This topic is in the 2014 archive. You can read it and like its posts here. It is read only.
Archive 2014Lens tests, 17mm/18mm
p.3 #1Lens tests, 17mm/18mm
Peter Figen wrote:
To my eye, the 17 TS-E was still the best of the bunch by a small margin, and also had the lowest amount of color fringing. What this test didn't show was the relative distortion between the different lenses, and what you'd probably find is that the 17 would likely have the lowest distortion of the bunch as well. A third phenomenon that's a bit harder to quantify of course, is curvature of field, which can made some lenses look better in certain situations than they might appear elsewhere.
The biggest virtue of the TS-17L over and above its baseline optical goodness is, of course, that it renders possible tilt and shift movements. At extreme ends of the shift range there is a penalty to be paid. It is also highly susceptible to flare (in a few instances, this can be used to creative advantage). On the whole, one heck of a lens.
To my eye, the 17 TS-E was still the best of the bunch by a small margin, and also had the lowest amount of color fringing. What this test didn't show was the relative distortion between the different lenses, and what you'd probably find is that the 17 would likely have the lowest distortion of the bunch as well. A third phenomenon that's a bit harder to quantify of course, is curvature of field, which can made some lenses look better in certain situations than they might appear elsewhere.
The biggest virtue of the TS-17L over and above its baseline optical goodness is, of course, that it renders possible tilt and shift movements. At extreme ends of the shift range there is a penalty to be paid. It is also highly susceptible to flare (in a few instances, this can be used to creative advantage). On the whole, one heck of a lens.
Jan 18, 2014 at 01:40 PM
p.3 #2Lens tests, 17mm/18mm
Rajan Parrikar wrote:
The TS-17L [...] is also highly susceptible to flare
Jim's test results, in this very thread, seem to show otherwise.
The TS-17L [...] is also highly susceptible to flare
Jim's test results, in this very thread, seem to show otherwise.
Jan 18, 2014 at 01:49 PM
p.3 #3Lens tests, 17mm/18mm
Monito wrote:
Jim's test results, in this very thread, seem to show otherwise.
In that case, my experience is at variance with what Jim found. See the first image here, for instance (where I use its susceptibility to flare) -
http://www.parrikar.com/blog/2013/04/01/berg-city/
Rajan Parrikar wrote
The TS-17L [...] is also highly susceptible to flareIn that case, my experience is at variance with what Jim found. See the first image here, for instance (where I use its susceptibility to flare) -
http://www.parrikar.com/blog/2013/04/01/berg-city/
Jan 18, 2014 at 01:51 PM
p.3 #4Lens tests, 17mm/18mm
Peter -- I agree. There is also the vignetting, which favors the 17TSE and Jim touched on it. Those four factors place the 17TSE well in advance of the 16-35II for flexibility. But, it is very encouraging that the zoom shows so well for its imaging: sharpness, contrast, color, etc. and most of the other minuses can be ameliorated with PP. Really a great showing, and a complete surprise to many readers who don't have either or both these lenses.
I think Dan's analysis is somewhat slanted, ignoring the fact that once again it is the most sophisticated imaging device (at the highest price) that turns in the best result. And the surprisingly close runner up is still a $1700 lens -- nothing to sneeze at, and second highest cost. If anything, the test helps confirm that you generally get what you pay for. Work-around are available, trading mental sweat equity and computer/software expense for lower priced lens alternatives.
I think Dan's analysis is somewhat slanted, ignoring the fact that once again it is the most sophisticated imaging device (at the highest price) that turns in the best result. And the surprisingly close runner up is still a $1700 lens -- nothing to sneeze at, and second highest cost. If anything, the test helps confirm that you generally get what you pay for. Work-around are available, trading mental sweat equity and computer/software expense for lower priced lens alternatives.
Jan 18, 2014 at 02:05 PM
p.3 #5Lens tests, 17mm/18mm
Gunzorro wrote:
And the surprisingly close runner up is still a $1700 lens -- nothing to sneeze at, and second highest cost. If anything, the test helps confirm that you generally get what you pay for.
