4/3 and depth of view

That part of the article on Four-Thirds is, actually, mine.

Wikipedia doesn't like articles that portray things, especially products, in either unambiguously positive or negative terms; it's not quite kosher to say "4/3 has more DOF, but if you know what you're doing you can still get the shot you want."

So here's my actual experience.

4/3 has more dof at any given f/number and focal length than your friend's Nikon. That's just a fact. But it still has a LOT less than any compact camera. Depth of field is determined by the crop factor -- very roughly, the lower the crop factor, the less DOF.

So here are the crop factors:

Film/full-frame: 1
Canon/Nikon/Sony DSLR: 1.6
Four-Thirds: 2
High-end compact (Panasonic FZ50 or LX3, for instance): 5
Most compacts: 7

As you see, there's not much difference between 4/3 and Canon/Nikon. In practice, I have never had problems wanting less dof, and often want more, on 4/3. Those "portrait-style shots" you want have more to do with lens focal length, aperture, subject distance, background distance, etc., and not that much to do with the camera system.

If you're wanting to absolutely demolish the background when doing a full-body portrait at any distance, then you're going to need a very expensive lens on whatever system you use. You'll also have a nearly impossible time getting the focus right.

But if you want to be able to get a pleasantly out-of-focus background in reasonable shooting conditions, 4/3 can certainly do it. You do NOT need a super-wide-aperture lens on fullframe to get pleasantly out-of-focus backgrounds. You do need that sort of stuff to get one eyeball in focus and one eyeball out at 50 feet, but how often do you really want that?

You'll have a harder time using the kit lens, which like all kit lenses has a narrow aperture and short-(ish) focal length. It's too short to shoot soccer in any case. You want a longer lens for that. If you want to blur backgrounds from closer distances, you'll need a fast portrait lens.

Fortunately, you can get those for 4/3.

The Olympus 50/2 is widely seen as the portrait lens on 4/3. It's not cheap ($400-$500 I think), but it is amazingly sharp and has a narrow enough DOF to do what you want. For even more aperture, there's the Sigma 30/1.4 and 50/1.4.

If you want something more practical but slightly less sharp and with a little less aperture, get the Sigma 18-50/2.8 ($350?)

The 40-150 that you get in a two-lens kit will actually do a respectable job, and it's cheap. If you want something even longer, get the Olympus 70-300/4-5.6 ($270). You can use these long lenses to do "portrait-style" shots too -- here's one of mine with the 70-300.

 
...all of which were taken with lenses than are within reach of most forum participants (as opposed to the 90-250mm for example), helped to dispel some of the myths regarding 4/3's and put a few minds at ease. Though, I doubt it.

I'm sure we'll be seeing yet another 4/3's DOF issue thread within a couple of days. :-(
feel the need to post an inordinate number of images in an attempt to belittle the benefits of FourThirds.
Where is Kaizer Soze when we need him? :-)

Bill Turner
--
Olympus E-3, E-620 and E-420

Zuiko 9-18mm, 25mm pancake, 50mm, 14-42mm, 12-60mm, 50-200mm, and 8mm fisheye. FL-36R and FL-50R Flashes. HLD-4 Grip.
Canon PowerShot TX1
Ricoh GR-D
Sony DSC-V3
 
In regards to that US$ 700 kit lens (Nikon 18-200) of yours, it seems like it's worse than the US $200 Oly kit lens when comparing their blur charts at maximum aperture at http://www.slrgear.com . Take a look at it yourself.
May i ask which camera did you use for these pictures?
Did you shoot raw at that time and what post processing did you do on them?
Please, i like to know....
All shot in raw, developed in Capture NX. Lenses; Nikon 18-200, 70-200, 85/1.8, Sigma 150, I believe that is all.

http://pbase.com/sngreen
 
Gee Sergey

About $11,000 worth of gear (in Oz: D700 body = $3,800 + f2.8/300 = $7,800), and he still has his pics over-sharpened ... Why the sharpening halos around the dog's head, ears, whiskers and muzzle if this stuff is so flaming good?

Just asking ... ;)

Well taken, though ...

--
Regards, john from Melbourne, Australia.
-- -- --

The Camera doth not make the Man (or Woman) ...
Perhaps being kind to cats, dogs & children does ...

