Ad
Ad
Ad
Pages: [1] 2 3 ... 8 »   Bottom of Page
Print
Author Topic: Newbe question 35mm vs MF and dof  (Read 22468 times)
HCHeyerdahl
Newbie
*
Offline Offline

Posts: 47


WWW
« on: December 31, 2010, 10:40:23 AM »
ReplyReply


Say I take a Leica S2 with a 70mm f2.5 and compare shots with a Nikon d3x with a 50mm f 1.4, how will the dof compare:
At what f stop will the Nikon dof be approx equal to the Leica @ f 2,5?
Will this change if Nikon comes with a D4x at say 30 mp?
Is there som kind of multiplyer I can use across focalranges to get som idea of the dof similarities/differences?

Christopher
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7658


WWW
« Reply #1 on: December 31, 2010, 10:58:53 AM »
ReplyReply

Hi,

I didn't do the math but I guess about f/1.7. DoF will be same on a D4X regardless of MP, albeit you may get a bit more demanding. Anyway the difference between 30 MP and 24.5 MP is quite ignorable, it's about 10% on linear scale.

You can assume something like one stop difference.

Focusing is quite critical, BTW, and you cannot really rely on AF for pinpoint focus, but neither can you rely on your eyes. Live view is probably the best way to achieve dead on focus.

Best regards
Erik

Say I take a Leica S2 with a 70mm f2.5 and compare shots with a Nikon d3x with a 50mm f 1.4, how will the dof compare:
At what f stop will the Nikon dof be approx equal to the Leica @ f 2,5?
Will this change if Nikon comes with a D4x at say 30 mp?
Is there som kind of multiplyer I can use across focalranges to get som idea of the dof similarities/differences?

Christopher
Logged

David Klepacki
Full Member
***
Offline Offline

Posts: 185


« Reply #2 on: December 31, 2010, 10:59:12 AM »
ReplyReply

http://www.dofmaster.com/
Logged
HCHeyerdahl
Newbie
*
Offline Offline

Posts: 47


WWW
« Reply #3 on: December 31, 2010, 12:17:48 PM »
ReplyReply

Thanks!

However,  I don`t think any of the cameras there are MF?

Christopher
Logged
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7658


WWW
« Reply #4 on: December 31, 2010, 12:37:58 PM »
ReplyReply

Hi,

DoF is not really related to format, there are four parameters Circle of Confusion (CoC), focal length, aperture and focusing distance. The image size may have to do with CoC. With a larger format you would enlarge the sensor image less for a given print size, so you may accept a larger CoC.

My best suggestion is to ignore DoF. Focus on what supposed to be sharp and stop down, hoping for the best. Avoid stopping down beyond f/16 if possible because you start loosing sharpness massively due to diffraction.

You may check:

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/29-handling-the-dof-trap

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures

Best regards
Erik


Thanks!

However,  I don`t think any of the cameras there are MF?

Christopher
Logged

Doug Peterson
Sr. Member
****
Online Online

Posts: 2838


WWW
« Reply #5 on: December 31, 2010, 12:40:24 PM »
ReplyReply

However,  I don`t think any of the cameras there are MF?

An informed medium format dealer can help you establish an appropriate CoC value for a specific digital back relative to the 35mm dSLR you are comparing to. Due to the omission of an AA filter there is a slight fudge factor in addition to the different micron size.

DOF depends on a large variety of factory including micron size of the sensor, print size and intended viewing distance (if you wish to go based on a print size rather than 100% pixel sharpness), type of sensor (CMOS w/AA filter or CCD without), use of tilt (if any*), and sensor size.

If you work with a good dealer you won't have to trudge through such questions on your own :-).