But isn't that generally the case? It is that incremental improvement in spec that costs a great deal more. Some will claim that that incremental delta is of no consequence in real world situations and is not worth the increase in cost. Other will disagree. This is what underlies most of the disagreements.
Edited on Jan 18, 2014 at 02:14 PM · View previous versions
And the surprisingly close runner up is still a $1700 lens -- nothing to sneeze at, and second highest cost. If anything, the test helps confirm that you generally get what you pay for.
But isn't that generally the case? It is that incremental improvement in spec that costs a great deal more. Some will claim that that incremental delta is of no consequence in real world situations and is not worth the increase in cost. Other will disagree. This is what underlies most of the disagreements.
Edited on Jan 18, 2014 at 02:14 PM · View previous versions
Jan 18, 2014 at 02:12 PM
p.3 #6Lens tests, 17mm/18mm
My flare experience with the 17TSE.
I believe the lens is excellent at controlling flare and produces good flare performance when the sun is in the frame. But like all bulbous lenses without shades, it can pick up sun that is out of the frame. This is something the flatter shaded lenses are less prone to do.
I believe the lens is excellent at controlling flare and produces good flare performance when the sun is in the frame. But like all bulbous lenses without shades, it can pick up sun that is out of the frame. This is something the flatter shaded lenses are less prone to do.
Jan 18, 2014 at 02:13 PM
p.3 #7Lens tests, 17mm/18mm
I'd like to separate the two types of "flare".
One is the specular ghosting (prismatic color trail) cast off a predominant point light source in the image -- I consider these to be "ghosts" or ghosting flares. This is the one that seems to be mainly addressed on this thread.
The more general term "flare" denotes an across-the-image contrast reduction or fogging caused by bright light sources -- usually caused by poor lens coatings or other poor baffling causing light to scatter throughout the elements.
I find the 17TSE to be a very low flare lens by the general definition of flare. But its hemispherical front element catches bright light sources will outside the image area causing the prismatic flaring unless guarded against by a well placed flag. Once the ghosts have been eradicated, the lens performs very well, even with large bright areas or multiple point light sources/reflections within the picture -- contrast is very high.
One is the specular ghosting (prismatic color trail) cast off a predominant point light source in the image -- I consider these to be "ghosts" or ghosting flares. This is the one that seems to be mainly addressed on this thread.
The more general term "flare" denotes an across-the-image contrast reduction or fogging caused by bright light sources -- usually caused by poor lens coatings or other poor baffling causing light to scatter throughout the elements.
I find the 17TSE to be a very low flare lens by the general definition of flare. But its hemispherical front element catches bright light sources will outside the image area causing the prismatic flaring unless guarded against by a well placed flag. Once the ghosts have been eradicated, the lens performs very well, even with large bright areas or multiple point light sources/reflections within the picture -- contrast is very high.
Jan 18, 2014 at 02:19 PM
p.3 #8Lens tests, 17mm/18mm
Rajan Parrikar wrote:
But isn't that generally the case? It is that incremental improvement in spec that costs a great deal more. Some will claim that that incremental delta is of no consequence in real world situations and is not worth the increase in cost. Other will disagree. This is what underlies most of the disagreements.
Ha-ha! You don't have to convince ME!
Try getting that across to Dan. (Dan, you know you double posted your last comments, right?)
Yes, Rahan, we all have to decide where to draw the line based our finances and the relative importance of some gear. Jim's test show that for top quality UWA, we are going to need to look at these sort of top end solutions.
Gunzorro wrote
And the surprisingly close runner up is still a $1700 lens -- nothing to sneeze at, and second highest cost. If anything, the test helps confirm that you generally get what you pay for.Ha-ha! You don't have to convince ME!
Try getting that across to Dan. (Dan, you know you double posted your last comments, right?)Yes, Rahan, we all have to decide where to draw the line based our finances and the relative importance of some gear. Jim's test show that for top quality UWA, we are going to need to look at these sort of top end solutions.