Gallery: http://canopuscomputing.com.au/gallery2/main.php
Hints & Tips (temporary link, as under construction):
http://canopuscomputing.com.au/index.php?p=1_9



Bird Control Officers on active service.

Member of UK (and abroad) Photo Safari Group
looks sharpened fine to me, can't see the halos you mentioned
--
http://illy.smugmug.com
 
Gidday Illy
Gee Sergey

About $11,000 worth of gear (in Oz: D700 body = $3,800 + f2.8/300 = $7,800), and he still has his pics over-sharpened ... Why the sharpening halos around the dog's head, ears, whiskers and muzzle if this stuff is so flaming good?

Just asking ... ;)

Well taken, though ...
looks sharpened fine to me, can't see the halos you mentioned
Well mate, it is certainly clear as anything from where I am sitting!

I always reckon that if one can tell that an image has been manipulated, then one has over-done it - unless that is one's aim, of course. It is often difficult to tell this about any image at web resolutions, so to be able to see it at all, I would suggest that the image is pretty rough at full resolution.

As an example, here is one of my OoC JPEGs re-sized for the web, converted to sRGB and with an USM of 28%, 1.8 pixels applied (automatic PS action I wrote; no other PP). Can you see any sharpening or other artifacts in this image?



Taken with my E-510 and f2/50.

Is this image technically better than the dog jumping?

It most certainly was nowhere near as hard to take, as the flower was just moving in the capricious wind (stormy day in the Melbourne foothills), and I had to wait for a break in the movement.

And my camera body and lens cost how much?

--
Regards, john from Melbourne, Australia.
-- -- --

The Camera doth not make the Man (or Woman) ...
Perhaps being kind to cats, dogs & children does ...

Gallery: http://canopuscomputing.com.au/gallery2/main.php
Hints & Tips (temporary link, as under construction):
http://canopuscomputing.com.au/index.php?p=1_9



Bird Control Officers on active service.

Member of UK (and abroad) Photo Safari Group
 
Gee Sergey

About $11,000 worth of gear (in Oz: D700 body = $3,800 + f2.8/300 = $7,800), and he still has his pics over-sharpened ... Why the sharpening halos around the dog's head, ears, whiskers and muzzle if this stuff is so flaming good?

Just asking ... ;)

Well taken, though ...
looks sharpened fine to me, can't see the halos you mentioned
Well mate, it is certainly clear as anything from where I am sitting!

I always reckon that if one can tell that an image has been manipulated, then one has over-done it - unless that is one's aim, of course. It is often difficult to tell this about any image at web resolutions, so to be able to see it at all, I would suggest that the image is pretty rough at full resolution.

As an example, here is one of my OoC JPEGs re-sized for the web, converted to sRGB and with an USM of 28%, 1.8 pixels applied (automatic PS action I wrote; no other PP). Can you see any sharpening or other artifacts in this image?



Taken with my E-510 and f2/50.

Is this image technically better than the dog jumping?

It most certainly was nowhere near as hard to take, as the flower was just moving in the capricious wind (stormy day in the Melbourne foothills), and I had to wait for a break in the movement.

And my camera body and lens cost how much?

--
Regards, john from Melbourne, Australia.
-- -- --

The Camera doth not make the Man (or Woman) ...
Perhaps being kind to cats, dogs & children does ...

Gallery: http://canopuscomputing.com.au/gallery2/main.php
Hints & Tips (temporary link, as under construction):
http://canopuscomputing.com.au/index.php?p=1_9



Bird Control Officers on active service.

Member of UK (and abroad) Photo Safari Group
i was just saying that i don't see any over sharpening, not a comparison, not denigrating any system, at full res i would imagine it would be pretty similar looking to 100% of other dslr shots
--
http://illy.smugmug.com
 
Hi again Illy
Gee Sergey

About $11,000 worth of gear (in Oz: D700 body = $3,800 + f2.8/300 = $7,800), and he still has his pics over-sharpened ... Why the sharpening halos around the dog's head, ears, whiskers and muzzle if this stuff is so flaming good?

