Help Understanding Olympus DOF

[AnandaSim]

You made the diagram yourself?

Mentioning CoC too early just makes more MEGO for strugglers.....



--



Ananda
http://anandasim.blogspot.com/
http://olympuse510.wikispaces.com/
http://picasaweb.google.com/AnandaSim/
http://www.flickr.com/photos/32554587@N00/

 

[AnandaSim]

AnandaSim wrote:
Not because the lens is different, not because the camera sensor is
different, but because you used your legs to walk back and you
changed the distance to the subject.

So to break this down a bit, the DOF is based on the combination of
the lens and the length of your legs?

Yes, the SDFS being the Stride Distance From Subject.

Just kidding. Thank you for a good explanation Ananda. Ive been
curious about this for a while.

I'm glad I helped in the epiphany. It's sometimes really hard to get it even though it is simple because the explanation is difficult.

I spend a lot of time in spreadsheets and have taught absolute and relative addressing of cells since 1990 and yet, each new non math person blinks heavily to grasp it.

--



Ananda
http://anandasim.blogspot.com/
http://olympuse510.wikispaces.com/
http://picasaweb.google.com/AnandaSim/
http://www.flickr.com/photos/32554587@N00/

 

[AnandaSim]

True, but for the photographer who buys a lens because of it's angle
of view, this viewpoint is irrelevant. The photographer is more
likely to ask how much depth of field you can get with a normal lens,
for example. Here, the focal length obviously changes with the format.

True Simon, but in this case, the OP has already bought the e-510 and is in the E-system. He is probably reading bits here and there about DOF and wants somehow to understand his camera and lenses, not compare different sensor sizes.

In this case, it is much better to gain practical and pragmatic experience by looking at DOF charts, going to the field and shooting, with DOF Preview, coming back to computer and create his own feedback loop then to spend time trying to extract nuance from off the cuff remarks of books or discussions not related to his sensor size.

--



Ananda
http://anandasim.blogspot.com/
http://olympuse510.wikispaces.com/
http://picasaweb.google.com/AnandaSim/
http://www.flickr.com/photos/32554587@N00/

 

[Venkat Venkatesan]

Perhaps someone could clear this up for me.

If I've followed the discussion so far, a 50 mm f/2 lens on a 4/3 body should result in the same depth of field (and field of view) as a 100 mm f/4 lens on a 35 mm body (let's ignore the difference in aspect ratio for the purposes of this discussion).

Now suppose the 35 mm camera above (with a 100 mm lens set to f/4) requires a shutter speed of 1/100 sec at ISO 800 for proper exposure. Then the 4/3 camera (with a 50 mm lens set to f/2, for the same depth of field and field of view) would achieve the same exposure with a shutter speed of 1/100 sec at ISO 200. Right?

If so, then wouldn't it be meaningful to compare the noise at ISO 800 of the 35 mm camera with the noise at ISO 200 (not 800) of the 4/3 camera, since it corresponds to the same shutter speed and depth of field?

Venkat

 

[AnandaSim]

Hi Venkat,

I appreciate that you want to ask equivalence. But there are lots of people who are struggling with the idea of DOF.

I hope you and others who want to talk about equivalence start a new thread, not add to this one as equivalence to ISO and DOF makes discussion even more complicated because this is another parameter that has nothing to do with DOF, it has to do with noise and exposure.

Some people have already bought a new camera, and they are learning how to make background blur or not blur. That is their aim for the moment. They may not care about how their camera compares against a Nikon D3 now or ever in terms of noise and high ISO until they master DOF first. If they have saved and saved like the OP and bought their first camera, it makes no sense to say "oh, but a Nikon D3 will have so and so DOF and so and so high ISO noise".

Perhaps someone could clear this up for me.

If I've followed the discussion so far, a 50 mm f/2 lens on a 4/3
body should result in the same depth of field (and field of view) as
a 100 mm f/4 lens on a 35 mm body (let's ignore the difference in
aspect ratio for the purposes of this discussion).

