100 macro vs 100 macro bokeh

[gasrocks]

Just ran into a site that was listing bokeh reviews. Claims that the older Canon 100 macro was a 10 while the new Canon 100 macro is only a 7. Anyone with experience on this? Any major differences in construction that would lead to this? Yes, I know it is a very subjective subject. (gee, is that an oxymoron? pun?)

[MagentaJoe]

Interesting. Where is this site you speak of?

[gasrocks]

I did not make a note of it, sorry. One person listing what they thought about 2 dozen different lenses - bokeh.

[Jarrad]

Interesting. Where is this site you speak of?

Google: Mike Johnston Bokeh Ratings

[MagentaJoe]

Free downloadable pdf. Thanks!

http://www.lulu.com/content/129691

[Chandler.]

Good find. I'm surprised the 85L hasn't been rated.

[sebr]

The 100/2 should have a better bokeh because of shallower depth of field if shot wide open, but the reason for the 100 macro's better rating may be that it allows to take photos closer (for macro purpose) and depth of field also depends on distance to subject.

[Mark_Cohran]

The 100/2 should have a better bokeh because of shallower depth of field if shot wide open, but the reason for the 100 macro's better rating may be that it allows to take photos closer (for macro purpose) and depth of field also depends on distance to subject.

Bokeh isn't be a matter of the amount depth of field, but rather the quality of the out of focus areas, and more specifically how the out of focus specularities are rendered. It is a qualitative/subjective criteria rather than a measurable, quantifiable criteria such as depth of field. So, if the lenses are rated for bokeh using the the degree of DOF and the rating is incorrect for the terminology used.

[photobitz]

The 100/2 should have a better bokeh because of shallower depth of field if shot wide open, but the reason for the 100 macro's better rating may be that it allows to take photos closer (for macro purpose) and depth of field also depends on distance to subject.

That's probably true, but the OP stated the difference between the original 100mm macro and the current 100mm macro USM.

I've had a look at the block diagrams for both lenses and they are quite different internally. The older lens has 10 elements in 9 groups whereas the USM has 12 elements in 8 groups. I'm not sure if there is any correlation between the number of groups and bokeh though.... I was of the belief it was more about a combination of aperture shape and depth of field.

[Consensus Trance]

bokeh is for amateurs :)

Edge to edge sharpness, DOF before diffraction, resistance to flare, lack of chromatic aberrations, lack of vignetting, lack of barrel distortion

That's the stuff that really matters

[Mark_Cohran]

bokeh is for amateurs :)

Edge to edge sharpness, DOF before diffraction, resistance to flare, lack of chromatic aberrations, lack of vignetting, lack of barrel distortion

That's the stuff that really matters

Ah, that's the stuff of amateurs. A real photographer knows it's all about the tripod and ball head. :)

[silvex]

he rates the 50f1.8 better than 50f1.4! We all know the f1.4 has a better bokeh. So I question his "test".

[silvex]

he rates the 50f1.8 better than 50f1.4! We all know the f1.4 has a better bokeh. So I question his "test".

Nevermind the FD50f1.8 has better bokeh NOT the EF50f1.8....:oops:

[sebr]

That's probably true, but the OP stated the difference between the original 100mm macro and the current 100mm macro USM.

Sorry, I missed that. Not sure what the difference could be then. Do they have the same aperture blades ?

[Consensus Trance]

Ah, that's the stuff of amateurs. A real photographer knows it's all about the tripod and ball head. :)

shhhhh...don't give away all our secrets hehe

[runninmann]

Sorry, I missed that. Not sure what the difference could be then. Do they have the same aperture blades ?Same number (eight). Not sure about aperture shape, though.

[Canon Bob]

..... but the reason for the 100 macro's better rating may be that it allows to take photos closer (for macro purpose) and depth of field also depends on distance to subject.