Just asking ... ;)

Well taken, though ...
looks sharpened fine to me, can't see the halos you mentioned
Well mate, it is certainly clear as anything from where I am sitting!

I always reckon that if one can tell that an image has been manipulated, then one has over-done it - unless that is one's aim, of course. It is often difficult to tell this about any image at web resolutions, so to be able to see it at all, I would suggest that the image is pretty rough at full resolution.

As an example, here is one of my OoC JPEGs re-sized for the web, converted to sRGB and with an USM of 28%, 1.8 pixels applied (automatic PS action I wrote; no other PP). Can you see any sharpening or other artifacts in this image?

http://canopuscomputing.com.au/gallery2/d/7846-3/E-510_JAK_2009-_7045043_Ew.jpg

Taken with my E-510 and f2/50.

Is this image technically better than the dog jumping?

It most certainly was nowhere near as hard to take, as the flower was just moving in the capricious wind (stormy day in the Melbourne foothills), and I had to wait for a break in the movement.

And my camera body and lens cost how much?
i was just saying that i don't see any over sharpening, not a comparison, not denigrating any system, at full res i would imagine it would be pretty similar looking to 100% of other dslr shots
Sorry then. I was just stating that there are no sharpening artifacts in my humble image, where the ones linked by Sergey do show these. I will not compound what he has done by further linking someone else's images here; nor embed them.

I can easily see the sharpening artifacts. If one expands the first or second of the images linked by Sergey by zooming in one level, one can see the halo very easily. I mean absolutely no disrespect to the original photographer, quite the contrary - I could never take an action shot like that!

However, I would never be satisfied with this kind of image processing either, whether done in-camera or in PP, I do not know.

It is Sergey who is claiming some kind of superiority of one camera/system over another (as per usual ... ); not me. I am merely posting one of my own images to demonstrate my point in defence of my system of choice ... After all, this IS the Olympus dSLR Forum ; not the Nikon Forum ... ;)

--
Regards, john from Melbourne, Australia.
-- -- --

The Camera doth not make the Man (or Woman) ...
Perhaps being kind to cats, dogs & children does ...

Gallery: http://canopuscomputing.com.au/gallery2/main.php
Hints & Tips (temporary link, as under construction):
http://canopuscomputing.com.au/index.php?p=1_9



Bird Control Officers on active service.

Member of UK (and abroad) Photo Safari Group
 
Sergey, I think you just proved my point. Amin is comparing 2 lenses with the same field of view and what one has to do to achieve a similar DOF on the 2 lenses. The fact is the 25mm f2.8 lens is still a 25mm f2.8 lens and not a 50mm f1.4.
The fact is that the 28 mm ƒ/2.8 is indeed a 25 mm ƒ/2.8 and it produces results equivalent to those of a 50 mm ƒ/5.6 on full frame.

Not to a 50 mm ƒ/1.4 lens on full frame; that would require a thus far hypothetical 25 mm ƒ/0.7 lens.
 
feel the need to post an inordinate number of images in an attempt to belittle the benefits of FourThirds.
we think alike
I'm not surprised.
Where is Kaizer Soze when we need him? :-)
call it the Keyser Söze syndrome
Thanks for the link and the correction! :-)
http://en.wikipedia.org/wiki/Keyser_S%C3%B6ze

"A lot of recent films seem unsatisfied unless they can add final scenes that redefine the reality of everything that has gone before"....

--
ʎǝlıɹ

plɹoʍ ǝɥʇ ɟo doʇ uo ǝɹɐ ǝʍ 'ɐılɐɹʇsnɐ uı
Bill Turner
 
OK, I get it.I think...

I have a gut feeling, though, that the OP wrote about "shallow/wide field of view" meaning DOF and mixing those entities (FOV vs DOF) up a bit...

It looked that way to me reading his post over once again that he meant DOF the whole time, and sometimes called it "field of view".
How, otherwise, could he talk about "shallow" field of view?

Aim & Frame ;-)
It´s not a "shame" just stating the fact, that the DOF at same FOV, same print size & same aperture is "deeper" on 4/3 than on fex APS-C or FWIW on FF.
Yes, it is a fact. If you are comparing FOV side by side, there will be a difference.

