gavinlg
Veteran
The sample pictures are at f8.... there's next to no out of focus parts available for scrutiny.
spreadsheet showing why M8 and FF have same DOF
- Thread starter ampguy
- Start date
- Latest activity Latest activity:
- Replies 120
- Views 12K
- Status
The sample pictures are at f8.... there's next to no out of focus parts available for scrutiny.
antiquark
Derek Ross
Here's an old post by Pickett Wilson that bears repeating:
Panasonic LX-3, 24mm (35mm equiv), f/2.8
EOS 5D, 24mm, f/2.8
gavinlg
Veteran
end of thread.
FrankS
Registered User
That's pretty conclusive.
And completely expected. A smaller sensor or film format requires a shorter focal length lens for an equivalent view. And guess what you get with a shorter focal length: greater DOF (when the aperture and focused distance is the same.)
And completely expected. A smaller sensor or film format requires a shorter focal length lens for an equivalent view. And guess what you get with a shorter focal length: greater DOF (when the aperture and focused distance is the same.)
Last edited:
FrankS
Registered User
Off the top, use a wide lens aperture which gives a more limited DOF, in order to compare differences, not f8.
ampguy
Veteran
hmmm
hmmm
according to DOF Master online, the LX3 should give about 6 cm total DOF, while the 5D should give about 30 cm total DOF, assuming focus is at 1m, and would be even less at a shorter focal point.
Sure looks like the LX3 has a little more DOF, doesn't it?
hmmm
according to DOF Master online, the LX3 should give about 6 cm total DOF, while the 5D should give about 30 cm total DOF, assuming focus is at 1m, and would be even less at a shorter focal point.
Sure looks like the LX3 has a little more DOF, doesn't it?
antiquark
Derek Ross
My two cents:
- Normalize the angle of view. So for the M6, you'd use a 50mm lens, and for the M8, you'd use a 35mm lens. That would give a similar AOV. Or, 35mm on the M6 and 28mm on the M8.
- Use the widest aperture possible, to enhance the depth of field effect. If both lenses are capable of F/2, then use F/2.
- Photograph an object at a near distance to enhance the depth of field. Say, the distance should be 1m or so.
- The object should be something that's marked so that variations are easy to see. For example, a yardstick tilted at a 45 degree angle. If you focus on a specific point, the blur should increase gradually before and after the point of focus.
Actually, you might not see much of a difference. According to DOFmaster, the depth of field is pretty close between the M6 and M8.
ampguy
Veteran
tried this scenario
tried this scenario
I could try this, it gets very hard even with my longest throw lenses to have any room to move focus to one stop more wide open.
If you have suggestions for specific lenses or distances to try, I'll give it a shot.
What Finder/Roland and others propose is even more radical than the Zeiss PDF which indicates 1 stop between APS-C and FF (the table in the PDF gives no data point for a ~32mm diag. sensor, which is a lot closer to their 40mm diag (which they use for FF, not 43.2?), than to their APS-C diag size (26mm).
Perhaps Finder and Roland may want to revise their estimate to an appropriate fraction?
I'm all ears here ...
tried this scenario
I could try this, it gets very hard even with my longest throw lenses to have any room to move focus to one stop more wide open.
If you have suggestions for specific lenses or distances to try, I'll give it a shot.
What Finder/Roland and others propose is even more radical than the Zeiss PDF which indicates 1 stop between APS-C and FF (the table in the PDF gives no data point for a ~32mm diag. sensor, which is a lot closer to their 40mm diag (which they use for FF, not 43.2?), than to their APS-C diag size (26mm).
Perhaps Finder and Roland may want to revise their estimate to an appropriate fraction?
I'm all ears here ...
ampguy
Veteran
That's out of scope
That's out of scope
What we were all sincerely trying to help the OP's question was if his lens barrel DOF markings for a given Leica lens would still work they way they did on his MP as he transitioned to his M8.
I gave my own results, based and experience with prints and enlargements between my M6 and M8 (I don't have an MP).
The plan called for same lens, same focal distance (tripod in set position). Also, no magnification in pp. I don't think of sensors in terms of crop factors, but for what they are size and quality wise. If a FF with a given lens gives some overscan, like TV's, fine, I will concentrate comparing DOF on the range that is common to both images. I made this clear up front, and it was understood by all who can read.