*sure DOF doesnt change in technical terms but placing the plane of focus in line with your subject (when possible) is effectively the same as increasing DOF.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Leaf, Leica, Cambo, Arca Swiss, Canon, Apple, Profoto, Broncolor, Eizo & More

National: 877.217.9870  |  Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off
Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
David Klepacki
Full Member
***
Offline Offline

Posts: 185


« Reply #6 on: December 31, 2010, 02:26:24 PM »
ReplyReply

DOF depends on a large variety of factory including micron size of the sensor, print size and intended viewing distance (if you wish to go based on a print size rather than 100% pixel sharpness), type of sensor (CMOS w/AA filter or CCD without), use of tilt (if any*), and sensor size.

If you work with a good dealer you won't have to trudge through such questions on your own :-).

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration

This is very interesting, especially about how DOF depends on print size.   No matter how large or small I print my images, I don't see any change to the DOF.  Can you elaborate more on how you find DOF to depend on print size?

Also, I have taken identical image captures from a 36x48 CCD back with 9 micron pixels (Hasselblad CF22) and when compared to identical images taken with the same camera and lens but with a 36x48 CCD back having 7.2 micron pixels (Sinar e75LV), there is no difference in DOF either.  You probably have more experience with many other digital backs.  So, can you elaborate on your experience where you find the DOF to depend on the pixel size and not just the sensor size?

From what I can see in my own images, the DOF seems to depend only on the apparent object size relative to the sensor size.  It would be great if you could explain better how these other print and sensor issues also affect the DOF.

Thanks.
Logged
tho_mas
Sr. Member
****
Offline Offline

Posts: 1696


« Reply #7 on: December 31, 2010, 02:38:35 PM »
ReplyReply

This is very interesting, especially about how DOF depends on print size.   No matter how large or small I print my images, I don't see any change to the DOF.  Can you elaborate more on how you find DOF to depend on print size?
You can "simulate" this when you downsize your photos on the monitor. Let's say your image has a wide DOF from near distance to infinity but the very foreground is actually a bit soft. Now when you downrez your image to, say, 32% the foregound might appear sharp (the effect is immediately visible when you downsize your 39MP monster to 800x600 pixel for web puposes).
It's pretty much the same with print size.
Also comes into play when uprezzing... the more you enlarge your photo the smaller the DOF will be. Simply because the high contrast in the focus plane enlarges "better" (i.e. "sharper") than the somewhat lower contrast at the near + far end of DOF. Actually it's just a visuell effect... but, well, it's visible :-)

« Last Edit: December 31, 2010, 02:41:19 PM by tho_mas » Logged
David Klepacki
Full Member
***
Offline Offline

Posts: 185


« Reply #8 on: December 31, 2010, 02:53:42 PM »
ReplyReply

I understand your argument, but I still don't think you can claim the DOF has really changed, only that your perception of it perhaps, and even then it is a stretch of meaning.

If DOF truly changes with print size, then it should be possible to take any image captured at F1.0 and print it such that it appears to have the DOF as if captured at F22.

Logged
tho_mas
Sr. Member
****
Offline Offline

Posts: 1696


« Reply #9 on: December 31, 2010, 02:59:04 PM »
ReplyReply

I understand your argument, but I still don't think you can claim the DOF has really changed, only that your perception of it perhaps, and even then it is a stretch of meaning.

If DOF truly changes with print size, then it should be possible to take any image captured at F1.0 and print it such that it appears to have the DOF as if captured at F22.
ah, okay, I get it.
Yes, I was only referring to perception...
Your f1.0 image will look like captured at f22 when you print it at stamp size :-)
Logged
Doug Peterson
Sr. Member
****
Online Online

Posts: 2838


WWW
« Reply #10 on: December 31, 2010, 03:01:52 PM »
ReplyReply

This is very interesting, especially about how DOF depends on print size.   No matter how large or small I print my images, I don't see any change to the DOF.  Can you elaborate more on how you find DOF to depend on print size?

Also, I have taken identical image captures from a 36x48 CCD back with 9 micron pixels (Hasselblad CF22) and when compared to identical images taken with the same camera and lens but with a 36x48 CCD back having 7.2 micron pixels (Sinar e75LV), there is no difference in DOF either.  You probably have more experience with many other digital backs.  So, can you elaborate on your experience where you find the DOF to depend on the pixel size and not just the sensor size?

