Pages: [1]   Go Down

Author Topic: DoF article  (Read 6535 times)

larsrc

  • Full Member
  • ***
  • Offline Offline
  • Posts: 173
    • http://
DoF article
« on: January 18, 2007, 10:03:10 am »

A quite illuminating article, at least for me.  I had been wondering why the little throw-away cameras could manage so well in low light, now I know.  

There's some missing text towards the end, I guess some tags are messed up.  Particularly the sentence "The computed parameters for the S80 were and ; however, I settled for , the maximum focal length, and , the minimum value available at full zoom." is less than informative:)
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
DoF article
« Reply #1 on: January 18, 2007, 11:01:20 am »

The figures also don't show up for me.

Regards,
Bernard

apq65

  • Newbie
  • *
  • Offline Offline
  • Posts: 12
DoF article
« Reply #2 on: January 18, 2007, 11:14:50 am »

The article by Charles Johnson appears overall well written and researched. I would precede it, however, with the note that the concept of a crop factor is not incorrect and certainly a concept that is easy to comprehend. In other words, putting a 35mm lens on a less-than-full frame camera is equivalent to trimming a print obtained from the full frame image with a pair of scissors. That includes the depth of field (obviously, it would not change by trimming the picture). The scissors are equivalent to the light that ends up on the camera interior instead of on the sensor if the sensor is smaller than full frame. The caveat of this approach is that most photographers think (and compose the image) by choosing a certain focal length for any given distance to the object, thus obtaining a given perspective, depth of focus, etc.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
DoF article
« Reply #3 on: January 18, 2007, 11:21:09 am »

Quote
A quite illuminating article, at least for me.  I had been wondering why the little throw-away cameras could manage so well in low light, now I know. 

There's some missing text towards the end, I guess some tags are messed up.  Particularly the sentence "The computed parameters for the S80 were and ; however, I settled for , the maximum focal length, and , the minimum value available at full zoom." is less than informative:)
[a href=\"index.php?act=findpost&pid=96354\"][{POST_SNAPBACK}][/a]

I suppose you are talking about the most recent "Lens Equivalent" article by Charles Johnson.

This is really a terrible article to read. It's almost as though someone is trying to cram everything they know about a topic into a single article, John Cleese style.

I'm not sure I can pick any faults in the information presented. There might be some contention about choice of the diagonal of the sensor for DoF comparisons, when different aspect ratios are involved. This can tend to favour the squarer format.

There was an important point made about perspective, which perhaps needs to be repeated, and that is, it has more to do with distance to subject than lens focal length. Users of cropped format cameras such as the 20D and 30D should be aware that a 50mm lens used at the same distance as an 80mm lens with a 5D, will produce the same 'perspective', for portraiture for example.
Logged

michael

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5084
DoF article
« Reply #4 on: January 18, 2007, 11:27:56 am »

There are some formatting issues with the article when viewed on IE Windows (figures). I'll try and fix it ASAP.

Michael
Logged

Ken Tanaka

  • Full Member
  • ***
  • Offline Offline
  • Posts: 134
    • http://www.KenTanaka.com
DoF article
« Reply #5 on: January 18, 2007, 12:19:59 pm »

Quote
There are some formatting issues with the article when viewed on IE Windows (figures). I'll try and fix it ASAP.

Michael
[a href=\"index.php?act=findpost&pid=96378\"][{POST_SNAPBACK}][/a]
The formatting itself appears ok on Safari.  But there are some numbers missing from the text.

4th Paragraph from the end:
Quote
The computed parameters for the S80 were _______ and______ ; however, I settled for __________, the maximum focal length, and ________, the minimum value available at full zoom.
Fill in the blanks.

It looks like Charles just forgot to insert the results.
Logged
- Ken Tanaka -
 www.KenTanaka.com

E.J. Peiker

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 891
    • http://www.ejphoto.com
DoF article
« Reply #6 on: January 18, 2007, 04:15:13 pm »

It is a difficult article to read especially without the figures for IE Explorer users which hopefully get fixed.  But also interesting if the time is taken to unerstand the points being made.
Logged

Eric Myrvaagnes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 22814
  • http://myrvaagnes.com
    • http://myrvaagnes.com
DoF article
« Reply #7 on: January 18, 2007, 04:29:45 pm »

By the time I got to it, all the missing pieces were there, and the figures all show clearly (in Opera -- the best of all possible browsers     )

P.S. The best fix for IE users is to get a better browser (Opera, Firefox, Safari, Netscape, . . . ) all of which are less expensive (in the long run) than IE.
« Last Edit: January 18, 2007, 04:31:18 pm by EricM »
Logged
-Eric Myrvaagnes (visit my website: http://myrvaagnes.com)

alainbriot

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 796
  • http://www.beautiful-landscape.com
    • http://www.beautiful-landscape.com
DoF article
« Reply #8 on: January 18, 2007, 04:50:20 pm »

Quote
By the time I got to it, all the missing pieces were there, and the figures all show clearly (in Opera -- the best of all possible browsers     )

P.S. The best fix for IE users is to get a better browser (Opera, Firefox, Safari, Netscape, . . . ) all of which are less expensive (in the long run) than IE.
[a href=\"index.php?act=findpost&pid=96457\"][{POST_SNAPBACK}][/a]

On the Mac we don't have a choice but to do that since IE for Mac is no longer supported.
Logged
Alain Briot
Author of Mastering Landscape Photography
http://www.beautiful-landscape.com

MarkGoddard

  • Newbie
  • *
  • Offline Offline
  • Posts: 8
    • http://
DoF article
« Reply #9 on: January 18, 2007, 05:17:26 pm »

The rendering issues might be caused by some server-side tags that seem to be leaking through to the client.

