#Arkive KE: 20mm Lens comparison

Greetings All,

I've started a comparison of four 20mm lenses that I have. The lenses are:

1) Mamiya Sekor SX 21mm F4 (m42 mount)

2) Vivitar Auto Wide Angle 20mm F3.8 (m42 mount)

3) Yashica ML 21mm f2.8 (CY mount)

4) Canon nFD 20mm f2.8 (FD Mount)

Test lenses shown mounted on E-Mount Adapters

I have three test set-ups that I will be posting in this comparison:

1) Close Focus

2) Mid Range

3) Infinity Focus

All three set-ups are shot wide open, F8 and F11. The first posting is for close focus. Of the four lenses the Mamiya has the longest close focus at roughly 18" and the Vivitar the closest at around 3". The Canon and Yashica have comparable close focus. The crops are all at 100% to help illustrate the difference in focus distance as well as resolution.

Close Focus Test

I will post the rest of the tests this weekend

Nice selection, I'm following with interest.

Most 20mm lenses are reasonably sharp stopped down a little. What separates the men from the boys is flare. The Nikon 20mm f/3.5 AI was the best I ever owned in that regard. It could be shot straight into the sun without ghosts or veiling. Really remarkable.

Jack

Hello. It would be interesting to see a comparative test on the Sony A6000 camera.
"Test on banknotes from a distance of 2 meters"
One bill in the center of the frame, the other in the lower right corner.
Focusing on the tripod on the central bill. The shutter release timer or remote control.
One aperture of 5.6 is enough
Thank.

The matrix below compares the vignetting between the four lenses.  I used the base images from my mid range test setup. Of the four, the Mamiya vignettes less overall although it should be noted that it is also the slowest at F4.  The Canon vignettes quite a bit more than the other three lenses at all apertures.  To my eye, the ML is third worse wide open, but improves to second at subsequent apertures.

All images OOC JPEGS

The following matrices illustrate the mid-range focus characteristics of the four lenses.  The set up approximates a typical composition that I would shoot in the field.  In this case the subject is approximately 9' from the camera and the context trails away.  The fence is the subject in this set up.  All images were cropped at 100%.  Overall, I was surprised at the competitive performance of the Mamiya SX.  The Vivitar clearly lags behind the others in the mid and corner frame.  The ML performs very well in the center, mid and near corner.  The Canon FD performs very well in the center, mid and near corner.  It FD outperforms the ML in the far corner.

Greetings All,

Please find below the matrices comparing these lenses at infinity focus.  In this case, the infinity point is the horizon line of the opposite side of the valley, approximately 5 miles from the camera.  The matrix set up compares center point, mid-from and left and right corners.  Each crop is scaled at 100%.  The trees shown in the left corner are approximately 60' from the camera.  All other objects in frame are at least 100' from the camera.

I will write a conclusion for these studies in the next day or two.  Your comments are appreciated.

Camera Set Up - Horizon Line is Approximately 5 Miles Distant

Fully Open Aperture

F8

F11

filmluvr wrote:

Most 20mm lenses are reasonably sharp stopped down a little. What separates the men from the boys is flare. The Nikon 20mm f/3.5 AI was the best I ever owned in that regard. It could be shot straight into the sun without ghosts or veiling. Really remarkable.

Jack

I mostly agree.  I tried; Zeiss Flektogon, Canon FD S.S.C., Voigtlander 20 3.5, S-M-C Takumar 20 4.5, Pentax-M 20mm f4, Pentax K 20mm f4, and Pentax-A 20mm f2.8.  20mm is my most important lens for me, so I based my whole system around which 20 was the best.  The winner?  Pentax-A 20mm f2.8.

Reasons others failed:

Zeiss Flektogon had a lack of contrast, wasn't that sharp.  3d effect was non-existent.

Canon FD S.S.C. had winged coma issues when wide open with stars, effectively making it an otherwise gorgeously rendering 20.

Voigtlander had soft corners and vignetting.  Not quite enough glass to d othe job intended in my opinion.  Otherwise, I really liked the lens.

The Pentax's I am still using them all, but the Pentax-A 20mm f2.8 stands apart from the rest in general quality of look.  The floating element inside and most modern coatings, I think, are really helping in such a wide angle.   For overall quality of images, speed, and the whole lens.. The A series Pentax is the best in my opinion from what I have tried.

I don't doubt that the Nikon 20 3.5 is excellent.  I know it is.  Hmm.. And it focuses the "right" way for me...  mmm-hmm..

vyoufinder wrote:

filmluvr wrote:

Most 20mm lenses are reasonably sharp stopped down a little. What separates the men from the boys is flare. The Nikon 20mm f/3.5 AI was the best I ever owned in that regard. It could be shot straight into the sun without ghosts or veiling. Really remarkable.

Jack

I mostly agree. I tried; Zeiss Flektogon, Canon FD S.S.C., Voigtlander 20 3.5, S-M-C Takumar 20 4.5, Pentax-M 20mm f4, Pentax K 20mm f4, and Pentax-A 20mm f2.8. 20mm is my most important lens for me, so I based my whole system around which 20 was the best. The winner? Pentax-A 20mm f2.8.

Reasons others failed:

Zeiss Flektogon had a lack of contrast, wasn't that sharp. 3d effect was non-existent.

Canon FD S.S.C. had winged coma issues when wide open with stars, effectively making it an otherwise gorgeously rendering 20.

