DOF resolution and perspective

[SJP]

Advertisement (gone after registration)

Thank you, but what I need to understand is what are right figures must I use in your xls file for the following example?

 

M8 + Summicron 75/2 at f/4 for a distance of 10 m

 

f = ? mm [75 or 75/0.75 or 75*0.75]

f/# = 4 f/value

COC = 20 µm

 

So+Si = 10 m

 

Best regards

Michel

 

You should use

f = 75 mm

f/# = 4 f/value

COC = 20 µm

 

So+Si = 10 m

 

which gives:

8.742 - 11.685 meters or DOF = 2.94

 

It might be fun to check this in a real experiment to see how it compares to harsh reality (physics is an emprical science after all).

[MichelB]

 

It might be fun to check this in a real experiment to see how it compares to harsh reality (physics is an emprical science after all).

 

I'll try some day with other values and give you results.

Now all perfectly clear. In short, for a lens with focal F mounted on a 24x36 box or digital box with a % cropped sensor, calculations of DOF are same except COC.

 

BTW Leica DOF table gives for Summicron 75/2 at F/4 and 5 m: 4.488 - 5.645

To have same figures in your xls file, i.e. 4.485 - 5.648, COC must be 33 µm, just for information.

Before our correspondance here I thought that I must take F/% crop for calculations, 75 mm / 0.75 = 100 mm for example.

Thank you very much.

Best regards.

Michel

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[SJP]

In response to some questions:

 

Michel - thanks for the remark concerning the 'right' COC value, this indeed seems to be lens specific if you use the Leica datasheets. I found 31 micron worked well enough so far but maybe the 75/2 is calculated slightly different. Or does Leica measure the DOF? If so does anyone know how this is done?

 

The pfd file http://spicken59.googlepages.com/DOF...erspective.pdf has not been modified but I have added a IR focus calculator (guestimator) to the excel file:

http://spicken59.googlepages.com/resolutionanddof.xls

 

If someone has experience with IR photography with Leica glass please let me know if the calculation is close to the mark or not.

 

Enjoy!

[MichelB]

In response to some questions:

 

If someone has experience with IR photography with Leica glass please let me know if the calculation is close to the mark or not.

 

Enjoy!

 

No, I do not have experience with IR photography.

 

Please try in your Excel file the following values:

f=35 mm f/#=5.6 COC=31 µm and So+Si=10 m

You'll find:

So-max 1 = 13449

So-max 2 = 10000.000

So-max 3 = 10000.000

and in Approximate equation -23.817 ! Where is the error?

Good night dear Professor. I want to see the result tomorow !!!

Best regards.

Michel

[SJP]

"and in Approximate equation -23.817 ! Where is the error? Good night dear Professor. I want to see the result tomorow !!!"

 

Michel, Your wish is my command. Thanks for pointing this out I had overlooked that possibility. Inspecting the approximate equation in the pdf file it is clear that negative values can occur for 'beyond infinity' solutions - also known as a non-physical solution. This has been dealt with in the spreadsheet.

 

Also I have added a comment on the fact that computers do not like 'infinity' so I have truncated So-max to 100.000 meters, which is close enough for the present purposes.

 

Can no-one comment on the IR focussing?

[MichelB]

"

 

Also I have added a comment on the fact that computers do not like 'infinity' so I have truncated So-max to 100.000 meters, which is close enough for the present purposes.

 

 

Thank you.

Best regards.

Michel

[SJP]

Advertisement (gone after registration)

from LFI FAQ, see here(while it lasts...)

 

"Older lenses feature a red marking opposite the range scale, indicating to which extent the focal plane shifts in infrared photography. If the infrared percentage of light is not filtered out, then a blurred, slightly larger IR picture will overlay the sharp picture formed by the visible light."

 

So the plot thickens, do we need to refocus IR or not? The above would suggest that this is not the case and then 'fefe's question becomes even more interesting (and the IR part of my excel file can be sent to the bin).

 

Maybe there is a difference between old and new lens design with this? I would be surprised maybe this LFI blurb is just PR vaporware.

 

One red spot is enough - you do not need to put it on the lens barrel anymore.

[fefe]

"and in Approximate equation -23.817 ! Where is the error? Good night dear Professor. I want to see the result tomorow !!!"

 

Michel, Your wish is my command. Thanks for pointing this out I had overlooked that possibility. Inspecting the approximate equation in the pdf file it is clear that negative values can occur for 'beyond infinity' solutions - also known as a non-physical solution. This has been dealt with in the spreadsheet.

 

Also I have added a comment on the fact that computers do not like 'infinity' so I have truncated So-max to 100.000 meters, which is close enough for the present purposes.

 

Can no-one comment on the IR focussing?

 

Hi,

I did a few trys to figure if your IR focusing spreadsheet worked. I did not measure accurately the distances but here is what I did:

Took my 35 cron asph with a B+W 092 IR filter, focused it on various objects at F2 (checking by taking a picture without the filter that the RF patch was actually right), then check on the screen where was the actual plane of focus, then measure that distance using the RF again (tiles on the floor are good for that).

Focusing at ~80cm gave a sharp image at ~90cm, focusing at 3m gave a sharp picture at 4.5m which is what your spreadsheet predicted (my measurements are inacurate enough for now that the error is on my side). I don't have a ruler here so it won't be more accurate.

 

So, I am going to use your spreadsheet to estimate the shift of focus in IR , maybe it is not perfectly accurate, but I don't take that many pictures at f2 in IR so DoF will compensate for the error. Thanks

[SJP]

Excellent news fefe thanks!!!

 

I will keep the universal IR focus guestimator in the spreadsheet.

[fefe]

I think it works well:

 

35 cron Asph, f2 with B+W 092 at 1m focused around 1.10m (cropped)

[MichelB]

1. "Older lenses feature a red marking opposite the range scale, indicating to which extent the focal plane shifts in infrared photography..."

 

2. (and the IR part of my excel file can be sent to the bin).

 

 

1. I have a Planar 2/50 ZM, is it an old lens? No, but it features a red marking opposite the range scale for IR focusing.

 

2. Don't send your Excel file to the bin.

 

I saw in a french Leica forum that in the instructions book of the APO-MACRO-ELMARIT-R 1:2.8/100 mm it is said that thanks to the apochromatic correction it's not necessary to do the usual focus correction when taking IR pictures. It seems that is the same for all APO lenses.

 

Michel

[SJP]

Fefe great picture thanks for posting this!

 

MichelB interesting this APO information. I guess this is true - if you have apochromatic correction the deviations will be smaller when switching to IR. But they will not be zero as even in the ideal case the correction is done in the visible range. See e.g. wikipedia

[SJP]

For those that like to think about optics from a physics point view. If not feel free to skip all of this & go back to taking 101 pictures.

 

The pdf file has been upgraded.

 

Direct link to embedded excel file which includes a IR focussing calculator (see previous posts).

.

The main change in the document concerns the appendix page 13 & 14 which makes some comments on the effective focal length and the effective aparture when using a cropped sensor (in comparison to FF or 35 mm film).

 

Executive summary (specific for M8):

1 multiply focal length of the lens by 1.33 to get the equivalent focal length

2 multiply aperture setting of the lens by 1.15-1.20 to get the equivalent aperture

 

The second rule of thumb I believe is more accurate than the conversion proposed in the article 'Form follows Format' in LFI by Peter Karbe, Head of the Leica Optics Department, LFI 3/2006, pp 40-47.

 

Comments are welcome.

[jaapv]

I guess this is true - if you have apochromatic correction the deviations will be smaller when switching to IR. But they will not be zero as even in the ideal case the correction is done in the visible range. See e.g. wikipedia

 

The Apo Macro Elmarit is supposed to be corrected apochromatically far into the IR spectrum. As is supposed of the 280/4.0 apo. It would take a bit of testing to see whether this legend is true.