Sebr, this isn't strictly true....DOF is a function of aperture and magnification. The only way that subject distance comes into the equation is that lenses have a greater magnification when they get closer to the minumum focus distance.

Bob

[Lester Wareham]

The 100/2 should have a better bokeh because of shallower depth of field if shot wide open, ...

That is not what makes good or bad bokeh, it is about the quality of the out of focus blur and is aesthetic.

This is reputedly related to the combination of a number of things such as how well rounded the iris aperture is, how well matched the MTF characteristics are in the meridional and sagittal and how various aberrations (like spherical aberration) change as objects move away from the plane of sharp focus.

[photobitz]

That is not what makes good or bad bokeh, it is about the quality of the out of focus blur and is aesthetic.

This is reputedly related to the combination of a number of things such as how well rounded the iris aperture is, how well matched the MTF characteristics are in the meridional and sagittal and how various aberrations (like spherical aberration) change as objects move away from the plane of sharp focus.

I wouldn't think Canon would have put rounded aperture blades in the old version and not put them in the USM so quite possibly then, it is just down to the location of all the internal components...?

I wonder also whether the larger number of glass/air medium changes in the old one have some influence?

[R0ger]

This is from the newer lens.

Ignore the dust I was to lazy to remove, :P

[Lester Wareham]

I wouldn't think Canon would have put rounded aperture blades in the old version and not put them in the USM so quite possibly then, it is just down to the location of all the internal components...?

I wonder also whether the larger number of glass/air medium changes in the old one have some influence?

Quite possibly, but don't forget bokeh is an opinion orientated aesthetic thing, unlike MTF/SRF measurements which provide definite number and graph traces.

Just because some bloke has come up with a score value it does not mean you or others would necessarily feel the same way.

If perhaps he had published his test shots you could then make some sort of decision yourself.

But as it is just providing a score number is about as useful as the user surveys you see on lenses.

I wouldn't put any emphasis on something like that at all.

What you can put emphasis on is the major difference between the two of OLE and Internal focus systems. This is so significant operationally it more or less over rides anything else.

[photobitz]

Quite possibly, but don't forget bokeh is an opinion orientated aesthetic thing, unlike MTF/SRF measurements which provide definite number and graph traces.

Just because some bloke has come up with a score value it does not mean you or others would necessarily feel the same way.

If perhaps he had published his test shots you could then make some sort of decision yourself.

But as it is just providing a score number is about as useful as the user surveys you see on lenses.

I wouldn't put any emphasis on something like that at all.

What you can put emphasis on is the major difference between the two of OLE and Internal focus systems. This is so significant operationally it more or less over rides anything else.

That's a valid point. I guess having a scientific mind, I like to attribute effects to specific physical characteristics...

[Lester Wareham]

That's a valid point. I guess having a scientific mind, I like to attribute effects to specific physical characteristics...

Yes I agree absolutely, there are several physical characteristics that affect bokeh, although in some instances it is rather indirect.

It still remains bokeh is aesthetic, there are probably image metrics which could be use to evaluate bokeh like qualities though.

Here are some of the links I came across when I was designing my lens system and researching bokeh:

http://www.flarg.com/bokeh.html
http://www.vanwalree.com/optics/bokeh.html

[kshuler]

As a scientific minded person, I am always bothered by things that can't be measured. Bokeh is, as has been said, a very subjective quality of IMAGES, that depends on the characteristics of a lens, as well as all of the many things that go into the background- lighting, contrast, distance between subject and background, etc. With all of these various effects, it is very difficult to measure how a lens affects the background, teasing it out from the rest of the factors.

But that is just what I am going to try to do. I am proposing a new testing methodology that will examine single points of light from lenses and infer from that how they might affect the bokeh of an image. You may object to this, but before you do, please see my website, which explains WHY I think this will do. Let me know what you think.

http://bokehtests.com/Site/About_Bokeh.html

Klaus
--
http://www.bokehtest.com

[photobitz]

As a scientific minded person, I am always bothered by things that can't be measured. Bokeh is, as has been said, a very subjective quality of IMAGES, that depends on the characteristics of a lens, as well as all of the many things that go into the background- lighting, contrast, distance between subject and background, etc. With all of these various effects, it is very difficult to measure how a lens affects the background, teasing it out from the rest of the factors.