For example: <![if !supportEmptyParas]><![endif]>

These may be confusing some browsers.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
DoF article
« Reply #10 on: January 18, 2007, 06:25:14 pm »

Quote
On the Mac we don't have a choice but to do that since IE for Mac is no longer supported.
[a href=\"index.php?act=findpost&pid=96463\"][{POST_SNAPBACK}][/a]
Yes, so Firefox ("the browser formerly known as Mozilla") is now a good thing to have on any computer, at least for when the main browser (IE, Safari etc.) cannot handle something.
Logged

dtrayers

  • Newbie
  • *
  • Offline Offline
  • Posts: 44
    • http://
DoF article
« Reply #11 on: January 18, 2007, 10:19:29 pm »

What I don't understand is the statement that's it's a half-truth that a bigger sensor is required for a shallow DOF.  Though it may depend on the definition of 'shallow', I think in general it's completely true.

From Prof. Johnson's figures, a P&S has the same DOF at f/3.5 as an APS-C at f/14 or a FF at f/21.  But the reality is that many, many P&S cameras have lenses that start around f/2.8 and end at f/8.

Based on Prof. Johnson's example of a 2ft. object at 5ft. from the lens and his formulas, a FF camera at f/2 would have a DOF of about 1".  To get the same DOF from a P&S, an aperture of f/0.4 is necessary.  Since most P&S lenses start at f/2.8 (many at f/3.5), the best you can do on this example is a DOF of about 7.5" with a P&S.

Sure, if you want a lot of DOF, then a P&S seems to do just as well as a FF.  But if you want a shallow DOF, the *only* way to get more of it is with a larger sensor.  I suppose someone could make a f/0.4 lens for a P&S, but I'm not sure it's technically possible and certainly not affordable or pocketable.  It would have an optical diameter of about 45mm (about the same as my EF 85mm f/1.8) and a focal length of just 18mm!
Logged
- Dave

Sheldon N

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 828
DoF article
« Reply #12 on: January 18, 2007, 11:02:49 pm »

One other thing that I'm not sure the article completely addressed (perhaps I missed it) is that the this theory of equal maximum DOF only holds true within the enlargement capabilities of the format that you are shooting.

For example, if I shoot a small sensor 3 megapixel camera, the largest print I can get and still maintain a stringent COC is roughly a 5x7 print.

If I start to try and make DOF comparisons with a larger print size and a stringent COC, there is actually zero depth of field since even the sharpest plane of focus doesn't sufficiently resolve enough detail to meet the COC.

So, the 3 megapixel camera wouldn't have equal DOF as a 11 megapixel full frame DSLR on an 8x10 print, even if you apply his theory to equalize the apertures.  Similarly, a 11 megapixel DSLR shot wouldn't have equal DOF to 4x5 shot at a 20x30 print with equalized apertures.

Common sense stuff, but thought I would bring it up.
Logged
Sheldon Nalos
[url=http://www.flickr.com

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
DoF article
« Reply #13 on: January 19, 2007, 12:20:25 am »

I think Professor Johnson did mention that the P&S camera is more limited in its range of options. Generally, lenses for smaller formats tend to have bigger maximum apertures. As far as I know, only 35mm format offers a lens as fast as f1. I believe the Canon/F1 has long since been discontinued and now replaced with the 50/f1.2. Such lenses are expensive and the resolution performance at full aperture is not stellar, although the shallow DoF might be.

If it's possible to build an f1 lens for 35mm format, it should be possible to build an f 0.5 lens for a P&S camera, but it would be expensive and who would buy it. I'm not aware of any F1 or F1.2 35mm zooms. It seems if you are going to build a lens with the widest possible aperture, it has to be a prime.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
DoF article
« Reply #14 on: January 19, 2007, 12:40:22 am »

Quote
One other thing that I'm not sure the article completely addressed (perhaps I missed it) is that the this theory of equal maximum DOF only holds true within the enlargement capabilities of the format that you are shooting.

[a href=\"index.php?act=findpost&pid=96521\"][{POST_SNAPBACK}][/a]

I believe when the diffraction spot size exceeds the CoC, then DoF can be the same at any degree of enlargement. The entire image might then become unacceptably fuzzy, as many a shot is from a pin-hole camera.

I think professor Johnson also mentioned that prints ideally should be viewed in accordance with the perspective used when the photo was taken, for the best impression of DoF. Thus, a close-up taken with a wide-angle lens should be viewed from close up. Viewed from a distance, the relatively small background detail (in proportion to the foreground) becomes undiscernible and the DoF effect destroyed.

Perhaps more to the point, enlargement capability is now dependent on pixel count rather than sensor size. Consider and compare Canon's first DSLR, the 3mp D30, and it's latest DSLR, the 10mp 400D. Both sensors are the same size, but it's obvious which has the greater enlargement capability.
« Last Edit: January 19, 2007, 01:49:55 am by Ray »
Logged
Pages: [1]   Go Up