M4/3 and Bokah,

[Georgelee]

Hi, I keep reading on the internet that M4/3 does not do bokah well, I am no expert, but I think the attached looks good. I would be interested in comments. I took this image on Saltburn Pier a couple of weeks ago, I used and OMD EM5, and the 75mm F1.8, the exposure was 1/2500, F2, ISO 400.

Best wishes

George

[ajohnw]

A lot of people just assume shorter focal length so greater depth of field but forget that the sensor is smaller so for the same size of final image it needs enlarging more so the bokah actually on the sensor needs to be smaller than larger formats.

The actual relationship is shown by some depth of field calculators. Under the same conditions-field of view, for a 10x8 final image the out of focus size limit for full frame is usually stated as 0.030mm, aps 0.19mm and m 4/3 0.15mm. This goes some way to offsetting the increased depth of field caused by the shorter focal length. Bigger image and of course these limits have to become smaller or the image looked at from further away.

I don't think it's a relevant factor really in practice but wider apertures are needed but not by as much as some might suppose.

There is a DOF calculator some where on the web that plots DOF against distance but I no longer have the link.

John
-

[jcuknz]

I saw a post recently where a guy was trying out his new bug-eye lens on MFT and what struck me most was the blurred backgrounds over more than a dozen examples
In any case we have editing programmes to do what may not have been enough in-camera.

[Shadowman]

Examples.

http://www.thephotoargus.com/inspira...h-photography/

[inkista]

To me it's not that m4/3 "doesn't do bokeh well"; you can get creamy smooth dreamy bokeh with m4/3. It's just that it requires extra effort to get there, and is a little harder to do on m4/3 than with APS-C or full-frame.

For example, your shot is taken with the 75/1.8. A $1000 150mm f/1.8 equivalent. I can get similar thin DoF and background blur on my 5DMkII with a $400 EF 85mm f/1.8 USM lens. And while I'm sure the new Nocticron 42.5mm f/1.2 is going to kick ass, it's still not going to look like an 85/1.2L does on full-frame just because that's how focal lengths and subject distances affect DoF.

The crop factor means that to get an equivalent composition, with a smaller sensor, you're either going to use a wider lens or shoot from farther away than you would with a bigger sensor. Both of those things will increase DoF.

With micro four-thirds, I have a harder time getting a lot of background blurring without going to close subject distances with the 20/1.7. Nature of the beast.

[MattNQ]

George,
The image you posted is fine. The background is blurred nicely, but I do find that those vertical lines on the RH side are a little distracting still, even when blurred. Whether lightening or darkening them would work better, I'm not sure.


As a general comment, I don't find bokeh any harder on a m4/3 than an APS-C. (Never used a full frame, so can't comment on that - most users seem to agree full frame is the ultimate solution for the best bokeh). Comparing APS-C & m4/3,
it is more dependent on your lens, aperture, and shooting position than the sensor. For me, my biggest complaint with m4/3 is the shutter speed on my EPL1. At a max of 1/2000, you cannot shoot wide open at large apertures in bright daylight.
Does one need expensive lenses?...not at all.
My $100 Minolta Rokkor X 50/1.4 is a wonderful lens. I get some very nice results from it.

[Shadowman]

George,
The image you posted is fine. The background is blurred nicely, but I do find that those vertical lines on the RH side are a little distracting still, even when blurred. Whether lightening or darkening them would work better, I'm not sure.


As a general comment, I don't find bokeh any harder on a m4/3 than an APS-C. (Never used a full frame, so can't comment on that - most users seem to agree full frame is the ultimate solution for the best bokeh). Comparing APS-C & m4/3,
it is more dependent on your lens, aperture, and shooting position than the sensor. For me, my biggest complaint with m4/3 is the shutter speed on my EPL1. At a max of 1/2000, you cannot shoot wide open at large apertures in bright daylight.
Does one need expensive lenses?...not at all.
My $100 Minolta Rokkor X 50/1.4 is a wonderful lens. I get some very nice results from it.

You can get some crazy bokeh with a mirror lens on an APS-C camera.
http://www.bestbuy.com/site/rokinon-...&skuId=2523303

[William W]

Your image is fine. I like the Bokeh in your image.

But I hasten to add that Bokeh is NOT “Shallow Depth of Field”.
Bokeh is a SUBJECTIVE analysis/critique of the Out of Focus portion of the image.

That analysis and opinion is usually confined to ONLY discussing the Out of Focus portion of the image which is BEHIND the main Subject.

Also, Bokeh is usually associated with an image that has a shallow Depth of Field and Bokeh is usually discussed when critiquing and image made by a large maximum aperture lens, used at a large aperture.

BUT the analysis of Bokeh is NOT restricted to only those conditions.

Also Bokeh may change if the Background Texture and /or the Lighting on the Background is changed even if the same Lens and the same Aperture settings are used for the same framed shot.