From what I can see in my own images, the DOF seems to depend only on the apparent object size relative to the sensor size.  It would be great if you could explain better how these other print and sensor issues also affect the DOF.

I define Depth of Field as the inability to distinguish (in a meaningful/practical sense) between the sharpness of point A and the sharpness of point B.

It's easiest to see the effect of this with extreme examples.

Print a 60mp file as a postage stamp (say a 30mm square) and as a 3 meter print.

Even the very best (commercial photographic) printing technologies cannot get 60mp of detail onto the postage stamp. Look at a slightly out of focus area on the 3 meter print then look at the same subject area on the postage stamp. Even with a very close viewing distance (or a magnifying glass) you could not tell that this area of the subject was any less in focus than the sharpest part of the image.

Same goes with different micron sizes. Taking extremes again a 5.2 micron sensor (Aptus II 12) and a 12 micron sensor (H20): something that is slighlty out of focus on the 5.2 micron sensor cannot be shown as out of focus on the 12 micron sensor. So DOF will be moderately higher on the H20 (and the absolute level of detail in the in-focus areas will be higher in the Aptus II 12).

Said differently: DOF is infinite on a one pixel image. DOF is non-existent on a sensor with 1000 gigpixel resolution - the moment you look outside the plane of focus (at 100% on a monitor) you'll see the pixels are no longer as sharp as they were at the plane of focus.

Said differently again: If a camera/print is capable of resolving more detail then you'll more easily notice when it does not.

The difference between 7.2 and 9 microns will be subtle and will only appear if the 7.2 micron sensor is used with a lens fully capable of resolving on 7.2 micron pixels.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Leaf, Leica, Cambo, Arca Swiss, Canon, Apple, Profoto, Broncolor, Eizo & More

National: 877.217.9870  |  Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off
Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
Doug Peterson
Sr. Member
****
Online Online

Posts: 2838


WWW
« Reply #11 on: December 31, 2010, 03:08:07 PM »
ReplyReply

I understand your argument, but I still don't think you can claim the DOF has really changed, only that your perception of it perhaps, and even then it is a stretch of meaning. If DOF truly changes with print size, then it should be possible to take any image captured at F1.0 and print it such that it appears to have the DOF as if captured at F22.

As tho_mas says I'm dealing with "perception" as my definition (the ability to - in a practical sense - distinguish between the levels of sharpness of two points).

Depending on how you define it you have
- DOF of the raw file
- DOF of a given print

Getting away from extremes...
With a very high resolution file, such as a four image stitch from a P65+ on a tech camera it is very possible that an 11x14 print (even when viewed close) will show the entire field of view as equally in focus (DOF from front to back) but a 30x40 will show the front of the image is just slightly out of focus compared to the detail at mid-range (DOF does not quite extend front to back).

It is the above scenario that originally piqued my interest in the more-complicated-than-I-was-taught-in-school topic of Depth of Field and Sharpness.

In my experience most photographers define DOF as "where it's sharp at 100%" on the monitor. However, as the size of raw files goes up I beg us to consider moreso the application for which the file is being used. On shoots for the web* it is perfectly acceptable to use the digital loupe in Capture One / Aperture / LR etc at 25% to see if it's "sharp" rather than examining 100% detail. If the subject is slightly soft at 100% in an 80 megapixel raw file it will still appear indistinguishable from something "sharper" when processed at a 800x600 for e-commerce.