The whole 1.33, 1.5, 1.6, 2.0 4/3rds is internet and magazine jargon that can help you quickly compare equivalent focal lengths, but to do the math required for analyzing CoC and DOF, you need to know the size of the sensors you're working with, and you need to know what your target CoC is for that sensor.
Leica uses .023 for FF and for 27x18mm sensors. I find both old and newer Leica lenses to adhere to this equally on film and the M8.
Canon (and many other 35mm vendors) use .030 for CoC for full frame. Canon uses .023 for the CoC of it's M8 sized sensor, and .019 for its smaller APS-C 26mm diag - 1.6x crop sensors.
Others used their FF film images, and "emulated" what they thought an M8 would look like DOF-wise with photoshop, which of course is bogus, since you're making an apriori assumption that a) the film image is equal to the live view, and b) that photoshop is a virtual Leica M8 and lens 🙄
What others pointed out to the OP, were that to get equivalent hyperfocal settings with the M8, as on his MP, he would need to move the focus to one more stop wider than normal.
For me, when using hyperfocal, I already stop inwards 2 stops with FF, so what they were telling me was I needed to go 3 stops in.
That can't work if you're not >= 3 stops out from your wide open, and gives marginal rotation room, literally a degree or two max, depending on the lens.
The bottom line is that older lenses are perfectly fine on the M8, and even the RD1. I can't see any DOF differences in RD1 images magnified in print to 2ft x 3 feet compared with full frame images, and if anyone can, that would be most interesting.
That's out of scope
What we were all sincerely trying to help the OP's question was if his lens barrel DOF markings for a given Leica lens would still work they way they did on his MP as he transitioned to his M8.
I gave my own results, based and experience with prints and enlargements between my M6 and M8 (I don't have an MP).
The plan called for same lens, same focal distance (tripod in set position). Also, no magnification in pp. I don't think of sensors in terms of crop factors, but for what they are size and quality wise. If a FF with a given lens gives some overscan, like TV's, fine, I will concentrate comparing DOF on the range that is common to both images. I made this clear up front, and it was understood by all who can read.
The whole 1.33, 1.5, 1.6, 2.0 4/3rds is internet and magazine jargon that can help you quickly compare equivalent focal lengths, but to do the math required for analyzing CoC and DOF, you need to know the size of the sensors you're working with, and you need to know what your target CoC is for that sensor.
Leica uses .023 for FF and for 27x18mm sensors. I find both old and newer Leica lenses to adhere to this equally on film and the M8.
Canon (and many other 35mm vendors) use .030 for CoC for full frame. Canon uses .023 for the CoC of it's M8 sized sensor, and .019 for its smaller APS-C 26mm diag - 1.6x crop sensors.
Others used their FF film images, and "emulated" what they thought an M8 would look like DOF-wise with photoshop, which of course is bogus, since you're making an apriori assumption that a) the film image is equal to the live view, and b) that photoshop is a virtual Leica M8 and lens 🙄
What others pointed out to the OP, were that to get equivalent hyperfocal settings with the M8, as on his MP, he would need to move the focus to one more stop wider than normal.
For me, when using hyperfocal, I already stop inwards 2 stops with FF, so what they were telling me was I needed to go 3 stops in.
That can't work if you're not >= 3 stops out from your wide open, and gives marginal rotation room, literally a degree or two max, depending on the lens.
The bottom line is that older lenses are perfectly fine on the M8, and even the RD1. I can't see any DOF differences in RD1 images magnified in print to 2ft x 3 feet compared with full frame images, and if anyone can, that would be most interesting.
Last edited:
according to my favorite photo instructor
the factors affecting depth of field are only
1) focal length
2) f/stop
3) magnification of the image - (this used to be be enlarger magnification)
4) print or image size
5) viewing distance from eyeball to image
he believed format only affects dof as it affects the the real qualifier
-- the magnification necessary to achieve a certain size image.
Stephen
the factors affecting depth of field are only
1) focal length
2) f/stop
3) magnification of the image - (this used to be be enlarger magnification)
4) print or image size
5) viewing distance from eyeball to image
he believed format only affects dof as it affects the the real qualifier
-- the magnification necessary to achieve a certain size image.
Stephen
antiquark
Derek Ross
What was Leica's advice regarding DoF markings on lenses and the M8? I would just follow that, whatever they said.
ampguy
Veteran
This is from page 87 of M8/M8.2 manual
This is from page 87 of M8/M8.2 manual
here, and was pasted before. Further clarification is provided by Leica in an LFI issue. However, no one seems to care about what Leica says about their own product, heck, they have dofmaster !!
