jaapv
RFF Sponsoring Member.
Lens diffraction has nothing to do with digital, it is a lens property that is equally valid for film. Small apertures " bend" the light, lowering lens performance. With the best lenses it can be visible from f 4.0, usually it kicks in at 5.6 or 8.0.
Sparrow
Veteran
About the most practical advice that I've seen in this thread is to use the DOF field on the lens. It was set by the manufacture, using their set of assumptions about how happy you will be with the image, and how big you plan to enlarge it. Some manufacturers vary assumptions, some fudge them to make their lenses look better.
With an M8- start out using the DOF as marked. "Pixel Peep" some on the LCD, look at it on the computer, make some prints. Are you happy with the focus as indicated? If you are not, you will come up with your own subjective criteria. Absolute size of the circles of confusion is a function of focal length, aperture, actual point of focus, and distance of the point source from the actual point of focus. I've got an Excel spreadsheet that I made sitting on some Win98 machine in the house, lets you input focal length, aperture, focal point, and chosen COC size. It computes the endpoints, front and back of the main focal point, for that size COC.
Like I said, it's on some Win98 machine, have not looked at it in 10 years. I know what I like, and basically ignore the DOF as marked on the lens. Except on the Zoom-Nikkor 43~86 F3.5 and Zoom-Nikkor 80~200 F4.5. That's cause they're so cool.
With an M8- start out using the DOF as marked. "Pixel Peep" some on the LCD, look at it on the computer, make some prints. Are you happy with the focus as indicated? If you are not, you will come up with your own subjective criteria. Absolute size of the circles of confusion is a function of focal length, aperture, actual point of focus, and distance of the point source from the actual point of focus. I've got an Excel spreadsheet that I made sitting on some Win98 machine in the house, lets you input focal length, aperture, focal point, and chosen COC size. It computes the endpoints, front and back of the main focal point, for that size COC.
Like I said, it's on some Win98 machine, have not looked at it in 10 years. I know what I like, and basically ignore the DOF as marked on the lens. Except on the Zoom-Nikkor 43~86 F3.5 and Zoom-Nikkor 80~200 F4.5. That's cause they're so cool.
Finder
Veteran
I blasted you for a dumb article at Luminous landscape that concludes that DoF is not related to focal length. It clearly is.
I also blasted you for your assumption about DoF scales. The compensation we are talking about here is simply the change in sensor size. Not this idea that manufactures don't know how to make DoF scales--your opinion.
Changing your name to roland will not help you on either of those two points.
Finder
Veteran
You are forgetting the examples I posted that clearly show the change in DoF.
jaapv
RFF Sponsoring Member.
Sorry, DOF is NOT related to focal length. It is related to enlargement, and as a longer lens magnifies more, it appears to give less DOF. However, if you crop a shorter lens down to the same FOV when taking an image from the same distance, the DOF will be identical.
Finder
Veteran
If you can separate magnification from focal length, I will believe you.
Actually, DoF in your example will be ALMOST the same. DoF is not a linear property in that regard--you are still dealing with two different entrance pupils given the same f-number.
Finder
Veteran
Right now, I would not be too concerned. It appears that your lens will work anyway you would want it to.
ampguy
Veteran
Thanks Brian
Thanks Brian
I agree with this. 8x10's can be printed for $1.49 and 4x6's for $0.10 to 0.15 cents from either your M8 digital files, or your scanned film.
It confounds me that folks will spend countless hours on the internet, but not take an hour testing, and taking a trip to Costco or Walgreens to pickup the prints, or bother to have them mailed back to them to see these results for themselves.
I further cannot understand how folks think they can "emulate" an M8 and lens setup with photoshop. This is wrong. Unnatural, and probably illegal in Georgia
Btw, my results are over here:
http://www.rangefinderforum.com/forums/showthread.php?t=91281
Thanks Brian
I agree with this. 8x10's can be printed for $1.49 and 4x6's for $0.10 to 0.15 cents from either your M8 digital files, or your scanned film.
It confounds me that folks will spend countless hours on the internet, but not take an hour testing, and taking a trip to Costco or Walgreens to pickup the prints, or bother to have them mailed back to them to see these results for themselves.
I further cannot understand how folks think they can "emulate" an M8 and lens setup with photoshop. This is wrong. Unnatural, and probably illegal in Georgia
Btw, my results are over here:
http://www.rangefinderforum.com/forums/showthread.php?t=91281
jaapv
RFF Sponsoring Member.
Ok, you are right in that respect, and the formulas for DOF do introduce focal length in the third level. But it is of minimal significance.
Finder
Veteran
I agree with this. 8x10's can be printed for $1.49 and 4x6's for $0.10 to 0.15 cents from either your M8 digital files, or your scanned film.
It confounds me that folks will spend countless hours on the internet, but not take an hour testing, and taking a trip to Costco or Walgreens to pickup the prints, or bother to have them mailed back to them to see these results for themselves.
I further cannot understand how folks think they can "emulate" an M8 and lens setup with photoshop. This is wrong. Unnatural, and probably illegal in Georgia
Btw, my results are over here:
http://www.rangefinderforum.com/forums/showthread.php?t=91281
But I have done the tests and you are wrong. So wrong in fact that your first two images comparing the ff and M8 at f/8 show a change in DoF.
ampguy
Veteran
Not to my eyes
Not to my eyes
Not in the area that I am concerned with, which is the M8's area. The M6 at the same distance gives a little overscan, just like TV's, but my area that I am framing for, and looking for DOF differences is in my M8 photo. Just as I presume the OP had asked.
I think you are off on this multi-dimensional magnification rube goldberg concept where everything revolves around FF.