Now suppose the 35 mm camera above (with a 100 mm lens set to f/4)
requires a shutter speed of 1/100 sec at ISO 800 for proper exposure.
Then the 4/3 camera (with a 50 mm lens set to f/2, for the same depth
of field and field of view) would achieve the same exposure with a
shutter speed of 1/100 sec at ISO 200. Right?

If so, then wouldn't it be meaningful to compare the noise at ISO 800
of the 35 mm camera with the noise at ISO 200 (not 800) of the 4/3
camera, since it corresponds to the same shutter speed and depth of
field?

Venkat

--



Ananda
http://anandasim.blogspot.com/
http://olympuse510.wikispaces.com/
http://picasaweb.google.com/AnandaSim/
http://www.flickr.com/photos/32554587@N00/

 

[love_them_all]

Bokeh is the ability of the lens to render OOF shapes. Going wide open will give you a more OOF background from a shallower DOF. But nice Bokeh mainly comes from:

-the aperture opening of the lens. If the lens has a rounder aperture the OOF object would look more natural and pleasing. e.g. the contax zeiss 50 1.4 can generate shallow DOF but it's not famous for bokeh.

-shoot closer to the subject to get shallow DOF, and then have the background as far away as possible

-longer lenses help

I read that to get good bokeh, one needs an aperture of lower than
5.6. If this is true, which system would this have been referring to
and what is the Olympus minimum requirement? Thanks
--

--

 

[gianstam]

You made the diagram yourself?

Yes

Mentioning CoC too early just makes more MEGO for strugglers.....

The truth (and a more photographic view) is that that the DoF depends only on the area that passes the light (for a given distance). Knowing that, someone can explain why a 50mm lens @f2 achieves the DoF of a 100mm lens @f4. They both passes the light through the same [3.14X{(50/2) 2}^2]=[3.14X{(100/4) 2}^2]=491mm^2 area

 

[Dragos Jianu]

It's not something you can grasp through numbers. But here is a side by side comparison i've made. APSC + 50mm f/2 vs FF + 85mm f/2. Wish i could compare it to 4/3 but i didn't have an equivalent FL at f/2.

 

[baddboy]

Ananda and others,

The only modification I think might need to be made to your understanding is that the DOF is also dependent on the diameter of teh circle of confusion, Coc, which is surely dependent on the density of photosites (now commonly referred to as "pixels") on the sensor. Therefore we should not expect the DOF to be the same on an E-1 and an E-3, all other things being equal. It makes sense that if the resolution is higher, the DOF will be lower.

I might be absolutely wrong here, but I say what I say after having a go at doing some calculations using the equations in Wikipedia (yes, I confess, I use Wikipedia). In these equations there is a factor c which I took as the dimensions of the photosite (which seemed reasonable).

I just checked again and DOF Master appears not to vary the CoC between an E-1 and an E-3. However, if you manually select different CoCs, it can be seen that the DOF does vary.

I as a filmographer also thought in terms of the relations between formats and focal lengths etc still being quoted, and in the film era all applied because the resolution at the film was the basically the same in all cases, but this is obviously no longer true in teh digital age.

As I say, I did some calcs, but I never did finish my investigation. I have only an E-1 and cannot easily do any comparisons. I do not see people saying that there is a practical difference in DOF between an E-1 with only 5Mp and any of the other 4/3 cameras, so I may have got it wrong somewhere. Perhaps CoC is not strongly related to pixel pitch as I have assumed (but then why would a 10Mp sensor give more detail? We know we are not lens-limited).

I have been firebombed here in the past for saying this!

If anyone can explain where I've gone wrong, if I have, much appreciated.

Cheers, Baddboy.

 

[gianstam]

I do not see
people saying that there is a practical difference in DOF between an
E-1 with only 5Mp and any of the other 4/3 cameras, so I may have got
it wrong somewhere.

Theoretically it's true while it's difficalt to count the difference practically.