But the OP wrote (like many before him who have written similar things):

"I am especially concerned with field of view with not just macro shots

can anyone provide input on the wide/shallow field of view with 4/3.

Honestly I am looking at a SLR in the first place because of the constant "portrait" type pics my sister gets with a D40. Not possible with my point and shoot. "

I'm also stating a fact, you can get the same portrait type pics as you would with the mentioned D40, you just have to use the lens best suited for that on a 4/3's camera and perhaps physically take a couple of steps back if necessary.

--
Olympus E-3, E-620 and E-420

Zuiko 9-18mm, 25mm pancake, 50mm, 14-42mm, 12-60mm, 50-200mm, and 8mm fisheye. FL-36R and FL-50R Flashes. HLD-4 Grip.
Canon PowerShot TX1
Ricoh GR-D
Sony DSC-V3
 
Sergey, I think you just proved my point. Amin is comparing 2 lenses with the same field of view and what one has to do to achieve a similar DOF on the 2 lenses. The fact is the 25mm f2.8 lens is still a 25mm f2.8 lens and not a 50mm f1.4.
The fact is that the 28 mm ƒ/2.8 is indeed a 25 mm ƒ/2.8 and it produces results equivalent to those of a 50 mm ƒ/5.6 on full frame.

Not to a 50 mm ƒ/1.4 lens on full frame; that would require a thus far hypothetical 25 mm ƒ/0.7 lens.
Not so, actually. But since you keep peddling this "equivalence" rubbish, I think I will just add you to my 'ignore' list ... saves me the trouble of reading this nonsense; and you the aggravation of my responses ... ;)

BTW, an f2.8/25 lens on 4/3rds has an Effective Focal Length (EFL) or Angle of View (AoV) of a 50mm lens on a 36x24 sensor (almost, more or less ... ), except for the difference in sensor aspect ratio, and any number of other non-random variables ... However, it always remains an f2.8 lens .

What you are suggesting (peddling ... ?) means that my OM f3.5/28 (which is a lens designed for a 36x24 film format) suddenly becomes "equivalent" to an f6.3/56 lens when used on a 4/3rds camera, while remaining an f3.5 lens when used on a "FF" digital camera - what bunkum! It is an f3.5 lens (exactly the same aperture diameter, no matter which format it is placed in front of ... ) with more or less, approximately, sort of, the almost same AoV as a 56mm lens on a "FF" camera.

What you are suggesting is the same nonsense that Mr James tried to peddle here for years, and it has not gotten any more accurate with the re-telling - it is still b/s; sorry.

AND, just for completeness, my f2/50 macro is still razor sharp at f22 (see SLRGear test ... ) so "FF" cannot match the DoF advantage "any time it likes by stopping down". Almost no "FF" lens is sharp at f22 (none?), let alone critically sharp, as the f2/50 macro is at f22, BUT applying the "equivalence theory" nonsense, the "FF" lens would have to be as sharp at f64 as the f2/50 macro is at f22 ... pull the other leg, mate :P - NEVER going to happen!

Goodnight FG, I am off to bed, as late here.

--
Regards, john from Melbourne, Australia.
-- -- --

The Camera doth not make the Man (or Woman) ...
Perhaps being kind to cats, dogs & children does ...

Gallery: http://canopuscomputing.com.au/gallery2/main.php
Hints & Tips (temporary link, as under construction):
http://canopuscomputing.com.au/index.php?p=1_9



Bird Control Officers on active service.

Member of UK (and abroad) Photo Safari Group
 
...all of which were taken with lenses than are within reach of most forum participants (as opposed to the 90-250mm for example), helped to dispel some of the myths regarding 4/3's and put a few minds at ease. Though, I doubt it.
Your photos are most welcomed as opposed to an infinite number of Nikon photos posted for no other reason than to soothe an ego.
I'm sure we'll be seeing yet another 4/3's DOF issue thread within a couple of days. :-(
Sadly, how true and the Usual Suspects will return.
feel the need to post an inordinate number of images in an attempt to belittle the benefits of FourThirds.
Where is Kaizer Soze when we need him? :-)