Jan 18, 2014 at 02:25 PM
p.3 #9Lens tests, 17mm/18mm
I had an instructor at Art Center who defined flare as non image forming light. I think that's still a pretty good way of putting it. The 17 ts-e, overall is very good, but is prone to some pretty weird reflections of the many internal elements. It's almost always an easy retouch, but something to be aware of. Unfortunately, as great a lens as that one is, and there's simply nothing on the market that even comes close to what it can do, you have to be very very careful with any amount of tilt or swing. With even a degree of movement, the far corners really fall apart. May or may not matter depending on image, but it ain't a Super Angulon. That's for sure.
Jan 18, 2014 at 02:31 PM
p.3 #10Lens tests, 17mm/18mm
Peter Figen wrote:
[17 TS-E] you have to be very very careful with any amount of tilt or swing. With even a degree of movement, the far corners really fall apart.
It would be interesting to see a test comparing performance unshifted with shifted. When I tested a 24 TS-E (first version) I was disappointed in the shifted sharpness and the heavy vignetting (exposure reduction) at large shifts. The latter can be compensated, but not the former. My test was quick and not rigorous.
[17 TS-E] you have to be very very careful with any amount of tilt or swing. With even a degree of movement, the far corners really fall apart.
It would be interesting to see a test comparing performance unshifted with shifted. When I tested a 24 TS-E (first version) I was disappointed in the shifted sharpness and the heavy vignetting (exposure reduction) at large shifts. The latter can be compensated, but not the former. My test was quick and not rigorous.
Jan 18, 2014 at 02:48 PM
p.3 #11Lens tests, 17mm/18mm
I think there are a variety of types of flare, but most can be categorized as either;
(i) veiling flare, or
(ii) flare structures.
Sometimes, veiling flare covers the whole frame, and sometimes parts of the frame. I think that all of the flare artifacts shown in this thread are flare structures, except for maybe the light, pinkish blob on the Oly. The 35mm tests that I'll soon publish on the Alt forum show regions of medium-density, red-ish veiling flare for some conditions, as well as some interesting flare structures.
(i) veiling flare, or
(ii) flare structures.
Sometimes, veiling flare covers the whole frame, and sometimes parts of the frame. I think that all of the flare artifacts shown in this thread are flare structures, except for maybe the light, pinkish blob on the Oly. The 35mm tests that I'll soon publish on the Alt forum show regions of medium-density, red-ish veiling flare for some conditions, as well as some interesting flare structures.
Jan 18, 2014 at 03:13 PM
p.3 #12Lens tests, 17mm/18mm
Gunzorro, Ben, and I have been amongst the few strong supporters of 16-35II, still recognizing its limitations of course. Your results very much confirm experience I have had comparing the lens with the likes of the excellent 24tse, and nikon 14-24 2.8. the 16-35II is far from sharp in the corners wide open, but from 5.6 on, turns in a very good performance, corner to corner quite adequate of 30x20 prints from 5d2/5d3. it also does ok with flare (much better than the original) and has awesome starburst effect. its a very versatile lens, and I think still the best ultrawide zoom choice for full frame canon.
Jan 18, 2014 at 03:35 PM
p.3 #13Lens tests, 17mm/18mm
Rajan Parrikar wrote:
The "rage" that you see is not visible to the naked eye. May I point out that disagreement with your views, even in instances where it is strong, is not tantamount to "rage."
I see that you have softened your view about TS lenses now. Earlier I recall you describing them as "exotic" or some such.
Yep, Dan needs to fit in somewhere. He will side against Zeiss even if he has to plug his nose and accept TSE glass as being great. Funny how things go isn't it.
gdanmitchell wrote
To try to soften the rage that sometimes ensues, let me begin by saying that the 17TS is demonstrably a really, really fine lens and is exactly the lens that some photographers need for some kinds of photography. So, I am not here to say anything negative about the 17TS or any other particular lens.I see that you have softened your view about TS lenses now. Earlier I recall you describing them as "exotic" or some such.
Yep, Dan needs to fit in somewhere. He will side against Zeiss even if he has to plug his nose and accept TSE glass as being great. Funny how things go isn't it.
Jan 18, 2014 at 04:19 PM
p.3 #14Lens tests, 17mm/18mm
Gunzorro wrote:
Peter -- I agree. There is also the vignetting, which favors the 17TSE and Jim touched on it. Those four factors place the 17TSE well in advance of the 16-35II for flexibility. But, it is very encouraging that the zoom shows so well for its imaging: sharpness, contrast, color, etc. and most of the other minuses can be ameliorated with PP. Really a great showing, and a complete surprise to many readers who don't have either or both these lenses.
I don't know if I would say well in advance for flexibility when the 16-35 is a zoom and has AF. Those are two pretty damn big components to flexibility in my book and why I own it.
Peter -- I agree. There is also the vignetting, which favors the 17TSE and Jim touched on it. Those four factors place the 17TSE well in advance of the 16-35II for flexibility. But, it is very encouraging that the zoom shows so well for its imaging: sharpness, contrast, color, etc. and most of the other minuses can be ameliorated with PP. Really a great showing, and a complete surprise to many readers who don't have either or both these lenses.
I don't know if I would say well in advance for flexibility when the 16-35 is a zoom and has AF. Those are two pretty damn big components to flexibility in my book and why I own it.
Jan 18, 2014 at 04:59 PM
p.3 #15Lens tests, 17mm/18mm
ggreene wrote:
I don't know if I would say well in advance for flexibility when the 16-35 is a zoom and has AF. Those are two pretty damn big components to flexibility in my book and why I own it.
That is kind of my overall point in these posts - that once you get to a certain level of, say, resolution (which can vary a bit depending upon your subject and so forth), resolution differences become less significant that other factors that sometimes get left in the dust.
As an example, let's consider two kinds of photographers - out of a wide range of types that you might imagine or even be yourself.
One is a product photographer who works in a relatively controlled shooting environment - backgrounds might be constructed rather than found, lighting is artificial and controllable, subjects rarely move, and so forth. Such a photographer might well find one of the longer TS lenses to be nearly idea for quite a bit of his/her work and the potential advantages of larger aperture or zoom might not be very important at all. For this person, the definition of "flexibility" might largely be about flexibility to tilt the plane of focus. (I have friends who specialize in food photographer who fit this category perfectly.)
Another is a landscape photographer who produces very large fine art prints of a wide range of subjects and often works far out in the landscape. Shooting at relatively smaller apertures diminishes potential resolution differences among lenses, most of which can produce quite good resolution in any case. Being able to adapt the focal length to a particular composition is very important, and for this photographer "flexibility" might be defined by what the zoom lens can do. For this photographer, who may virtually always manually focus, things like the speed or accuracy of the AF system or the ability to tilt and shift or image stabilization may have little or no appeal.
Another is photographing active subjects - lets say some sports or things like moving wildlife at a distance. For this person a zoom might not help in the flexibility area that much since he/she may simply need to shoot with the longest reasonable focal length available. If shooting in natural light, it might be important to get the largest aperture in order to keep shutter speed as high as possible and to deal with possibility that one will shoot handheld and more. Along the same lines the effectiveness of the AF system may be paramount to this shooter and IS may be indispensable.
Another photographer may shoot street photography in the "classic" manner, working quickly with a single focal length prime and always shooting handheld and not worrying too much about perfect focus or elimination of motion blur, preferring instead the spontaneity of the simplest possible system rather than the adaptability of zooms and TS and so forth.
Without telling any of you which lenses you should or should not use, I think it is reasonable to say a few things about all of this. First, that clearly there is much more to lens selection that answering the "which is sharpest" question. Second, once you consider the type of photography, the functional differences among lenses start to count for much more than the IQ differences. Third, the balance of which variables are most important will vary tremendously depending upon the nature of the shooting. Fourth, for some photographers doing more specialized types of work, more specialized lenses that function less well in more general photography may be appropriate. Fifth, a much larger group of photographers - a group that includes a lot of very serious and critical photographers - find that zooms are quite often the best choice.