From P. 87 of the Leica M8/8.2 manual:
"The extension factor
The nominal focal lengths of Leica M lenses are based
on the 35mm-format, i.e. on a film format of
24x36mm. In comparison, with its 18x27mm, the
sensor in the LEICA M8 is somewhat smaller though –
by a factor of 0.75. Therefore, when used on the
LEICA M8, these lenses have angles of view corresponding
to lenses with focal lengths that are longer
by a factor of 1.33 (1.33 = reciprocal of 0.75). This
has the respective effects on their perspective, but
not on their depth of field, which, with the LEICA M8,
can also be read directly off the lens (see the lens
instructions for more details).
The bright-line frame in the viewfinder of the LEICA M8
of course always shows the “correct” field of view for
this camera, i.e. it takes account of the increased
focal length. You can therefore compose your pictures
in the normal way, just as for other cameras in the
Leica M series (see also “The bright-line view- and
rangefinder “, p. 102)."
This is from page 87 of M8/M8.2 manual
here, and was pasted before. Further clarification is provided by Leica in an LFI issue. However, no one seems to care about what Leica says about their own product, heck, they have dofmaster !!
From P. 87 of the Leica M8/8.2 manual:
"The extension factor
The nominal focal lengths of Leica M lenses are based
on the 35mm-format, i.e. on a film format of
24x36mm. In comparison, with its 18x27mm, the
sensor in the LEICA M8 is somewhat smaller though –
by a factor of 0.75. Therefore, when used on the
LEICA M8, these lenses have angles of view corresponding
to lenses with focal lengths that are longer
by a factor of 1.33 (1.33 = reciprocal of 0.75). This
has the respective effects on their perspective, but
not on their depth of field, which, with the LEICA M8,
can also be read directly off the lens (see the lens
instructions for more details).
The bright-line frame in the viewfinder of the LEICA M8
of course always shows the “correct” field of view for
this camera, i.e. it takes account of the increased
focal length. You can therefore compose your pictures
in the normal way, just as for other cameras in the
Leica M series (see also “The bright-line view- and
rangefinder “, p. 102)."
ampguy
Veteran
Hi Stephen
Hi Stephen
I don't recall all of the details of my photo classes (in the 1980s), and optical physics classes and labs (also in the 1980s), but with a prime lens with focal length already physically determined, the last 3 items below can be embodied in the CoC value.
The CoC can contain the source and target magnification factors so diagonal of the film and now sensors is used in the computation.
The most common CoC formula is the "Zeiss formula" where:
CoC = diagonal of sensor/fim divided by the z-constant
where the z-constant varies by manufacturers and over time, but is usually a value around 1500. Leica, unlike Zeiss and Canon, designs its lenses with a very conservative .023 value, while Zeiss, Canon and others design their lenses for a CoC value of .030mm (Canon used .023 for their APS-H 1.3x 1d sensors for DOF calculations, and .019 for their APS-C family). Other than DOF markings, the CoC has no other impact of the lens.
The CoC is later used with other variables for DOF calculations that involve those first 2 factors above, along with magnification and viewing factors.
Hi Stephen
I don't recall all of the details of my photo classes (in the 1980s), and optical physics classes and labs (also in the 1980s), but with a prime lens with focal length already physically determined, the last 3 items below can be embodied in the CoC value.
The CoC can contain the source and target magnification factors so diagonal of the film and now sensors is used in the computation.
The most common CoC formula is the "Zeiss formula" where:
CoC = diagonal of sensor/fim divided by the z-constant
where the z-constant varies by manufacturers and over time, but is usually a value around 1500. Leica, unlike Zeiss and Canon, designs its lenses with a very conservative .023 value, while Zeiss, Canon and others design their lenses for a CoC value of .030mm (Canon used .023 for their APS-H 1.3x 1d sensors for DOF calculations, and .019 for their APS-C family). Other than DOF markings, the CoC has no other impact of the lens.
The CoC is later used with other variables for DOF calculations that involve those first 2 factors above, along with magnification and viewing factors.
antiquark
Derek Ross
Then the problem is solved!
I don't think Leica is saying the DoF is identical. It's more likely that, for practical purposes, there's not a big enough difference to matter in regular photographs.
dcsang
Canadian & Not A Dentist
I believe there is confusion in semantics here.