*If there is a moderate chance of having the images re-used for large prints then obviously it needs to be sharp at 100% pixel view.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Leaf, Leica, Cambo, Arca Swiss, Canon, Apple, Profoto, Broncolor, Eizo & More

National: 877.217.9870  |  Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off
« Last Edit: December 31, 2010, 03:10:25 PM by dougpetersonci » Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
tho_mas
Sr. Member
****
Offline Offline

Posts: 1696


« Reply #12 on: December 31, 2010, 03:18:45 PM »
ReplyReply

If there is a moderate chance of having the images re-used for large prints then obviously it needs to be sharp at 100% pixel view.
exactly! When you enlarge the image 300% and print it at 300ppi than the actual pixel size (i.e. 100%) of your original (unenlarged) image infact represents the real outcome on your ~100ppi monitor quite good (of course it looks "different", but it gives you a good idea about the apperance of the print re DOF).
Logged
HCHeyerdahl
Newbie
*
Offline Offline

Posts: 47


WWW
« Reply #13 on: December 31, 2010, 03:22:57 PM »
ReplyReply

Hi,

DoF is not really related to format, there are four parameters Circle of Confusion (CoC), focal length, aperture and focusing distance. The image size may have to do with CoC. With a larger format you would enlarge the sensor image less for a given print size, so you may accept a larger CoC.

My best suggestion is to ignore DoF. Focus on what supposed to be sharp and stop down, hoping for the best. Avoid stopping down beyond f/16 if possible because you start loosing sharpness massively due to diffraction.

You may check:

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/29-handling-the-dof-trap

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures

Best regards
Erik



Hmmmm,
I am not sure I really understand this.
Thing is, I very often use my Nikon at f 9-11 which I understand is aproaching the diffraction limit. Does that make a swap for a leica sort of pointless since it would have to be used at f 22 to get same dof and thus diffraction will eat upp the improvement in number of pixels?
Assume the same print size and viewing distance.

Christopher
« Last Edit: December 31, 2010, 03:25:39 PM by ChristopherHeyerdahl » Logged
David Klepacki
Full Member
***
Offline Offline

Posts: 185


« Reply #14 on: December 31, 2010, 03:38:03 PM »
ReplyReply

As tho_mas says I'm dealing with "perception" as my definition (the ability to - in a practical sense - distinguish between the levels of sharpness of two points).

Depending on how you define it you have
- DOF of the raw file
- DOF of a given print

Getting away from extremes...
With a very high resolution file, such as a four image stitch from a P65+ on a tech camera it is very possible that an 11x14 print (even when viewed close) will show the entire field of view as equally in focus (DOF from front to back) but a 30x40 will show the front of the image is just slightly out of focus compared to the detail at mid-range (DOF does not quite extend front to back).

It is the above scenario that originally piqued my interest in the more-complicated-than-I-was-taught-in-school topic of Depth of Field and Sharpness.

In my experience most photographers define DOF as "where it's sharp at 100%" on the monitor. However, as the size of raw files goes up I beg us to consider moreso the application for which the file is being used. On shoots for the web* it is perfectly acceptable to use the digital loupe in Capture One / Aperture / LR etc at 25% to see if it's "sharp" rather than examining 100% detail. If the subject is slightly soft at 100% in an 80 megapixel raw file it will still appear indistinguishable from something "sharper" when processed at a 800x600 for e-commerce.

*If there is a moderate chance of having the images re-used for large prints then obviously it needs to be sharp at 100% pixel view.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Leaf, Leica, Cambo, Arca Swiss, Canon, Apple, Profoto, Broncolor, Eizo & More

National: 877.217.9870  |  Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off


No, this is all wrong.  You cannot arbitrarily change the definition of DOF, just as I can't arbitrarily decide to change the definition of a circle.  DOF already has a universally accepted definition that can be found in most textbooks.

To prevent an unhealthy and possibly drawn out thread, I will cut to the heart of the issue.  What you are describing as "DOF" is known as "viewing resolution".  Viewing resolution is a perceptual quantity that indeed depends on print size, viewing distance, pixel size and sensor size.  However, DOF does not depend on any of these things, except the sensor size as it relates to the object size as measured at the sensor.