Take care,
Dan
Edited on Jan 19, 2014 at 10:22 AM · View previous versions
Gunzorro wrote
Peter -- I agree. There is also the vignetting, which favors the 17TSE and Jim touched on it. Those four factors place the 17TSE well in advance of the 16-35II for flexibility. But, it is very encouraging that the zoom shows so well for its imaging: sharpness, contrast, color, etc. and most of the other minuses can be ameliorated with PP. Really a great showing, and a complete surprise to many readers who don't have either or both these lenses.That is kind of my overall point in these posts - that once you get to a certain level of, say, resolution (which can vary a bit depending upon your subject and so forth), resolution differences become less significant that other factors that sometimes get left in the dust.
As an example, let's consider two kinds of photographers - out of a wide range of types that you might imagine or even be yourself.
One is a product photographer who works in a relatively controlled shooting environment - backgrounds might be constructed rather than found, lighting is artificial and controllable, subjects rarely move, and so forth. Such a photographer might well find one of the longer TS lenses to be nearly idea for quite a bit of his/her work and the potential advantages of larger aperture or zoom might not be very important at all. For this person, the definition of "flexibility" might largely be about flexibility to tilt the plane of focus. (I have friends who specialize in food photographer who fit this category perfectly.)
Another is a landscape photographer who produces very large fine art prints of a wide range of subjects and often works far out in the landscape. Shooting at relatively smaller apertures diminishes potential resolution differences among lenses, most of which can produce quite good resolution in any case. Being able to adapt the focal length to a particular composition is very important, and for this photographer "flexibility" might be defined by what the zoom lens can do. For this photographer, who may virtually always manually focus, things like the speed or accuracy of the AF system or the ability to tilt and shift or image stabilization may have little or no appeal.
Another is photographing active subjects - lets say some sports or things like moving wildlife at a distance. For this person a zoom might not help in the flexibility area that much since he/she may simply need to shoot with the longest reasonable focal length available. If shooting in natural light, it might be important to get the largest aperture in order to keep shutter speed as high as possible and to deal with possibility that one will shoot handheld and more. Along the same lines the effectiveness of the AF system may be paramount to this shooter and IS may be indispensable.
Another photographer may shoot street photography in the "classic" manner, working quickly with a single focal length prime and always shooting handheld and not worrying too much about perfect focus or elimination of motion blur, preferring instead the spontaneity of the simplest possible system rather than the adaptability of zooms and TS and so forth.
Without telling any of you which lenses you should or should not use, I think it is reasonable to say a few things about all of this. First, that clearly there is much more to lens selection that answering the "which is sharpest" question. Second, once you consider the type of photography, the functional differences among lenses start to count for much more than the IQ differences. Third, the balance of which variables are most important will vary tremendously depending upon the nature of the shooting. Fourth, for some photographers doing more specialized types of work, more specialized lenses that function less well in more general photography may be appropriate. Fifth, a much larger group of photographers - a group that includes a lot of very serious and critical photographers - find that zooms are quite often the best choice.
Take care,
Dan
Edited on Jan 19, 2014 at 10:22 AM · View previous versions
Jan 18, 2014 at 05:13 PM
p.3 #16Lens tests, 17mm/18mm
Jim, thanks for taking the time to research and post this information.
As a person who is trying to have a minimum number of lenses that suit my needs and budget, this is the type of post that keeps me coming back.
I don't have a problem with buying expensive glass if I will use it a lot (hence not much duplication/overlap).
Right now I can't think of a lens or body I'm willing to plunk down the dough for.
As a person who is trying to have a minimum number of lenses that suit my needs and budget, this is the type of post that keeps me coming back.
I don't have a problem with buying expensive glass if I will use it a lot (hence not much duplication/overlap).
Right now I can't think of a lens or body I'm willing to plunk down the dough for.
Jan 18, 2014 at 06:18 PM
p.3 #17Lens tests, 17mm/18mm
Tapeman wrote:
I don't have a problem with buying expensive glass if I will use it a lot (hence not much duplication/overlap).
I don't have a problem with Jim buying expensive glass!
I'm even fine with Jim using it a lot!
I don't have a problem with buying expensive glass if I will use it a lot (hence not much duplication/overlap).
