarge a smaller film negative more, which makes the
blurry bits blurrier, which means less DOF. Your numbers show
greater DOF for APS, which is opposite of what happens if you
compare to the same focal length on 35mm.
Click to expand...
First his Rows are swapped. The numbers for the 35mm are the APS numbers and visa versa. Second he "rounded down" everything (even if the next digit was 9).
Member said:
If you do a 2:1 comparison, going from a 50mm lens to a 25mm lens
will (to a first approximation) increase the DOF by 4x. Going from
35mm film to 17.5mm film will decrease the DOF by 2x. Combine
these two (the 50mm on 35mm and 25mm on 17.5mm will produce
identical FOV), and you get a net 2x increase in DOF.
Click to expand...
Another good thing to try is using the same film/sensor size (and thus the same CoC), and the same F-number, Double the Focal length and Double the Subject Distance. The Feild of view at the focus point AND the DoF will be very close to the same.

Example using FCalc:

50mm, 35mm film (CoC=02501 -- FCalc's Default), Distance=8ft, F2.8
-> Total DoF = 1.086260 ft
-> Vertical FoV = 3.761260 ft Horizontal FoV= 5.641890 ft

100mm, 35mm Film, Distance = 16ft, F2.8
-> Total DoF = 1.082535 ft (less than 1% different)
-> Vertical FoV = 3.761260 ft Horizontal FoV= 5.641890 ft

What this shows is that to "Control" the DoF, KEEPING the subject the SAME SiZE, you can't get there by changing the focal length and moving the camera to keep the subject the same size. Using the same film/sensor size (and thus the same CoC), the ONLY whay to get there is to change the F-number.

Thus it is often said that the "F-number controls the DoF." But what a lot of people miss is that the SIZE of the FoV at the focus distance controls it by a the Square of the size of the FoV.
--
Karl
 
Of couse DOF varies most by apeture size - the subject I was disscusing was which is more important - sensor size or focal length - or more specifically does sensor size equate to DOF at all. Everyone jumped on the Canon engineer when in fact what he said was quite correct.

I take you and anyone else on a very simple mental excersize:

1. Take a shot of cards mounted along a rail and slightly offset so you can see every card in a line. These cards are mounted a few inches apart.

2. Take this shot at a given apeture and focal length

3. Take the shot again at a different focal length and same apeture

Print out both shots. Weather you crop some of the picture to change the view or not doesn't change which cards are in focus. So in a real sense sensor size doesn't change DOF at all. Moving my eye around the picture and looking at one card to the next does not change the range of focus on these printed pictures. Taking a pair of sissors and cutting the picture after it's printed does not magically increase or decrease what is in focus.

So in one perspective sensor size itself has nothing to do with DOF from the point of view of physics. (don't jump or confuse yourself now, think of a picture you have then think of sissors and cutting it up, slicing your picture up does not increase the DOF)

But, everyone knows that a G2 has lots more Depth of Field than a D60. Well, this must be since it's sensor size is smaller - right? Well, no, because the sensor size being smaller is exactly like a crop. It is because for a given perspective result and field of view of a subject I must use a wider focal length as my sensor size gets smaller to get the same result as a camera with a bigger sensor. This is why what I'm saying is true and when people say if the sensor size shrinks DOF goes up is also true. But cutting a picture up (cropping) does not make more things in the pieces you have left any sharper than the picture you started with.

I know people will jump on me with this for whatever reason they choose, but people that aren't so brainwashed will understand the simple excersize and parable of printing a picture and cutting (cropping) it with scissors and will realize what is the best way to think and understand this DOF issue.

Flame on!
 
what a way to take things out of context

The discussion was about sensor size vs focal length as it relates to DOF

Thanks for finding my errors and totally missing the point
 
Member said:
Print out both shots. Weather you crop some of the picture to
change the view or not doesn't change which cards are in focus. So
in a real sense sensor size doesn't change DOF at all. Moving my
eye around the picture and looking at one card to the next does not
change the range of focus on these printed pictures. Taking a pair
of sissors and cutting the picture after it's printed does not
magically increase or decrease what is in focus.
Click to expand...
You continue to overlook enlargement. After a crop, you must enlarge the print to an 8x10 and view at two feet. It is not an apples to apples comparison if you simply crop an 8x10 to a 4x5. A 4x5 is NOT an 8x10. You must enlarge the cropped 4x5 to an 8x10 and then compare to other 8x10 prints.