Voigtlander had soft corners and vignetting. Not quite enough glass to d othe job intended in my opinion. Otherwise, I really liked the lens.

The Pentax's I am still using them all, but the Pentax-A 20mm f2.8 stands apart from the rest in general quality of look. The floating element inside and most modern coatings, I think, are really helping in such a wide angle. For overall quality of images, speed, and the whole lens.. The A series Pentax is the best in my opinion from what I have tried.

I don't doubt that the Nikon 20 3.5 is excellent. I know it is. Hmm.. And it focuses the "right" way for me... mmm-hmm..

I had a Sigma 18mm years ago that was so bad, I couldn't even shoot under a couple 60W bulbs in my living room without significant veiling flare. I replaced it with the Nikkor 20mm, and that was the last third-party UW I ever bought.

Sorry if this sidetracks the OP's intent for this thread, but really, no test of an UW for sharpness or contrast is useful if the lens exhibits a propensity for flare. This is the first thing I test with any newly acquired UW, prime or zoom. If it fails, it's out the door immediately. Why waste time testing all the other characteristics if lens performance will be degraded by a real-world light source?

Jack

Does the Mamiya glass have a yellow coating? The images from that one are a different color.

Jack

filmluvr wrote:

Does the Mamiya glass have a yellow coating? The images from that one are a different color.

Jack

I have almost a full line of Mamiya SX lenses and they all render significantly warmer than my other lenses from the same era.  I'm sure that there is something in the Mamiya coatings that is different, but its outside of my understanding.

borissimo86 wrote:

Kevner wrote:

The matrix below compares the vignetting between the four lenses. I used the base images from my mid range test setup. Of the four, the Mamiya vignettes less overall although it should be noted that it is also the slowest at F4. The Canon vignettes quite a bit more than the other three lenses at all apertures. To my eye, the ML is third worse wide open, but improves to second at subsequent apertures.

All images OOC JPEGS

Why is the color rendition of the Mamiya so different from the rest?

Greetings Brossimo,  It is a common characteristic of all my Mamiya SX lenses which render images in a warmer tone than my other lenses.  There isn't any yellowing like you see with radioactive lenses.  I just think it has to do with the coatings Mamiya used or the design philosophy of the SX lens line.  To the eye, the lenses are clear.  I do have a few yellowed lenses for comparison.  I would be interested in learning more at some point.

I finally had a chance to bring this test to a conclusion (of sorts). Firstly, let me say that I have no illusions that these lenses would perform as well as a modern 20mm designed for use on the Sony filter stack. While these are all lenses that were designed over thirty years ago, I was pleasantly surprised by the results. Here is how the test comparisons were performed:

Close Focus/Resolution: This test is fairly self explanatory. I set up a typical close focus target (US $5 Bill) and established what the closest focus for each lens. Three shots for each lens, wide open, F8 and F11. These are typical aperture settings for the way I shoot. The set up probably affected the outcomes and I will use a standard distance setting for resolution rather than the close focus distance for each lens. The performance for the Mamiya SX 20mm also lends me to believe that the close focus distance for that lens was incorrect.

Vignetting: Vignetting was measured in the sky from the shots taken for the mid-range target for all three aperture settings. The wide open shots are hard compare as the four lenses do not share the same maximum aperture. As would be expected, the Mamiya SX performed best wide open. The performance at F8 and F11 is a truer comparison between the lenses.

Mid-Range Target: For the mid-range target I use a fence in my back yard. The camera is set up so that the center frame target is 10' from the camera and the fence oblique to the camera. Within the image frame, the fence in the left corner of the image is approximately 40' from the camera and the fence in the right corner of the image is approximately 6' from the camera. The image is compared in the center, mid-frame left, left and right corners. As the aperture is changed, each test point is compared using the target (fence) and similar points throughout the focal range.

Infinity Target: For the infinity target I used the horizon line on the opposite side of the valley I live in seen from my back yard. The camera is set up so that the center frame target is 5 miles from the camera and focused on the conical peak seen in the center frame.   The image is compared in the center, mid-frame left, left and right corners. As the aperture is changed, each test point is compared using the target (conical peak) and set target points throughout the focal range.

Attached below are the excel charts indicating how these lenses performed in comparison to each other.

Close Focus

Mid-Range

Infinity Target

This is good info. Unfortunately, it's not enough if your purpose is to help someone make a decision as to which lens is best for them. You need to take all these lenses outside on a sunny day and shoot an image with A) the sun outside the frame, to test for veiling flare and B) the sun in the corner of the frame, to test ghosting.

Jack

Greetings Jack,

Thanks you for the input.  I recognize that this is not a comprehensive comparison.  This is was something I did out of interest in how these lenses compared to shooting situations similar to how I most commonly compose the landscape.  Given where I live and my interests, I rarely compose images with the sun where flare would be considered an issue.  I do understand how that would be centrally important to someone whose shooting style uses the sun in that manner.  The one test that I wish I had done was distortion as architectural subjects is one of my principal interests.

This is the third comparison I have done and the one that within its limited scope is the most rigorous.  I intend to revise the 24mm comparison I did a few weeks back to this format and redo the 28mm comparison I did a while ago.  Looking ahead, I also intend to do a 35mm and 50/55mm comparison.

To be honest, I do these because work has been quite busy lately and don't have much time to actually go out and spend time in the field.  These are pretty quick to set up and I can work on them for a couple weeks and, it's something I can share.  Its also a way for me to get a feel for the lenses I end up collecting.

Best,

Kevin