Personally, I don't adhere to everything mentioned in Wikipedia however I do feel they have their ducks in a row on this one:
When you look at Leica's literature, it does not take into account "the same distance". They refer to "angle of view" only. To achieve the "same picture" mentioned in the Wiki entry, you're going to have to change your distance when using the same lens on two different formats (Full Frame versus Cropped sensor).
Cheers,
Dave
Personally, I don't adhere to everything mentioned in Wikipedia however I do feel they have their ducks in a row on this one:
- from http://en.wikipedia.org/wiki/Depth_of_field
When you look at Leica's literature, it does not take into account "the same distance". They refer to "angle of view" only. To achieve the "same picture" mentioned in the Wiki entry, you're going to have to change your distance when using the same lens on two different formats (Full Frame versus Cropped sensor).
Cheers,
Dave
rxmd
May contain traces of nut
Lens maker puts conservative DOF markings on their lens, gets lucky because the crop factor difference is small enough. In particular for living room shots with a 35 at f/8 with nothing in the foreground. News at 11.
You put a lot of effort into this thread, what is it that you're really trying to prove here?
Last edited:
ampguy
Veteran
Hi Dave
Hi Dave
Dave,
Actually it depends - the smaller sensor size with decreased CoC decreases DOF relative to a larger sensor, with larger CoC.
But, a shorter lens, paired with the smaller sensor, to equal the larger sensor and lens FOV will possibly make up for, and potentially increase, the overall DOF range over larger sensor, larger lens, assuming contant target print size, but different magnification factors in the CoCs.
See here: http://www.ronbigelow.com/articles/depth-of-field-1/depth-of-field-1.htm
Hi Dave
Dave,
Actually it depends - the smaller sensor size with decreased CoC decreases DOF relative to a larger sensor, with larger CoC.
But, a shorter lens, paired with the smaller sensor, to equal the larger sensor and lens FOV will possibly make up for, and potentially increase, the overall DOF range over larger sensor, larger lens, assuming contant target print size, but different magnification factors in the CoCs.
See here: http://www.ronbigelow.com/articles/depth-of-field-1/depth-of-field-1.htm
Last edited:
ampguy
Veteran
craygc
Well-known
From P. 87 of the Leica M8/8.2 manual:
The bright-line frame in the viewfinder of the LEICA M8
of course always shows the “correct” field of view for
this camera, i.e. it takes account of the increased
focal length.
Even the Leica manual cant get it right ...there is no increase in focal length due to a crop factor! And nor does the non-term "effective focal length" doesnt even belong here either. 😱
This thread confuses me... about the logic going on here, not the subject. There is terminology, which is specific, although it sometimes tends to get perverted in general use. If you clearly understand the terminology there is nothing to discuss here...
Whether you can visually see the difference in a print is irrelevant to the facts
The bright-line frame in the viewfinder of the LEICA M8
of course always shows the “correct” field of view for
this camera, i.e. it takes account of the increased
focal length.
Even the Leica manual cant get it right ...there is no increase in focal length due to a crop factor! And nor does the non-term "effective focal length" doesnt even belong here either. 😱
This thread confuses me... about the logic going on here, not the subject. There is terminology, which is specific, although it sometimes tends to get perverted in general use. If you clearly understand the terminology there is nothing to discuss here...
- For a given focal length the physical size of the CoC as represented on the film/sensor is the same (but as you dont view the film/sensor, this point finishes here)
- Its when you make same sized print from both you enlarge the CoC of the smaller film/sensor more and that action in turn decrease the perceived DoF on the smaller film/sensor
- If you try and equalise the FoV then the smaller film/sensor must have a shorter focal length, which in itself increases your DoF and then its no longer about focal length DoF vs film/sensor size
Whether you can visually see the difference in a print is irrelevant to the facts
ampguy
Veteran
right ...
right ...
but on that last bullet, we're not trying to do that. One of the questions I asked was same camera on tripod, same lens, focus, f-stop, but image 1 is FF film, image 2 is a smaller piece of film, say 27x18mm, develop and print them both, cutting off the excess overscan of image 1 to equal image 2 FOV.
Is the DOF the same? Some adamantly said no.
right ...
but on that last bullet, we're not trying to do that. One of the questions I asked was same camera on tripod, same lens, focus, f-stop, but image 1 is FF film, image 2 is a smaller piece of film, say 27x18mm, develop and print them both, cutting off the excess overscan of image 1 to equal image 2 FOV.
Is the DOF the same? Some adamantly said no.
- Status