For example here is (large) sample of Bokeh made with a quite slow zoom lens and with that lens used toward the wide angle end:

Portrait of Woman Walking in the Rain – Milan 2013



WW

[DanK]

(Never used a full frame, so can't comment on that - most users seem to agree full frame is the ultimate solution for the best bokeh)

Nope. For two reasons.

First, as WW pointed out, bokeh is not the same thing as shallow depth of field.

Second, if shallow DOF is what you want, then you just need to change aperture to offset differences in sensor size. The best summary I have seen of the effect of sensor size on DOF is here: http://photo.net/learn/optics/dofdigital/. So, assume that you are comparing a small and large sensor camera, and you have chosen lenses that will give you the same field of view on both. Then the larger sensor will give you shallower DOF at any given aperture. Open up the aperture more on the small-sensor camera, and you can offset this. This works as long as you have room to change apertures. There is a calculator on one of the tutorials on this site that will calculate how much you have to change the aperture: https://www.cambridgeincolour.com/tu...ensor-size.htm.

[ajohnw]

Nope. For two reasons.

First, as WW pointed out, bokeh is not the same thing as shallow depth of field.

Second, if shallow DOF is what you want, then you just need to change aperture to offset differences in sensor size. The best summary I have seen of the effect of sensor size on DOF is here: http://photo.net/learn/optics/dofdigital/. So, assume that you are comparing a small and large sensor camera, and you have chosen lenses that will give you the same field of view on both. Then the larger sensor will give you shallower DOF at any given aperture. Open up the aperture more on the small-sensor camera, and you can offset this. This works as long as you have room to change apertures. There is a calculator on one of the tutorials on this site that will calculate how much you have to change the aperture: https://www.cambridgeincolour.com/tu...ensor-size.htm.

That is a good link Dan. Pity it doesn't put that in extra aperture needed terms due to smaller sensor size.

There is another aspect as well. An Olympus 45mm F1.8 isn't that expensive a lens. And yet another aspect. Scale that up to a 6x9 camera and the DOF at F1.8 would be great for taking shots of a flat piece of paper. That in some ways is an advantage of formats that size because the lens would be stopped down leading to better optical quality but a longer exposure if ISO is the same.

One days some one will do a meaningful comparison - aperture against the same depth of field. Gets complicated though as the next aspect would be just how far behind the subject background would need to be to get the same effect..

This area to me is a little like some reviewers comparing lenses and as one is 1/2 or even 1 stop slower than another saying it is less usable in low light work when in real terms the lens is unlikely to be used at max aperture anyway.

Debates on both areas tend to take a very simplistic view.

John
-

[William W]

One days some one will do a meaningful comparison - aperture against the same depth of field. . . but a longer exposure if ISO is the same.

If you mean ONLY Aperture and DoF related to Camera Format and then add in the ISO required - then there is a massive study of that.

You might research "Equivalence Theory" as it relates to Photography.

+++

Gets complicated though as the next aspect would be just how far behind the subject background would need to be to get the same effect. . .

Then we are discussing the matrix of the OoF circle (which is closer to what Bokeh is all about) and for that, as a starting point you can research Bob Atkins' downloadable, "Blur Calc" tool and his mathematics associated with it.

+++

Debates on both areas tend to take a very simplistic view.

Not really. Depends where one looks.

WW

[RustBeltRaw]

A summary of the points everyone's making:

• Bokeh quality is determined by the lens. See the distinctive circular bokeh from Lomography's Petzval lens.
• Bokeh size is coupled to depth-of-field-related factors. It can be increased by any combination of wider aperture, larger sensor/media, and greater distances between subject and background.

Getting deeply into nerd territory, according to Michael Reichmann of Luminous Landscape, MTF charts provide insight into a lens's bokeh. The closer the meridonial and sagittal (solid and dashed) lines are on a lens's chart, the more likely is it to have pleasing bokeh. This ignores effects such as aperture shape, and there will always be aesthetic judgments in play. To get an idea of how deeply committed bokeh fanatics can get, see any Leica forum.

When shooting an MFT camera and aiming for good bokeh, decreasing your depth of field via any of the techniques above will amplify the lens's bokeh. But the lens will always be the deciding factor.

[rpcrowe]

Since, I shoot more often with longer focal length lenses than many photographers, bokeh is quite important to me, My using the longer focal lengths will provide a more narrow depth of field so the aesthetic quality of the OOF areas is meaningful to me...


7D with 70-200mm f/4L IS lens. 165mm (264mm equivalent) @ f/5.6

I love the 70-200mm f/4L IS lens with its rounded aperture blades because of the creamy smooth bokeh it will produce. Many photographers automatically assume that a fast lens is necessary for shallow DOF. The above image shows, that under the right circumstances (with the subject being close to the camera and the background being a distance away) a smaller aperture combined with a longer focal length will produce a shallow DOF. However, there is another factor to consider when you are talking about bokeh: what the background is consists of. If you have a background with considerable spectral highlights, that BG will be busy, even when OOF and shot with a lens that traditionally produces very smooth bokeh...