Bill Turner
--
Olympus E-3, E-620 and E-420

Zuiko 9-18mm, 25mm pancake, 50mm, 14-42mm, 12-60mm, 50-200mm, and 8mm fisheye. FL-36R and FL-50R Flashes. HLD-4 Grip.
Canon PowerShot TX1
Ricoh GR-D
Sony DSC-V3
--
Bill Turner
 
AND, just for completeness, my f2/50 macro is still razor sharp at f22 (see SLRGear test ... )
Mine isn't razor sharp way beyond the diffraction limit. Interesting. Anyway, CoC and thus what is in acceptable focus of course depends on how much you enlarge.
so "FF" cannot match the DoF advantage "any time it likes by stopping down". Almost no "FF" lens is sharp at f22 (none?), let alone critically sharp, as the f2/50 macro is at f22, BUT applying the "equivalence theory" nonsense, the "FF" lens would have to be as sharp at f64 as the f2/50 macro is at f22 ...
No, no, the "equivalence nonsense" operates with two stops between 4/3rds and FF. That means stopping down to ƒ/44 on FF in your example (was your ƒ/64 just a typo?)

You probably don't care, but when you set the two systems' parameters to produce equivalent pictures, they will be diffraction limited at the same point. If 4/3rds is not diffraction limited at ƒ/8, neither will FF be at ƒ/16.
Goodnight FG, I am off to bed, as late here.
Sleep well, JK.

P.S. I warmly recommend Warren J. Smith, Modern Optical Engineering . Any edition.
 
Dear Chris,

first of all, I think your pictures are superb. I really enjoy looking at them. Unfortunately, that didn't happen when I was reading what you wrote. first, not because someone thinks (and it's convinced) about something different than you, it's a troll. Then you will be a troll for me and I will be one for you, and certainly, I don't consider you a troll.

Second, when people talk about equivalences, many (not all) are based in a camera model, and there are some hundreds of years of studies, measurements, and calculations that support those models. I've been helping on a computational photography course, we built a pinhole camera, and we explained the students how to use the models.

I think the other big problem is that when anyone want to bash the 4/3 system they compare it with 35mm sensors, and then they (somehow) conclude that APS is much better than 4/3, even though the difference is very small.

So, for me the important part is to separate the trolls/fanboys that will use any argument to bash 4/3, from the ones that want to do a quantitative comparison.

cheers,

Jose
 
As long as you have the necessary space and realize that you change the perspective.
well, if you are trying to dissolve the background, I don't think the change of perspective will be that important. The space can be important sometimes.
 
Hi,

I have a question that maybe you can answer to me. Why the diffraction limit doesn't depend on the focal length of the lens? I'm seriously asking, since I haven't found an answer anywhere. for me it looks like that there should be a strong dependence, but i haven't found it anywhere.

Thanks,

Jose
 
If I understand correctly, even though some people believe the kit lenses are "better than typical" for olympus, to really make these camera bodies work, you need to upgrade the lenses. So even if I bought a nikon or canon, the same people would recommend new lenses. Fine, I am in the same situation no matter what I buy.

BestBuy currently sells the 2 lens 620 kit for $719, and for a 1st SLR I believe this is too hard to pass up. I will have plenty to monkey with for awhile.

I can buy lenses as needed, if and when the hobby grows over time.

I love your reference to canon/nikon 1.6, olympus 2, and most compacts 7. Currently I feel I have a 7. I understand this math.

I fully believe the d40 or any SLR for that matter would really make me thrilled - even though many here refer to "other companies entry models as junk." I also believe the e620 for couple hundreds more does give me lots more camera. I just wanted to make sure that when I shell out $800 I am able to get some "bokeh." One eye in focus, the other out, is nowhere near what I would ever expect. A defender soccer player in focus with the net in the background 20 yds further behind slightly out of focus would be just about right.
 
Really, the difference in height between 4/3rds and APS-C is very minimal (width difference is more pronounced). For example the Sigma APS-C is 13.8mm tall and the 4/3rds is 13.5mm tall. Three tenths of a mm folks. Nikon APS-C is a little bigger at 15.8mm tall, but even then when you consider that 35mm sensors are 24mm tall and typical digicam sensors are 4.6mm tall well...
Don't compare the heights. Compare the diagonals.
 

Keyboard shortcuts

Back
Top