I don't have a problem with Jim buying expensive glass!
I'm even fine with Jim using it a lot!
Jan 18, 2014 at 06:34 PM
p.3 #18Lens tests, 17mm/18mm
ggreene wrote:
I don't know if I would say well in advance for flexibility when the 16-35 is a zoom and has AF. Those are two pretty damn big components to flexibility in my book and why I own it.
I completely agree, but Jim's comparison is really about lens IQ and amongst a small sub-set of focal lengths: 16-18mm. It's very rare to find a lot of choices in that area (another one that might be interesting to see is the Tokina 16-28). I would venture that it's a wash between TS and AF/zoom -- just depends on what you need. The zoom is one of the reasons I use the 16-35II fairly often as an interior lens.
Gunzorro wrote
Peter -- I agree. There is also the vignetting, which favors the 17TSE and Jim touched on it. Those four factors place the 17TSE well in advance of the 16-35II for flexibility. But, it is very encouraging that the zoom shows so well for its imaging: sharpness, contrast, color, etc. and most of the other minuses can be ameliorated with PP. Really a great showing, and a complete surprise to many readers who don't have either or both these lenses.I completely agree, but Jim's comparison is really about lens IQ and amongst a small sub-set of focal lengths: 16-18mm. It's very rare to find a lot of choices in that area (another one that might be interesting to see is the Tokina 16-28). I would venture that it's a wash between TS and AF/zoom -- just depends on what you need. The zoom is one of the reasons I use the 16-35II fairly often as an interior lens.
Jan 18, 2014 at 06:36 PM
p.3 #19Lens tests, 17mm/18mm
Thanks for the tests Jim.
It does not seem you are taking in consideration field curvature. The TS-E 17mm has a mild field curvature and therefore you need to focus on the center to evaluate center resolution and extreme edges to evaluate that area.
An easy field curvature test is to use Live View with the lens wide open.
A test chart or newspaper taped to a wall can work well. Focus at center, then move the focusing point towards the edges, observing image sharpness. Refocus off-center and determine if the image becomes sharper-if so, you're seeing field curvature (assuming you've aligned the camera squarely to the target)
Regarding the vignetting test. The exposures must be identical for this test to be valid. Also using auto-contrast may have altered the results a bit.
Best,
Fred
It does not seem you are taking in consideration field curvature. The TS-E 17mm has a mild field curvature and therefore you need to focus on the center to evaluate center resolution and extreme edges to evaluate that area.
An easy field curvature test is to use Live View with the lens wide open.
A test chart or newspaper taped to a wall can work well. Focus at center, then move the focusing point towards the edges, observing image sharpness. Refocus off-center and determine if the image becomes sharper-if so, you're seeing field curvature (assuming you've aligned the camera squarely to the target)
Regarding the vignetting test. The exposures must be identical for this test to be valid. Also using auto-contrast may have altered the results a bit.
Best,
Fred
Jan 19, 2014 at 08:41 AM
p.3 #20Lens tests, 17mm/18mm
Hi Fred,
Thanks for the comments. You're right on all counts; however, I prefer to evaluate sharpness from a single exposure, because that's what I get from a single exposure. I often do multiple, sequential exposures to vary EV, but not to vary point of focus, except sometimes for macro. If field curvature is a problem at f/8, then the lens probably won't stay in my bag for very long.
With respect to vignetting, most of the central portions of the images have the same brightness (which you can check in the Set 1 and Set 2 images), and so the comparison should generally be OK, except for those I noted in the comments.
Thanks,
Jim
Thanks for the comments. You're right on all counts; however, I prefer to evaluate sharpness from a single exposure, because that's what I get from a single exposure. I often do multiple, sequential exposures to vary EV, but not to vary point of focus, except sometimes for macro. If field curvature is a problem at f/8, then the lens probably won't stay in my bag for very long.
With respect to vignetting, most of the central portions of the images have the same brightness (which you can check in the Set 1 and Set 2 images), and so the comparison should generally be OK, except for those I noted in the comments.
Thanks,
Jim
Jan 19, 2014 at 09:39 AM