My only point in this thread is to please not confuse these terms, especially to someone new here.  There is already so much confusion on the internet with regard to the understanding of DOF.
Logged
David Klepacki
Full Member
***
Offline Offline

Posts: 185


« Reply #15 on: December 31, 2010, 03:44:39 PM »
ReplyReply

Hmmmm,
I am not sure I really understand this.
Thing is, I very often use my Nikon at f 9-11 which I understand is aproaching the diffraction limit. Does that make a swap for a leica sort of pointless since it would have to be used at f 22 to get same dof and thus diffraction will eat upp the improvement in number of pixels?
Assume the same print size and viewing distance.

Christopher

Basically, if you need more depth of field with the same angle of view, then moving to a larger format will make that goal harder to achieve.
Logged
Doug Peterson
Sr. Member
****
Online Online

Posts: 2838


WWW
« Reply #16 on: December 31, 2010, 03:48:55 PM »
ReplyReply

To prevent an unhealthy and possibly drawn out thread, I will cut to the heart of the issue.  What you are describing as "DOF" is known as "viewing resolution".  Viewing resolution is a perceptual quantity that indeed depends on print size, viewing distance, pixel size and sensor size.  However, DOF does not depend on any of these things, except the sensor size as it relates to the object size as measured at the sensor.

Call it what you want, but if you want to present an image with sharp detail from front to back (as the OP does) then pixel size, print size, viewing distance, and sensor size all matter :-).
Logged

DOUG PETERSON (dep@digitaltransitions.com), Digital Transitions
Dealer for Phase One, Mamiya Leaf, Arca-Swiss, Cambo, Profoto
Office: 877.367.8537
Cell: 740.707.2183
Phase One IQ250 FAQ
Chris Livsey
Sr. Member
****
Offline Offline

Posts: 285



« Reply #17 on: December 31, 2010, 03:53:40 PM »
ReplyReply

Christopher

The first point to appreciate is that any lens can only focus at a precise distance, at that plane, often it is not a flat plane which would be best for flat copy work hence "planar" lenses, but curved, it is sharp and in focus at exactly that distance no more or less. The decrease in sharpness both in front and behind that point/plane is gradual. The distance where it becomes sufficiently unsharp, out of focus, to be noticeable is the problem. You need to define where unsharp is which is where circle of confusion comes in and that varies with the size of image you are viewing cf Doug's stamp and poster, (note not the format) the distance you are viewing the image from and now to the lens bit, focal length and aperture (crudely, as you know less DOF at f1 than f8).

Note that a point and shoot will give an image that appears to be "sharp" (using a generous definition of sharp !) over a very wide depth of field. That is usually because the system is such that no sharp focus distance point is seen (the image is usually diffraction limited and "blurred") so a transition to unsharp is not seen, it all appears the same.

What this boils down to is depth of field is what you say it is. One photographers acceptable image is not anothers and all the equations in the world won't allow for subjective judgement.

Those lovely calculators, and the markings on your lenses, all make assumptions about what is sharp and where it becomes unsharp based on a print size and viewing distance that they will not usually quote. They may look like hard figures but they are not in real life.

No one has talked about focus stacking BTW which is really neat way of demonstrating depth of field and what it looks like when you "fiddle" it away.

Logged

ATB
Chris Livsey
http://www.flickr.com/photos/red_eyes_man/

It's not what you look at that matters, it's what you see.
David Klepacki
Full Member
***
Offline Offline

Posts: 185


« Reply #18 on: December 31, 2010, 03:54:25 PM »
ReplyReply

ah, okay, I get it.
Yes, I was only referring to perception...
Your f1.0 image will look like captured at f22 when you print it at stamp size :-)

The viewing resolution changes, but the DOF remains the same.

Logged
David Klepacki
Full Member
***
Offline Offline

Posts: 185


« Reply #19 on: December 31, 2010, 03:56:50 PM »
ReplyReply

Call it what you want, but if you want to present an image with sharp detail from front to back (as the OP does) then pixel size, print size, viewing distance, and sensor size all matter :-).

Exactly, since the only way to achieve such sharp detail is by understanding how viewing resolution is affected by these parameters.

Logged
Pages: [1] 2 3 ... 8 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad