RichA
New member
Nikon sharper, more DOF. Odd, but interesting.
http://www.pbase.com/andersonrm/unusual_dof_of_olympus_50mm_lens
http://www.pbase.com/andersonrm/unusual_dof_of_olympus_50mm_lens
The GetDPI Photography Forum
Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!
Olympus 50mm versus Nikon 50mm DOF at f1.8
- Thread starter RichA
- Start date
DOF is not a lens characteristic per se. It depends on the assumed circle of confusion and viewing conditions (frame magnification , viewing distance, eyesight). Two lenses of the same focal length, focused at the same distance and set to the same aperture will produce on the focus plane discs of confusion of the same size (diameter). All other parameters being equal, also the DOF will be the same.
pellicle
New member
which is perhaps why RichA thought it was interesting ... I find it interesting too and would like to know more (such as was there an accidental setting of the aperture. This is of course not recorded in the EXIF data.
RichA
New member
No, the battery row and the ruler were shot on separate days. The aperture was wide open. No way it could be actuated on the G1 either.
I think the world is more complicated. A DOF calculator is based on a theoretical thin lins and is as such doomed not to work in real life. Well, a DOF calculator works, but it isn't that exact.
For example, I have several times read about Zeiss 50mm lenses giving a more distinct "limit" between the in DOF area and the OOF area. Then we humans either perceive this as a more shallow DOF than the other 50mm lens the Zeiss is compared to, or as a not very smooth transition from in-focus and out towards out of focus.
As there are very many different lens constructions I assume there also are many different results with regards to the DOF, for example the exact depth of it, the distribution of it, how it ends (suddenly or smooth).
just mentioning a few things I got from here and there,
/Jonas
In reality it looks different.
Once I compared different 50mm on my M8 at the SAME f-stop (Noctilux, Summilux, Elmarit, Sonnar) and for some reason I dont know the DOF "appears" different.
I believe in digital world DOF things have change and depend on different more factors, maybe including AA filter for example, contrast of a lens and other things.
It's somehow difficult to believe for me that "in real world" discs of confusion produced by lenses of the same f-length and at the same true aperture and focussing distance can differ*. M8 (which I also have) is not the best measuring equipment as issues of backfocus , rangefinder adjustment, lens adapter misalignment and such can skew any experiment.
One needs to start with a manual focus with visual confirm, such as in any "live view" camera.
You need to make sure that the DOF measuring plate/scale is always at the same distance from the camera's plane of focus.
You must ensure that the target and camera is rock solid and mirror/shutter shake are also eliminated (preferably an isolated optical bench).
You must focus bracket on the target many times to pick the one 100% clear rendition.
You can of course use only one and the same camera for all comparisons. You must set the the lenses to one common optimum, not diffraction limited, aperture (say f4 in m43) to minimize aberrations**.
You must exclude any influence of the jpeg machine/raw developer (develop raws in dcraw).
The lighting, visibility and ambient temperature as well as temperature of the sensor must be kept consistent throughout.
YOU MUST BE SURE THAT F-LENGTH AND F-NUMBERS FROM THE LENS 100% CORRESPOND TO REALITY***
You must magnify the recorded frame to the same size (minimum 10x8 inch) at each take when printing or always use a 100% crop on computer screens.
All people looking at the prints/crops(always 1:1 pixel for pixel) do it from the same distance with the same eyesight strength (best corrected to 20/20 vision).
If under such circumstances the perceived DOF is different, I'd start looking for explanations beyond the disc of confusion/circle of confusion (CoC) model.
* I would need to see some evidence from optical engineering, I am not an expert
** One thing real in the "real world" is aberrations which are always present
***Despite markings say 50mm, the lens can actually have 47mm or 53mm
The aperture marked/communicated as f4 in reality can be different due to manufacturing inaccuracies. Also light transmission losses occur. This in cinematography world is remedied with Tstop figure--the lenses are actually measured all individually for light transmission and marked with "real world" T-stops rather then F-stops (T for transmission). Eg. T2 can be equivalent to F 2.3. Also distance scale is marked from real measurments.
In cinematography nobody heard of DOF variations dependent on lens brand. The 35mm cine DOF tables are made for one standard circle of confusion of 1/1000 inch, and they must be true for all approved pro glass marked with T-stops. The photography world variations, if any, must come from inaccurate methodologies--those of lens manufacturers or the users.
One needs to start with a manual focus with visual confirm, such as in any "live view" camera.
You need to make sure that the DOF measuring plate/scale is always at the same distance from the camera's plane of focus.
You must ensure that the target and camera is rock solid and mirror/shutter shake are also eliminated (preferably an isolated optical bench).
You must focus bracket on the target many times to pick the one 100% clear rendition.
You can of course use only one and the same camera for all comparisons. You must set the the lenses to one common optimum, not diffraction limited, aperture (say f4 in m43) to minimize aberrations**.
You must exclude any influence of the jpeg machine/raw developer (develop raws in dcraw).
The lighting, visibility and ambient temperature as well as temperature of the sensor must be kept consistent throughout.
YOU MUST BE SURE THAT F-LENGTH AND F-NUMBERS FROM THE LENS 100% CORRESPOND TO REALITY***
You must magnify the recorded frame to the same size (minimum 10x8 inch) at each take when printing or always use a 100% crop on computer screens.
All people looking at the prints/crops(always 1:1 pixel for pixel) do it from the same distance with the same eyesight strength (best corrected to 20/20 vision).
If under such circumstances the perceived DOF is different, I'd start looking for explanations beyond the disc of confusion/circle of confusion (CoC) model.
* I would need to see some evidence from optical engineering, I am not an expert
** One thing real in the "real world" is aberrations which are always present
***Despite markings say 50mm, the lens can actually have 47mm or 53mm
The aperture marked/communicated as f4 in reality can be different due to manufacturing inaccuracies. Also light transmission losses occur. This in cinematography world is remedied with Tstop figure--the lenses are actually measured all individually for light transmission and marked with "real world" T-stops rather then F-stops (T for transmission). Eg. T2 can be equivalent to F 2.3. Also distance scale is marked from real measurments.
In cinematography nobody heard of DOF variations dependent on lens brand. The 35mm cine DOF tables are made for one standard circle of confusion of 1/1000 inch, and they must be true for all approved pro glass marked with T-stops. The photography world variations, if any, must come from inaccurate methodologies--those of lens manufacturers or the users.
Last edited:
There will never be anyone managing to show you images you can accept then. Still, we have folks very good at the trade every now and then discussing the different DOF they get from different lenses.
Why would every lens sold act as an ideal thin lens with regards to DOF? Do they all have the same amount of the DOF in front, and behind, of the focal plane? I'm just thinking out loud, not really asking.
regards,
/Jonas
Mathematically calculated "general" DOF tables are accurate to within couple of inches.
The actual DOF (or rather the size of CoC) for a given lens depends also on its entrance pupil size and position.
Entrance pupil is the projection of the physical aperture as seen from the front of the lens. Therefore for ultimate accuracy each lens model should have it's own DOF table. This is done for cine lenses such as eg. Zeiss Master Prime: http://tinyurl.com/28qqhqq
But cine lenses have extremely accurate focusing systems and the 1-2 inch focussing accuracy is possible. In photography such precision can be achieved perhaps with focus bracketing, practical only in time consuming tests.
Zeiss do not deliver individual stills lenses DOF tables, as the accuracy of the manufacturing process (also on the camera mount/sensor side) and user technique make such issues moot.
DOF perception can also differ because of resolving power of the lens. When the objects in the focus plane are very detailed the DOF seems deeper.
Other factors such as soft/harsh lighting can influence the perception.
So indeed to be exact:
DOF can vary with the lens construction and also depends on manufacturing accuracy (also of the camera) and the final user/viewer practice.