If, let's say, a CoC is spreaded on an area of 300X300 pixels (square) on the E-1 sensor, the same sensor area at E-3 sensor will have quite more pixels. If we print these two images at 300dpi the discussed CoC will be 1sq inch in the E-1 print and larger in the E-3. But note that the paper size will be different in this case.

If we reduce the E-3 paper size to much the E-1 print the two areas will be the same.
It is just an opoinion

 

[simpy]

It is still relevant in this age of multiple sensor sizes. For
example a 50 mm lens on a Canon full frame sensor camera (e.g. 5D)
and on an APS sized sensor camera (e.g. 40D) will have the same debth
of field at the same aperture depite it being a "normal" lens on a 5D
but a "portrait" lens on a 40D.

Also if I am using a legacy Olympus OM 50 mm f1.4 lens on both an OM
film body as well as an E-520 with an adapter, the depth of field at
f1.4 is the same even though it is a normal lens on the former camera
body but a long lens on the four thirds body.

True, if you intend to use the same lens on multiple formats. In my comment I was assuming this wasn't the case for most photographers.

This has always relevant in the past when comparing lenses for 35 mm
film vs medium format vs large format.

Only if the photographer uses the same lens, right? Because once you have different lenses for each format, you will consider each lens at its 'effective focal length' and 'effective aperture (for DOF purposes).

I do believe that English Bob is right in the sense that the aperture
is a ratio of the aperture diameter to the focal length of the lens.
Therefore the diameter of the aperture of a 50 mm lens set to f2 is
probably the same as a 100 mm lens set to f4. I do need to double
check this though.

Correct.

Simon

 

[simpy]

True, but for the photographer who buys a lens because of it's angle
of view, this viewpoint is irrelevant. The photographer is more
likely to ask how much depth of field you can get with a normal lens,
for example. Here, the focal length obviously changes with the format.

True Simon, but in this case, the OP has already bought the e-510 and
is in the E-system. He is probably reading bits here and there about
DOF and wants somehow to understand his camera and lenses, not
compare different sensor sizes.

That could indeed be the case, although the original question was about a comparison. Also, in all fairness, your explanation started from a comparison as well

Simon

 

[AnandaSim]

That could indeed be the case, although the original question was
about a comparison. Also, in all fairness, your explanation started
from a comparison as well

Again Simon, that is not incorrrect. I'm trying to put myself in a newbie's shoes - usually when we learn a new language or we come to a new country, we want to interpret what the new thing is in terms of what we know as the old thing. Therefore, we can't avoid discussing making quick comparisons with a different language or a different country / culture.

Very soon however, we have to operate well in this new country - doesn't matter how people do it in other countries, for the time being, all that matters is, the country we're in.



--



Ananda
http://anandasim.blogspot.com/
http://olympuse510.wikispaces.com/
http://picasaweb.google.com/AnandaSim/
http://www.flickr.com/photos/32554587@N00/

 

[profborg]

Hi everybody, sorry for not posting at all, havent been on the forums for a day or two but many thanks for all of your informative responses, especially Ananda.

My point in posting was really to fully understand DOF/aperture to get better creative control. What Anandasim said made sense, but then English Bob went and ruined it all!

If if I am standing with an Olypmus 5mm f/2 and a FF 100mm f/2 at an equal distance from a subject, why wouldnt the DOF be the same?
--
Daniel

 

[simpy]

If if I am standing with an Olypmus 5mm f/2 and a FF 100mm f/2 at an
equal distance from a subject, why wouldnt the DOF be the same?

Because the aperture of the full frame lens is twice as large as that of the 4/3 lens (50mm vs. 25mm). To really 'understand' it, you'll have to visualize lens diagrams or play with the equations.