But that is just what I am going to try to do. I am proposing a new testing methodology that will examine single points of light from lenses and infer from that how they might affect the bokeh of an image. You may object to this, but before you do, please see my website, which explains WHY I think this will do. Let me know what you think.

http://bokehtests.com/Site/About_Bokeh.html

Klaus
--
http://www.bokehtest.com

Wow... Are you writing a thesis or something? ;)

[Lester Wareham]

As a scientific minded person, I am always bothered by things that can't be measured. Bokeh is, as has been said, a very subjective quality of IMAGES, that depends on the characteristics of a lens, as well as all of the many things that go into the background- lighting, contrast, distance between subject and background, etc. With all of these various effects, it is very difficult to measure how a lens affects the background, teasing it out from the rest of the factors.

But that is just what I am going to try to do. I am proposing a new testing methodology that will examine single points of light from lenses and infer from that how they might affect the bokeh of an image. You may object to this, but before you do, please see my website, which explains WHY I think this will do. Let me know what you think.

http://bokehtests.com/Site/About_Bokeh.html

Klaus
--
http://www.bokehtest.com

That looks like an excellent site Klaus. I think there might be some images missing right at the end defining the 3 CoC types.

Have you looked at imbalance of meridonal and sagittial MTF and iris shape as indicators of poor bokeh?

[silvex]

As a scientific minded person, I

http://bokehtests.com/Site/About_Bokeh.html

Klaus
--
http://www.bokehtest.com

Scientific indeed! Very well put. I hope the mods put it in the faq.

[tom s]

Macro lenses are known for less than stellar bokeh when used as medium telephoto portrait lenses wide open. The bokeh is not great because of the spheric aberration overcorrection. That's why many macro lenses have ''no bokeh'' in macro shots (everything except the macro subject is blurred, as in the R0ger's photo, see above, but even p&s cameras can do this) and average-to-bad bokeh in portrait shots wide open (too much nisen bokeh, and double lining). An example: http://www.pbase.com/miljenko/image/40762759

Nice bokeh (busy background is blurred, w/300 2.8 ):
http://img339.imageshack.us/img339/7186/dudupaolopotta3vjw8.jpg

[kshuler]

Well, no, I haven't looked at the MTF lines to see how they affect bokeh. Truth be told, I don't see how this possibly COULD affect bokeh, at least in regards to the spherical aberration that is so important. I think the closeness of sagital and meridional lines affects coma and astigmatism, and it is true that this will affect the shape of the out of focus areas (making them oval in the case of astigmatism, and comet tailed in the case of coma). This certainly will affect the out of focus spots, I suppose. In the future I may try to build up a few more of those blurred image diagrams to investigate, but they are SUCH A PAIN TO MAKE. On another forum, I posted a request from some people that stated that this was the main contributor to bokeh for them to explain it to me, so I can model it. No responses yet. Bob Atkins of photo.net himself wrote a few times about this when I searched old posts, but I have never seen anybody explain WHY this is the case. I have read it so many times and never seen it explained, which always has me concerned this is an old wives tale, like colds being caused by the cold. No evidence but it keeps getting passed down through generations and accepted as the truth. So I will pose it to any of you out there... can you explain WHY MTF results should have anything to do at all with bokeh? Are you only referring to coma and astigmatism, or is there more than you can tell me? I would appreciate it.