I still don't think these factors are critical for the practice of photography.
The actual DOF (or rather the size of CoC) for a given lens depends also on its entrance pupil size and position.
Entrance pupil is the projection of the physical aperture as seen from the front of the lens. Therefore for ultimate accuracy each lens model should have it's own DOF table. This is done for cine lenses such as eg. Zeiss Master Prime: http://tinyurl.com/28qqhqq
But cine lenses have extremely accurate focusing systems and the 1-2 inch focussing accuracy is possible. In photography such precision can be achieved perhaps with focus bracketing, practical only in time consuming tests.
Zeiss do not deliver individual stills lenses DOF tables, as the accuracy of the manufacturing process (also on the camera mount/sensor side) and user technique make such issues moot.
DOF perception can also differ because of resolving power of the lens. When the objects in the focus plane are very detailed the DOF seems deeper.
Other factors such as soft/harsh lighting can influence the perception.
So indeed to be exact:
DOF can vary with the lens construction and also depends on manufacturing accuracy (also of the camera) and the final user/viewer practice.
I still don't think these factors are critical for the practice of photography.
RichA
New member
This is referred to in the amateur astronomy world as focus "snap" where higher quality (accuracy) lenses produce a more pronounced in-focus point than lesser lenses. You know exactly when you hit focus.
I compared lens-specific ZeissMaster DOF table figures with the "general" ones (dofmaster.com).
For the Zeiss Master Prime I linked above, cine Distagon 32mm/T1.3 the Zeiss table at T2 and distance 2 meters gives:
1.88-2.21m
For the same cine CoC of 0.025mm (US 1/1000 inch) and f2/2 meter distance, dofmaster.com:
1.82-2.21m
The boundaries only mean where the CoC achieves the diameter of 0.025mm. DOF does not suddenly disappear at 1.82 m or 1.88m, the CoCs become gradually bigger or smaller and it's a smooth process. The whole thing is purely conventional and the perception of DOF also depends on the viewing conditions (display size, distance, visibility, eyesight).
Definitive answer to all DOF questions by Dr Nasse of Zeiss:
http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/CLN_35_Bokeh_EN/$File/CLN35_Bokeh_en.pdf
For the Zeiss Master Prime I linked above, cine Distagon 32mm/T1.3 the Zeiss table at T2 and distance 2 meters gives:
1.88-2.21m
For the same cine CoC of 0.025mm (US 1/1000 inch) and f2/2 meter distance, dofmaster.com:
1.82-2.21m
The boundaries only mean where the CoC achieves the diameter of 0.025mm. DOF does not suddenly disappear at 1.82 m or 1.88m, the CoCs become gradually bigger or smaller and it's a smooth process. The whole thing is purely conventional and the perception of DOF also depends on the viewing conditions (display size, distance, visibility, eyesight).
Definitive answer to all DOF questions by Dr Nasse of Zeiss:
http://www.zeiss.com/C12567A8003B8B6F/EmbedTitelIntern/CLN_35_Bokeh_EN/$File/CLN35_Bokeh_en.pdf
Last edited:
Maybe, but... When I think of a lens with good focus "snap" it is a lens having high contrast wide open and less SA than the "lesser" lens you mention. I don't think of a lens with more shallow DOF than another lens of the exact same speed and focal length.
I suppose Wer est tombé misst, misst est tombé le brouillard, no?
OK. Then we agree. I never thought we discussed anything having a lot to do with practical photography...
Cheers
RichA
New member
Your criteria are correct. Snap is produced by the lens being stripped of aberrations, taken to a specific quality level (waves accurate) so it will be apparent when it is in focus. There are about a dozen aberrations and physical characteristics that effect ultimate image sharpness. However, with two lenses of equal correction and different focal ratios, the faster lens will appear to meet focus more distinctly because its range of best focus is (like its DOF) narrower than the higher ratio lens.