Remember, nothing is in focus except objects on the focus plane. In other words, DoF is related to perception. That is, objects "appear" to be in focus. For objects to "appear" to be in focus, we must view the print. When viewing prints, we must "standardize" the viewing. CoC and DoF calculations assume a standard print size and viewing distance. You can not simply crop an 8x10 to a smaller print and view the smaller print. After the crop, you must enlarge the smaller cropped image to an 8x10 and view at two feet.
  • Christopher
 
I just did a search on my post and the word "aperture" isn't even in there. I don't see where you got the idea I was talking about aperture. I assumed you were holding aperture constant. Everything I posted was about how sensor size and focal length changes DOF.
John Mason said:
what a way to take things out of context

The discussion was about sensor size vs focal length as it relates
to DOF

Thanks for finding my errors and totally missing the point
Click to expand...
 
We're just trying to correct your understanding of this, as well as cut down any damage your incorrect explanation might be doing to others trying to understand this. I doubt any of us hold any personal animosity towards you.
John Mason said:
So in one perspective sensor size itself has nothing to do with DOF
from the point of view of physics. (don't jump or confuse yourself
now, think of a picture you have then think of sissors and cutting
it up, slicing your picture up does not increase the DOF)
Click to expand...
The claim you're making - that using a camera with a smaller sensor will result in a larger DOF if you switch focal length lenses to maintain FOV - is true if you just state it like that without trying to explain it. Even your claim that the change in focal length is the predominant factor is true. If you'd stuck to that you'd be just fine. People who want the one-sentence answer without bothering to understanding the nuts and bolts will do just fine with those statements.

Unfortunately you're not just stating it. You're also trying to explain it, and your explanation is just plain wrong. This is driving those of us who like proper explanations nuts, because your incorrect explanation will confuse those who want to understand the nuts and bolts of what's really going on.

Contrary to your explanation, "from the point of view of physics," sensor size does affect DOF. Smaller sensors reduce DOF. If you have a smaller sensor, you need to enlarge more to produce a (say) 8x10 print, which makes the blurry bits blurrier, which is just another way of saying DOF has decreased. The amount of decrease is proportional to the reduction in sensor size - halving the sensor size halves the DOF.

The smaller focal length you need to maintain the same FOV with that smaller sensor results in increased DOF. The magnitude of the increase is the inverse square of the change in focal length - halving focal length quadruples DOF.

When you combine these two, the net effect is that DOF increases proportionally to the reduction in sensor size for identical FOV shots - halve the sensor size and you double the DOF.
Member said:
But, everyone knows that a G2 has lots more Depth of Field than a
D60. Well, this must be since it's sensor size is smaller - right?
Well, no, because the sensor size being smaller is exactly like a
crop. It is because for a given perspective result and field of
view of a subject I must use a wider focal length as my sensor size
gets smaller to get the same result as a camera with a bigger
sensor. This is why what I'm saying is true and when people say if
the sensor size shrinks DOF goes up is also true. But cutting a
picture up (cropping) does not make more things in the pieces you
have left any sharper than the picture you started with.
Click to expand...
As Christopher has pointed out, you're neglecting the enlargement stage. Cropping and enlarging makes things blurrier. If you try to ignore this, people who insist a D60 has the same DOF as a EOS 1 will try to claim the effect of smaller sensor and shorter f.l. will cancel out, yielding the same DOF for both cameras. The two effects are in opposite directions, but the f.l. effect is second order while the sensor size effect is first order, resulting in a net first order effect of the D60 having more DOF.
 
John Kim said:
You've just taken two pictures which look different because they're
taken from different distances from the subject.
Click to expand...
Yep, this is true.

Note that I didn't suggest taking comparison photos because I was in any doubt of the outcome. :-) I'd just rather see people actually find out what the situation is rather than throw theory at each other.

-Dave-
 
Few Qs for Karl.

Consider first only a film camera, which is full frame. Consider the distance from the back of the lens to the film plane as a fixed distance, say 50mm.

1) If, with that film camera, I take a photo using a 35mm lens at f/2 and make a print it will have a defined, measurable, viewable DOF, no?

2) If then I cut out the middle of that print, did that suddenly change the DOF of my print?

3) Is not the d60 backside of lens to film plane distance the same as in a film camera? (In this example, 50mm)

4) Does the lens used above (35mm f/2) change in any way when I mount it on a D60?

5) Thus, is not that same lens, separated by the same distance from the film/CCD plane creating the exact same size circle of confusion in both the d60 and film bodies?

