Depth of field on M8 vs MP... Sorry for being thick!

Murphy's Law

Murphy's Law

Sparrow said:
All these DOF threads should be merged into one mammoth thread if confusion ..... a 50mm lens at f2 has the same DOF whatever it projects the image onto, then use reason from there
Click to expand...

It's important to remember a corollary to Murphy's Law: "All constants are variables."

OK, let me try to use some reason and logic on this. Assume that I display only un-cropped images and at 20 x 30 units. I take two shots from the same vantage point with the same lens and f stop, but photo 1 is a full frame M9 and photo 2 is a 1.33 crop factor M8. I end up with two different photos, with #1 having more info and a wider field of view. Blow them both up, un-cropped, to my specified size, and I will have had to blow up #2 more by a factor of 1.33. In doing so, I have also magnified the threshold circle of confusion by the same 1.33 factor.

So I assume that I will routinely have to blow up the M8 images more than what I'm used to for whatever I'm doing, regardless of other variables. That's why I rely more on the next wider stop markings on the lens for hyperfocal determinations.
 
Mike Richards said:
It's important to remember a corollary to Murphy's Law: "All constants are variables."

OK, let me try to use some reason and logic on this. Assume that I display only un-cropped images and at 20 x 30 units. I take two shots from the same vantage point with the same lens and f stop, but photo 1 is a full frame M9 and photo 2 is a 1.33 crop factor M8. I end up with two different photos, with #1 having more info and a wider field of view. Blow them both up, un-cropped, to my specified size, and I will have had to blow up #2 more by a factor of 1.33. In doing so, I have also magnified the threshold circle of confusion by the same 1.33 factor.

So I assume that I will routinely have to blow up the M8 images more than what I'm used to for whatever I'm doing, regardless of other variables. That's why I rely more on the next wider stop markings on the lens for hyperfocal determinations.
Click to expand...

Not quite.

Digital sensors don't record circles of confusion. They record a single colour in a pixel which has no size. It is built from countless CoCs overlapping each other onto a single pixel sensor. Any attempt to try and compare to magnifying image capture from film based on CoC is invalid.

Printing an M8 image at a size of 6x4 inches provides 25.82 pixels per mm which is 13 line pairs per mm. i.e. there is no magnification required to print an M8 image at same size as from a 35mm neg when CoC required for the lens on film is 0.03mm. But there is 4.16 times magnification required from film. So where does that leave your thinking?
 
Last edited:
Mike Richards said:
It's important to remember a corollary to Murphy's Law: "All constants are variables."

OK, let me try to use some reason and logic on this. Assume that I display only un-cropped images and at 20 x 30 units. I take two shots from the same vantage point with the same lens and f stop, but photo 1 is a full frame M9 and photo 2 is a 1.33 crop factor M8. I end up with two different photos, with #1 having more info and a wider field of view. Blow them both up, un-cropped, to my specified size, and I will have had to blow up #2 more by a factor of 1.33. In doing so, I have also magnified the threshold circle of confusion by the same 1.33 factor.

So I assume that I will routinely have to blow up the M8 images more than what I'm used to for whatever I'm doing, regardless of other variables. That's why I rely more on the next wider stop markings on the lens for hyperfocal determinations.
Click to expand...

Very good, you tested the hypothesis that; a 50mm lens at f2 has the same Field of View whatever it projects the image onto, and found it to be untrue
 
tlitody said:
Not quite.

Digital sensors don't record circles of confusion. They record a single colour in a pixel which has no size. It is built from countless CoCs overlapping each other onto a single pixel sensor. Any attempt to try and compare to magnifying image capture from film based on CoC is invalid.

Printing an M8 image at a size of 6x4 inches provides 26.24 pixels per mm which is 13 line pairs per mm. i.e. there is no magnification required to print an M8 image at same size as from a 35mm neg when CoC required for the lens on film is 0.003mm. But there is 4.16 times magnification required from film. So where does that leave your thinking?
Click to expand...

You're trying to make variables out of the constants I posited. I compared full frame M9 to M8 digital to digital. If the given lens on an M9 has hyperfocal markings that indicate certain DOF characteristics, no matter what it is based on -- CoC, pixel overlap or whatever, or derived on film or anything else -- then the M8 image, under the conditions I stated, will have to be blown up relatively by a factor of 1.33. Thus it's prudent to use next stop markings if one wants M9-like DOF characteristics as a basis. It's been pointed out that Leica claims the standard hyperfocal markings are valid for the M8. If one agrees with that, conceivably one could use the next wider hyperfocal marks for M8-like DOF characteristics on the M9.
 
ampguy said:
You may not realize it, but the smaller sensor is more diffraction limited than the larger sensor, (given same pixel pitches, sensor pattern design) so let's get on with my 2nd favorite topic today, what were the apertures and cameras used?
Click to expand...

Nice try although has nothing to do with the topic. The aperture was around f/2.8. So diffraction is not an issue. Not that diffraction is an issue in DoF.
 
Jeff S said:
Hey, you're the one who said focal length affects DOF. And, you're wrong...http://www.luminous-landscape.com/tutorials/dof2.shtml

And, obviously, DOF scales are only a guide. Duh. But, they are worse of a guide now than they were when established long ago. If you don't understand that, there's no point arguing with you.

Jeff
Click to expand...

Jeff, DoF s related to magnification. Focal length is is related to magnification. Here is the equation for calculating hyperfocal distance:

H = f^2/NxC wher f is focal length.