90mm Tamron macro @ f/11

The above was a choice of using a wider aperture and not being able to capture the entire insect in focus (the end of its body is going OOF even at f/11) but, being able to blur the BG even more or shooting at my chosen aperture and blurring the BG not as much as I would have preferred...

The reason, IMO, for selecting a shallow DOF is to center the viewer's interest on the subject. If the bokeh is ragged, or even if it is not ragged but, if the background has many busy highlights, the OOF areas can compete with the subject for the eye of the viewer...

[William W]

A summary of the points everyone's making:

• Bokeh quality is determined by the lens. See the distinctive circular bokeh from Lomography's Petzval lens.
• Bokeh size is coupled to depth-of-field-related factors. It can be increased by any combination of wider aperture, larger sensor/media, and greater distances between subject and background.

That's where I disagree. And that's where I think it is very important to be pedantically accurate.

'Bokeh' by definition is "the (aesthetic) quality of the Out of Focus area". And that is NOT 'determined' by the lens: however the lens is ONE OF the constituting factors of Bokeh - there are other factors, for example: the Aperture used; the lighting; the distance of the background from the Subject; the texture of the background; the angle of light on the background.

The SIZE of the Out of Focus "tiny bits" can be calculated by a tool such as 'Blur Calc' - but that is not calculating the "size of" the "Bokeh", because "Bokeh" does not have "size"


WW

[RustBeltRaw]

And that's where I think it is very important to be pedantically accurate.

Agreed. In my attempt at brevity, I may have sacrificed some precision.

From a technical perspective, I stand by my statements. Every camera-lens combination will have its own aesthetic bokeh qualities, which can be amplified by the same techniques used to achieve shallow depth of field. For instance, one will never achieve nonagonal bokeh with a 5-bladed aperture, and aperture geometry is only interchangeable on select lenses. Lenses prized for their bokeh tend to have small deltas in their meridonial and sagittal MTF curves. Reducing your depth of field will increase the rendered size of anything out of focus. These are immutable, technical facts, and you're right, they do not make aesthetic judgments.

You are also correct that bokeh has no "size," but the out-of-focus elements constituting it do. Shooting two different 35mm f/1.4 lenses on the same camera will produce identically-sized bokeh elements with different overall bokeh aesthetics. It is probably most useful to compare the differences in bokeh element size as a function of camera settings.

[William W]

... From a technical perspective, I stand by my statements. Every camera-lens combination will have its own aesthetic bokeh qualities ...

Hi Lex
To be quite clear I did NOT disagree nor argue that point.
What I was pointing out was that I disagree with this statement:

"Bokeh quality is determined by the lens"

-and I would also disagree if you wrote this:

"Bokeh is determined by the lens"

I disagree because of the use of the word "determined".
"determined" has a meaning of ‘absolute’; ‘conclusive’.
And my point being that the Lens, is only ONE factors of what will constitute the Bokeh in any given situation. And I think that it is very important not to lose sight of that important point.

*
I agree with this statement:

. . . “Every camera-lens combination will have its own aesthetic bokeh qualities”

WW

[ajohnw]

Agreed. In my attempt at brevity, I may have sacrificed some precision.

From a technical perspective, I stand by my statements. Every camera-lens combination will have its own aesthetic bokeh qualities, which can be amplified by the same techniques used to achieve shallow depth of field. For instance, one will never achieve nonagonal bokeh with a 5-bladed aperture, and aperture geometry is only interchangeable on select lenses. Lenses prized for their bokeh tend to have small deltas in their meridonial and sagittal MTF curves. Reducing your depth of field will increase the rendered size of anything out of focus. These are immutable, technical facts, and you're right, they do not make aesthetic judgments.

You are also correct that bokeh has no "size," but the out-of-focus elements constituting it do. Shooting two different 35mm f/1.4 lenses on the same camera will produce identically-sized bokeh elements with different overall bokeh aesthetics. It is probably most useful to compare the differences in bokeh element size as a function of camera settings.

The "Bokeh" can have size Lex. It relates to circle of confusion and the size of the detail that is being "bokehised". More often than not though people tend to regard this area as having a smooth creamy texture with no significant detail in it at all more or less as per the shot that started this post. The Lomgraphy shot is use of circles of confusion and can be calculated.

The basic relationship is in respect to formats is as I expected. Subject at 3m required depth of field 300mm

Full frame 90mm lens F4.8
APS (canon) 56mm lens F2.8
M 4/3 :-) Guess who 45mm F2.4

These assume the usual max circle of confusion sizes.

The bokeh or degree of circle of confusion then comes down to how far the back ground is behind the subject.. This is what happen in that respect. The APS is different as no exact F ratio was available.



I'm tempted to mention sucking eggs but will resist. But as said - often looked at in a rather simplistic fashion. This also all relates to a 10x8 in final image size that has some how turned to 12x8.

John
-