 

This is truly a valuable labor of love The only thing one could add would be a remark about the difference between film and a sensor, A sensor being a two-dimensional receptor and a film a three dimensional semi-opaque surface. This makes a difference in the perception of DOF.

[lct]

...Comments are welcome.

If you insist OK

The very concept of... how do you say?... 'equivalent aperture' is a pure heresy to me.

Apertures do not change with any sensor (or film) cropping in any way.

What such cropping does is changing the size of the circle of confusion, hence the DoF, but an f/1.4 lens remains an f/1.4 lens whatever body we use with it.

No wonder, with such 'equivalences', why people come to think that fast apertures are not important any more in digital photography.

The contrary is true. The more you crop a sensor, the wider the DoF is and the faster lenses you need to get blurred foregrounds or backgrounds actually.

IMHO of course!

[SJP]

lct, In fact that is exactly what the equivalent aperture is telling us. If you have a cropped sensor then a f/1 lens will act as if it was f/1.2 as far as the Depth of Field is concerned. There is no magic substitute for wide aperture optics in the digital age and with a cropped sensor you lose definition in the depth of field.

 

So in fact I think we actually agree, I do not think that 'think that fast apertures are not important any more in digital photography' in fact the opposite.

[lct]

Then there are no 'equivalent apertures' because apertures don't change in any way. There are simply different DoFs we do agree.

BTW i don't know who invented that concept of 'equivalent apertures' -- another Luts or LFI thing perhaps? -- but it is confusing at best.

[luigi bertolotti]

Well, i find the concept of "equivalent aperture", though a little nonsense, can be rather useful when numbers are "round"... if would happen that, using for instance my Summicron 50, I could remind that, say, at f4 its "equivalent aperture" on M8 is f 2,8, I'd enjoy an easy way of DOF evaluation... unfortunately, I think that in most cases one doesn't fall in such easy "equivalents"... and it becomes simply a confusing concept, mainly fot photogs less smart than US ... I remember a friend of mine, years and years ago, owner of a nice Pentax SLR with severals lenses, that when looked at me using a SuperIkonta 6x9 asked with surprise : "So with this old camera you must always use a tele ? I see the lens is a 105 mm... " Probably I replied it was about a "50mm equivalent"....

[SJP]

because apertures don't change in any way

 

Lct, thanks for the comment.

 

Not quite true. If a sensor has a crop factor compared to FF then to get the same perspective you need to compensate the focal length, so a 28 mm lens on the M8 'looks like' a 35 mm lens on film These are approximate numbers but good enough.

 

Similarly by using a cropped sensor you need to compensate the aperture of the lens for the changed resolution of the sensor, so setting f/2 with a 28mm lens on M8 will 'look like' a 35 mm lens on film with an aperture set at about f/2.4 to f/2.5 to produce the same depth of field.

 

So 28mm f/2 on the M8 'is the same' or 'looks the same' as 35mm f/2.4 on film regarding perpective and Depth of Field. This apparent equality is the definition of of the equivalent focal length and the equivalent aperture. If you don't like this as a concept and prefer to think in terms of depth of field instead that is fine with me.

 

'The aperture stays the same and so does the focal length' is an other point of view which is equally defendable as I discus in the document. From that point of view all you are doing is chopping the edge of the photo off (via the cropped sensor) and because the circle of confusion is smaller with the M8 the depth of field is less compared to film under the same circumstances (i.e. it looks like you are using a wider aperture than with film - yep, this is exactly opposite to the results above).

 

So .... it is not easy and it is not unambiguous and indeed it is confusing. Read the blurb and all will be clear

[Advertisement]

Hallo,

Look here: Leica M

[lct]

...I'd enjoy an easy way of DOF evaluation...

What about this old one of mine: use the DoF markings of the nearest faster f stop of your lenses i.e. f/5.6 when you choose f/8 for metering by example. Works fine with my APS-C digicams at least.