And iris shape certainly affects bokeh, but I don't think nearly as much as people seem to think. Even with a non-circular iris, spherical aberration will still blur the edges or not, and it is the degree of blurring of edges of an out of focus area that leads to the nisen bokeh effect, how smooth things are, etc. The one thing that will happen is that the diameter of blurred lines going in different directions on the image may be different (same as astigmatism), and any specular highlights that really stand out will have the shape of a hexagon, pentagon, or whatever shape the iris is. But the degree of bluring of the edges I think is still the most important.

Klaus
--
http://www.bokehtests.com

[gasrocks]

I hope everyone who mentions number and shape of blades (as it affects Bokeh) remembers that wide open the blades are not there, should not be affecting anything. Yes, number and shape of blades is something I look for in general but wide open (which is where most people are comparing Bokeh anyway) must be about other parameters.

[Lester Wareham]

Well, no, I haven't looked at the MTF lines to see how they affect bokeh. Truth be told, I don't see how this possibly COULD affect bokeh, at least in regards to the spherical aberration that is so important. I think the closeness of sagital and meridional lines affects coma and astigmatism, and it is true that this will affect the shape of the out of focus areas (making them oval in the case of astigmatism, and comet tailed in the case of coma). This certainly will affect the out of focus spots, I suppose. In the future I may try to build up a few more of those blurred image diagrams to investigate, but they are SUCH A PAIN TO MAKE. On another forum, I posted a request from some people that stated that this was the main contributor to bokeh for them to explain it to me, so I can model it. No responses yet. Bob Atkins of photo.net himself wrote a few times about this when I searched old posts, but I have never seen anybody explain WHY this is the case. I have read it so many times and never seen it explained, which always has me concerned this is an old wives tale, like colds being caused by the cold. No evidence but it keeps getting passed down through generations and accepted as the truth. So I will pose it to any of you out there... can you explain WHY MTF results should have anything to do at all with bokeh? Are you only referring to coma and astigmatism, or is there more than you can tell me? I would appreciate it.

And iris shape certainly affects bokeh, but I don't think nearly as much as people seem to think. Even with a non-circular iris, spherical aberration will still blur the edges or not, and it is the degree of blurring of edges of an out of focus area that leads to the nisen bokeh effect, how smooth things are, etc. The one thing that will happen is that the diameter of blurred lines going in different directions on the image may be different (same as astigmatism), and any specular highlights that really stand out will have the shape of a hexagon, pentagon, or whatever shape the iris is. But the degree of bluring of the edges I think is still the most important.

Klaus
--
http://www.bokehtests.com

That is my thought. There are some links out there that look into this. I am sure a search will turn them up.

[kshuler]

I will try to investigate how coma affects the bokeh of an image. As for astigmatism, seems like this should really only be important for specular highlights, right? It shouldn't turn any blurred edges into sharp edges, and, other then specular highlight shapes, the only effect I can foresee it having would be to make out of focus lines different diameters in different orientations. I think for a good look of what extreme curvature of field, astigmatism and coma all combined look like, you can examine pictures from a "lensbaby" and see what the bokeh looks like.

Klaus
--
http://bokehtests.com/Site/Introduction.html

[Lester Wareham]

I will try to investigate how coma affects the bokeh of an image. As for astigmatism, seems like this should really only be important for specular highlights, right? It shouldn't turn any blurred edges into sharp edges, and, other then specular highlight shapes, the only effect I can foresee it having would be to make out of focus lines different diameters in different orientations. I think for a good look of what extreme curvature of field, astigmatism and coma all combined look like, you can examine pictures from a "lensbaby" and see what the bokeh looks like.

Klaus
--
http://bokehtests.com/Site/Introduction.html

Sounds good. Astig will certainly show up in tight highlight areas (probably mostly specular then).

Not sure what the general effect would be. Simulation would be the easiest way. I might have some stuff I used a long while back for that to see the effect of lens MTFs on small 100% crops. I remember its spatial frequency modelling could cope with asymeticric frequencies although for implicity (as I was working with small crop areas) it was aligned to cartezian coordinates rather than polar.