Well, thankfully we have a lot of M lenses in various lengths that allow not only M8s, but RD1s, and even 4/3rds (possibly diffraction limited to f4) to make great photos with these lenses, including the DOF barrel markings, as is.
Not to my eyes
Not in the area that I am concerned with, which is the M8's area. The M6 at the same distance gives a little overscan, just like TV's, but my area that I am framing for, and looking for DOF differences is in my M8 photo. Just as I presume the OP had asked.
I think you are off on this multi-dimensional magnification rube goldberg concept where everything revolves around FF.
Well, thankfully we have a lot of M lenses in various lengths that allow not only M8s, but RD1s, and even 4/3rds (possibly diffraction limited to f4) to make great photos with these lenses, including the DOF barrel markings, as is.
Finder
Veteran
Nope. I am clearly right. Nice try though.
This is off to Rant-Land.
Next step will be to close it.
If someone wants to put up the formula for COC, feel free to.
Next step will be to close it.
If someone wants to put up the formula for COC, feel free to.
ampguy
Veteran
Hi Brian
Hi Brian
Well that's a bit of a trick question, since it involves a constant that varies among manufacturers, but here's what I have:
CoC = D (format diagonal = 43.3mm for 35mm FF) / z-constant
z-constant varies by vendor, but Zeiss has at times used 1500, and 1730, and Canon's online calculator uses 1443.
By re-arranging the formula to:
z-constant = D / CoC, we can infer that Leica uses a z-constant of 1883, since it has published that it uses a CoC value of .023mm for the design of it's lenses.
So one thing to check the next time you fire up your favorite dof calculator, is what that calculator is using for it's z-constant for calculations.
Hi Brian
Well that's a bit of a trick question, since it involves a constant that varies among manufacturers, but here's what I have:
CoC = D (format diagonal = 43.3mm for 35mm FF) / z-constant
z-constant varies by vendor, but Zeiss has at times used 1500, and 1730, and Canon's online calculator uses 1443.
By re-arranging the formula to:
z-constant = D / CoC, we can infer that Leica uses a z-constant of 1883, since it has published that it uses a CoC value of .023mm for the design of it's lenses.
So one thing to check the next time you fire up your favorite dof calculator, is what that calculator is using for it's z-constant for calculations.
Just put up the formula for calculating the size of the formed circle of confusion for focal length and aperture. What the manufacturers use for their cutoff is subjective, and may or may not be acceptable to an individual. That's why I've ignored the manufacturers marked DOF scale for years. I prefer to get to know what the lens does, not what the manufacturer assumes that I will find accecptable.
If no one else can find it, I'll look it up in one of my reference books and put it on my forum. Watching Monsters vs Aliens on DVD was kind of like reading this thread, but much more entertaining.
If no one else can find it, I'll look it up in one of my reference books and put it on my forum. Watching Monsters vs Aliens on DVD was kind of like reading this thread, but much more entertaining.
tlitody
Well-known
You can get some cool software from Linos. WinLens 4.3 and Pre Designer and play with all the parameters to your hearts content.
http://www.linos.com/pages/no_cache.../winlens/?sid=13476&cHash=6942b9ffc5#sid13476
It was free so all you need to do is go through the ordering process and there is nothing to pay (that is how it used to be).
Pre designer is the one you want but you need to install both.
http://www.linos.com/pages/no_cache.../winlens/?sid=13476&cHash=6942b9ffc5#sid13476
It was free so all you need to do is go through the ordering process and there is nothing to pay (that is how it used to be).
Pre designer is the one you want but you need to install both.
Last edited:
Jeff S
Well-known
Just as the Luminous Landscape article concludes...the author admits flaws, and cites real life prints, not exact science.
Same with DOF scales. You use them exactly. Congratulations, I don't. And, my resultant prints are all that matters.
Jeff
"Lenses in Photography", Kingslake, 1951.
"a depth of field table or scale should be taken with many mental reservations, and depths stated to small fractions of an inch can be somewhat misleading. Moreover, the observer's eye is not always situated at the correct center of perspective, and hence the depth may be multiplied or divided by a factor depending on his departure from the proper viewing conditions, which may reach as much as 2 or 3 or more. This hair-splitting arguments about depth of field data becomes absolutely worthless".
So- DOF is subjective. It's opinion. The absolute size of a circle of confusion is ruled by physics. The interpretation of what is and is not acceptable for an enlargement is subjective. The opinion of the person viewing the final results is all that matters with respect to DOF, and the absolute placement of the little lines on the lens are nothing more than "Feel Good" indicators on the part of the maker of the lens.
"a depth of field table or scale should be taken with many mental reservations, and depths stated to small fractions of an inch can be somewhat misleading. Moreover, the observer's eye is not always situated at the correct center of perspective, and hence the depth may be multiplied or divided by a factor depending on his departure from the proper viewing conditions, which may reach as much as 2 or 3 or more. This hair-splitting arguments about depth of field data becomes absolutely worthless".
So- DOF is subjective. It's opinion. The absolute size of a circle of confusion is ruled by physics. The interpretation of what is and is not acceptable for an enlargement is subjective. The opinion of the person viewing the final results is all that matters with respect to DOF, and the absolute placement of the little lines on the lens are nothing more than "Feel Good" indicators on the part of the maker of the lens.
Last edited:
Just wait for the next EP2 firmware update. You will create a little virtual reality gallery, with an enlarged print of the image that you are about to take, a person viewing that image, and placement of the person all via camera controls. Then, the live view will simulate you looking at the final print as you are composing the photograph. Change the F-Stop, realtime simulation through the viewfinder bases on the perspective of your little virtual reality world.
It's just software, why haven't they implemented it in release 1.1?
It's just software, why haven't they implemented it in release 1.1?