Simon

 

[profborg]

How is the aperture different if they have the same f stop?
--
Daniel

 

[haziz]

If if I am standing with an Olypmus 5mm f/2 and a FF 100mm f/2 at an
equal distance from a subject, why wouldnt the DOF be the same?
--
Daniel

Look at my posts in this thread. Precisely because the depth of field of a 50 mm lens at f2 is the same regardless of the size "sensor" behind it, whether the sensor is four thirds, a roll of 35 mm film in an Olympus OM2 camera or 120 roll film (6x7) in a Mamiya 7 or for that matter 4x5 sheet film in a large format camera. The image circles of these lenses (all do in fact exist, though you would have to substitute 47 mm or 55 mm in 4x5 in reality) will obviously need to be differnet to cover these differing "sensors", but that would open an entirely new can of worms. Their field of view is also different it is a normal lens on 35 mm film but a megawide on 4x5.

A 50 mm lens at f2 will in fact have the same DOF as a 100 mm lens at f4 at the same camera to subject distance. The diameter (not F stop) of the aperture in absolute numbers in mm will be the same.

BTW I assume you meant 50 mm rather than 5 mm which would be a megawide in all of these formats and technically impossible in most of them.

Thanks.

Sincerely,

Hany.

 

[simpy]

How is the aperture different if they have the same f stop?

The f-number is the ratio of the focal length to the effective aperture diameter (entrance pupil diameter). Google 'aperture' for more.

Two lenses with the same f-stop will be equally 'bright', i.e. illuminate the sensor with the same brightness, but the lens with the larger focal length (further away) will need a larger aperture to achieve this.

Simon

 

[Pedagydusz]

Badboy,

you introduced yet another variable, which just shows that these comparisons are more complex than we would think, on a first approach.

But I don't think that you are on the right track, though what you say about the CoC is correct.

Comparing an E-330 with an E-3 (both 4/3s, one 7 MP, other 10 MP), I found the pictures undistinguishable regarding OOF background.

I think that when we go from less to more MP, the resolution increases (everything else being equal). So, we have more detail in the subject and a slightly more blurred background (if the lens outperforms the sensor, which it may not in the corners, or in case of diffraction limited operation). But the difference, while it may show in the resolution of a sharp subject, is to small to notice in the OOF characteristics of the background.

Note that this is just an impression, based on observation. Most of what was written above by other posters, in answer to the OP, is hard science; here I am just trying to find a qualitative explanation to an observation, based on general knowledge. If I am wrong, I will be delighted to learn the truth!
--
Antonio

http://ferrer.smugmug.com/

 

[profborg]

If if I am standing with an Olypmus 5mm f/2 and a FF 100mm f/2 at an
equal distance from a subject, why wouldnt the DOF be the same?
--
Daniel

Look at my posts in this thread. Precisely because the depth of field
of a 50 mm lens at f2 is the same regardless of the size "sensor"
behind it, whether the sensor is four thirds, a roll of 35 mm film in
an Olympus OM2 camera or 120 roll film (6x7) in a Mamiya 7 or for
that matter 4x5 sheet film in a large format camera. The image
circles of these lenses (all do in fact exist, though you would have
to substitute 47 mm or 55 mm in 4x5 in reality) will obviously need
to be differnet to cover these differing "sensors", but that would
open an entirely new can of worms. Their field of view is also
different it is a normal lens on 35 mm film but a megawide on 4x5.

Are you saying that because the 50mm lens is wider, its DOF is greater because its focused area takes up more space in the frame, and when it is cropped, the focused area is blown up (magnified due to cropping of the other area) and then the DOF is greater?

A 50 mm lens at f2 will in fact have the same DOF as a 100 mm lens at
f4 at the same camera to subject distance. The diameter (not F stop)
of the aperture in absolute numbers in mm will be the same.

That makes no sense unless I have understood it wrong. Are you saying that the actual aperture size in mm is greater in FF than 4/3?

BTW I assume you meant 50 mm rather than 5 mm which would be a
megawide in all of these formats and technically impossible in most
of them.

Sorry, you are right. Too old to edit though.

--
Daniel

Olympus e-five10 + 14-42, 40-150