If the depth of field scales are worse now that they have ever been, then show it. My scales of my Horseman SW612, which is a scale focus camera, are spot on. The scales are accurate on both my Mamiya 6s. But show your proof...
 
tlitody said:
I haven't read all the answers here so my apologies if I repeat what someone has already said.

There are two answers. The first is purely scientific and the second is real world.

First is that the lens has not changed. It is the sensor that has changed. DOF is not affected by a sensor. It a property of the lens and not of a sensor. Therefore changing sensor has no effect on DOF. This is true and unarguable.

The second and real world answer depends on how you phrase your question.
If you mean "does the DOF change for the same subject area?" then yes it does. If you mean "does the DOF change for the cropped area caused by the reduced sensor size?" then no it doesn't.
The reason is simple to answer. The basics of DOF are that the nearer you are to the subject, then the shorter the DOF and the further you are from the subject then the greater the DOF.
Well with a smaller sensor and using the "same subject area", then you need to be further from the subject to get it all in. That means you get greater DOF. The DOF scale on the lens is still correct and you don't need to compensate because of smaller sensor. The compensation is automatic as a result of changing distance from the subject which will give you a new reading from your DOF scale.
If you stay the same distance from subject then nothing has changed and DOF will remain the same but you will get a smaller subject area on the sensor.

There is a third answer which is dependant on how sharply a sensor is capable of recording the subject but that is not factored into lens DOF scales as far as I know.

Leicas statement that there is no compensation required is true as far as I'm comcerned.

Finally, what this means is that with your M8 you will tend to be able to use a slightly wider aperture and benefit from increased shutter speed compared to your MP if you are working the same way and looking for the same DOF as you normally do on your MP. You will tend to be a little further from your subjects. I am of course assuming you are using the same lens.
Click to expand...

Exactly where is the "scientific" part of your answer? There is no discussion of circles of confusion. With out that, there is discussion of DoF.
 
tlitody said:
No you don't win because you have completely failed to take into account that different sensors produce different sharpness due to AA filters. So an M9 is claimed to be super sharp. An M8 isn't. That super sharpness will affect apparent DOF. But being scientifically challenged you have made a false assumption that it is a DOF issue. It's not. We live and learn. Well some of us do.
Click to expand...

The sensors are identical. But the AA filters do not affect DoF. At best the AA filter would affect depth of focus. You are confusing object space with image space.

You keep talking about science, but you clearly don't understand the issue.
 
tlitody said:
That site is highly misleading. It changes the CoC to suit the sensor size which is what you lot seem to think is correct and the reason that DOF is reduced.
But what you have not figured out is that the CoC it uses is the one that a lens would need to be able to produce to give the same resolution and DOF as a lens designed for Full Frame. Unfortunately just throwing a new CoC into a dumb online calculator does not change the design specification of the actual lens being used and which was designed to produce a CoC to suit a FF format.
To put it simply for you, you have interpreted the dumb online calculator incorrectly. i.e. The blind leading the blind.
Click to expand...

But if you actually understood depth of field, you would understand the coc at the sensor is proportional to sensor size. So the calculator is correct.
 
Sparrow said:
All these DOF threads should be merged into one mammoth thread if confusion ..... a 50mm lens at f2 has the same DOF whatever it projects the image onto, then use reason from there
Click to expand...

Then explain why Horseman uses a different DoF scale on its 55mm lens compared to Nikon's scale on its 55mm lens. The Horseman is a 6x12 camera and the Nikon a 35mm camera.
 
Spyro said:
Ι never understood why these threads always get so long.
What is so important about dof...
Click to expand...

In this case, the OP wants to know if he can trust his DoF scales on his 35mm lens if it is mated with a smaller sensor.

However, folks seem to think DoF is some kind of absolute quality, which, of course, it is not.
 
Lets have some more examples, but with some quantitative data. Here is two images of a micrometer placed at an angle taken with identical sensors, lens, aperture, and distance. The difference is sensor size.

http://www.rangefinderforum.com/photopost/uploads/3853/dof_test_pair.jpg

Here is a plot of the large indices of the left image:

http://www.rangefinderforum.com/photopost/uploads/3853/plot_1_left.jpg

Here is a plot of the large indices of the right image:

http://www.rangefinderforum.com/photopost/uploads/3853/plot_2_right.jpg

These clearly show the loss of depth of field. I am glad I get to deal with an audience that understands science.
 
Finder said:
So you agree that an M8 has different DoF than a 35mm camera. It is a matter of magnification, though, not FOV.
Click to expand...

Isn't that obvious? The projected image of a particular lens and aperture remains the same at the focal plane, after that it depends on the variables ... sensor size being just one
 
Last edited:
Look exactly the same

Look exactly the same

Plus that photo of the spikes is also totally in focus :D


Finder said:
Lets have some more examples, but with some quantitative data. Here is two images of a micrometer placed at an angle taken with identical sensors, lens, aperture, and distance. The difference is sensor size.

http://www.rangefinderforum.com/photopost/uploads/3853/dof_test_pair.jpg

Here is a plot of the large indices of the left image:

http://www.rangefinderforum.com/photopost/uploads/3853/plot_1_left.jpg

Here is a plot of the large indices of the right image:

http://www.rangefinderforum.com/photopost/uploads/3853/plot_2_right.jpg

These clearly show the loss of depth of field. I am glad I get to deal with an audience that understands science.
Click to expand...
 
You must log in or register to reply here.
Back
Top Bottom