6) Is not the d60 sensor size a crop?

7) Is not the DOF on that cropped image the same as the cutout of the print in Q (2)?
 
Jason Hutchinson said:
Does the G2 have more DOF than film? Everyone would answer "yes,
of course!" Well, the G2 is NO different than the D60, just a
smaller crop. Not at all different in any relevant way.
Click to expand...
Not quite. The G2 lens is designed only to produce a circle of confusion that is about the size of it's sensor. While it may also do some cropping, the lens on the G2 is not the same as a lens on the d60 or a film body.

The d60 does not change the size of the circle of confusion produced by lenses mounted to it. The only thing it does is crop (capture less).

See my questions to Karl above. The DOF is not different on the D60 than it is for film. The simplest explanation is, does the DOF change in a 35mm film print when you cutout the middle part? Nope.

--
Photos, tips and tests at:
http://www.geocities.com/glowluzid
 
glowluzid said:
2) If then I cut out the middle of that print, did that suddenly
change the DOF of my print?
Click to expand...
Actually, it does if you enlarge the center you cropped to the size of the original image. If you don't, no change.

Of course, with the D60 you wouldn't take that shot with a 35mm lens, you'd take it with a 22mm lens, which is how you'd get your increased DOF.
glowluzid said:
3) Is not the d60 backside of lens to film plane distance the same
as in a film camera? (In this example, 50mm)
Click to expand...
Yes.
glowluzid said:
4) Does the lens used above (35mm f/2) change in any way when I
mount it on a D60?
Click to expand...
No, you'd use the 22mm lens described before though unless you don't want to see the head of your subject.
glowluzid said:
5) Thus, is not that same lens, separated by the same distance from
the film/CCD plane creating the exact same size circle of confusion
in both the d60 and film bodies?
Click to expand...
Nope.
glowluzid said:
6) Is not the d60 sensor size a crop?
Click to expand...
Yes.
glowluzid said:
7) Is not the DOF on that cropped image the same as the cutout of
the print in Q (2)?
Click to expand...
You've left out too many variables such as what you've done to the image after cropping to say. See question 2's answer.
 
glowluzid said:
7) Is not the DOF on that cropped image the same as the cutout of
the print in Q (2)?
Click to expand...
You have neglected the affects of enlarging cropped images. It is important to understand that DoF is about perceived focus (objects off the focus plane are not really in focus, they just "appear" to be in focus when they are close enough to the focus plane). So, how close is close enough? To standardize the DoF calcuations, we must standardize the viewing of the prints (DoF does not make sense in the absence of viewing prints). It is a commonly accpeted practice to use 8x10 prints viewed at two feet (please note some other print size and distance could be used, but it is important that all prints be the same size and viewing distance).
  • Christopher
 
I hope the light bulb goes off in Glowluzid's head, so the thread can die.

Jason
Christopher Glaeser said:
Christopher Glaeser said:
7) Is not the DOF on that cropped image the same as the cutout of
the print in Q (2)?
Click to expand...
You have neglected the affects of enlarging cropped images. It is
important to understand that DoF is about perceived focus (objects
off the focus plane are not really in focus, they just "appear" to
be in focus when they are close enough to the focus plane). So,
how close is close enough? To standardize the DoF calcuations, we
must standardize the viewing of the prints (DoF does not make sense
in the absence of viewing prints). It is a commonly accpeted
practice to use 8x10 prints viewed at two feet (please note some
other print size and distance could be used, but it is important
that all prints be the same size and viewing distance).
  • Christopher
Click to expand...
 
Christopher Glaeser said:
It is
important to understand that DoF is about perceived focus
Click to expand...
Christopher Glaeser said:
... (DoF does not make sense in the absence of viewing prints).
Click to expand...
I see, now. You guys are talking about "perceived" DOF.

I was talking actual optical DOF. (The one that is dictated only by the lens and film plane, not by subjectivity of viewers).

Melles Griot (you know, optics company) gives the following definition of CoC
http://www.mellesgriot.com/glossary/wordlist/glossarydetails.asp?wID=132
"Circle of Confusion

The image of a point source that appears as a circle of finite diameter because of defocusing or the aberrations inherent in an optical system."

This definition has nothing to do with the viewer, enlargement, etc. It is all about the PSF of the optics...

But I will accept defeat on the meaning of DOF here. It is a perceived difference. In that sense, users of the D60 may very well perceive a different DoF than film users. Although, optically, it is the same.
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top