p.20 #1 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Fred Miranda wrote:
Received a bad copy (likely element tilt) and requested a replacement. I prefer not posting samples and wait for an optimal copy to compare with my other lenses.
I may have a few minutes this afternoon to do a formal centering test on my copy. My night shots were not all that instructive due to the long shutter times.
p.20 #2 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Holger wrote:
I have done a comparison, but not rigorously. My first impulse when getting the lens on Saturday was to check for centering
etc. I did some controlled tests indoors and outdoors and the lens seems fine.
When I have more time this week-end, I will repeat some tests. However, our A7riii will be replaced with the A7riv next week, so I would need to redo e.th. again.
p.20 #3 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
ottertack wrote:
"The bite", is it because of lack of CA or SA? It can be a tradeoff between sharpness vs bokeh smoothness.
Both lenses are great for astro. It depends on the intended use. For printing large or >100% viewing, batwing matters. Stitching is definitely a possibility, but a headache because of star trails (and accuracy of stars).
Do you feel the G20 AF is faster than GM24 (regardless of the motors)? And have you decided GM24's fate?
Yes, lenses that have little/no CA and SA make for harsher Bokeh but more bite. Which side of the trade-off you prefer is up to each shooter. I like softer less corrected optics generally.
As for which to keep, if I were a 24 FL shooter, GM wins hands down. I easily like the rendering better on the GM 24, and I think it’s one of the better modern lenses released in the last five years from anyone. It has that certain, don’t flame me-lol, 3-D quality to it and a richness of color where files are just gorgeous even beyond the testing block.
The new G 20 strikes me as a highly competent lens but lacks the top tier rendering quality. The files feel a little bit more like a Sigma or Nikon. That’s not bad, just my observation. A lot of people will appreciate and like it, I don’t think it will go down as having the magic of the GM 24.
I’ll probably sell the GM 24, not too decided yet. Tamron 17-28 is for sure gone.
And yes, I think the AF is faster on the G20, but other shooters might be better to ask that. I shoot manual focus 90% of the time, and have sold my A7r3 and A73 whee I would really see it.
p.20 #4 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Would love to see this lens compared to the Batis 18. I love my Batis 18, but a smaller lens with the same focal length (?) and over a stop faster would have my attention, as long as the rendering is as good as the Zeiss.
p.20 #5 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
RSSmith wrote:
When testing for centering, is it best to use a larger vs small aperture?
Largest possible. Increased DOF can hide decentering. I have an FE 28 that’s perfect centered, but only at F4 and up. Shoot at infinite, diagonally, and check the far corners.
p.20 #6 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Can anyone explain the optic differential cause between spherical aberrations and chromatic aberration?
I know chromatic abrasions are caused by misalignment of wavelengths not hitting the sensor the same time which results in color fringing/CA , both axial and lateral, purple and green. APO lenses address this.
I know that’s for spherical aberrations everything is veiled in a blanket of softness, and that there is an inverse relationship with contrast/clarity and SA. I also know that moving elements ever so slightly can increase or decrease spherical aberrations, but what actually causes SA?
In my experience, SA and CA seem to be linked in that we don’t get one without the other.
p.20 #7 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
nehemiahphoto wrote:
Largest possible. Increased DOF can hide decentering. I have an FE 28 that’s perfect centered, but only at F4 and up. Shoot at infinite, diagonally, and check the far corners.
Thanks! I'll give that a try when I get a chance to shoot outside.
nehemiahphoto wrote:
Yes, lenses that have little/no CA and SA make for harsher Bokeh but more bite. Which side of the trade-off you prefer is up to each shooter. I like softer less corrected optics generally.
As for which to keep, if I were a 24 FL shooter, GM wins hands down. I easily like the rendering better on the GM 24, and I think it’s one of the better modern lenses released in the last five years from anyone. It has that certain, don’t flame me-lol, 3-D quality to it and a richness of color where files are just gorgeous even beyond the testing block.
The new G 20 strikes me as a highly competent lens but lacks the top tier rendering quality. The files feel a little bit more like a Sigma or Nikon. That’s not bad, just my observation. A lot of people will appreciate and like it, I don’t think it will go down as having the magic of the GM 24.
I’ll probably sell the GM 24, not too decided yet. Tamron 17-28 is for sure gone.
And yes, I think the AF is faster on the G20, but other shooters might be better to ask that. I shoot manual focus 90% of the time, and have sold my A7r3 and A73 whee I would really see it.
You summarized it pretty well. ‘Sigma Art’-like. The considerations are 20mm vs rendering character.
I think I’ll stick to my GM24. I can stitch pano, sans aurora (which is rare here), if I need wider view. Sometimes I do travel portraits; GM24 bokeh never fails me. Might not say the same with double structure-like bokeh of G20.
---------------------------------------------
nehemiahphoto wrote:
Can anyone explain the optic differential cause between spherical aberrations and chromatic aberration?
I know chromatic abrasions are caused by misalignment of wavelengths not hitting the sensor the same time which results in color fringing/CA , both axial and lateral, purple and green. APO lenses address this.
I know that’s for spherical aberrations everything is veiled in a blanket of softness, and that there is an inverse relationship with contrast/sharpness and SA. I also know that moving elements ever so slightly can increase or decrease spherical aberrations, but what actually causes SA?
In my experience, SA and CA seem to be linked in that we don’t get one without the other. ...Show more →
p.20 #9 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
nehemiahphoto wrote:
Can anyone explain the optic differential cause between spherical aberrations and chromatic aberration?
I know chromatic abrasions are caused by misalignment of wavelengths not hitting the sensor the same time which results in color fringing/CA , both axial and lateral, purple and green. APO lenses address this.
I know that’s for spherical aberrations everything is veiled in a blanket of softness, and that there is an inverse relationship with contrast/clarity and SA. I also know that moving elements ever so slightly can increase or decrease spherical aberrations, but what actually causes SA?
In my experience, SA and CA seem to be linked in that we don’t get one without the other. ...Show more →
Spherical aberration happens when rays come through the periphery of the lens focus at a different point than rays coming through the center of the lens. That's why aspherical elements are so effective at correcting it. They change the focus point of those outer rays.
With spherical aberration, you often have the point of highest low frequency contrast being at a different focus position than the point of highest resolution.
p.20 #10 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
This is quite a good lens, a keeper for me since it gaps well with my 40mm and 28mm Arts, and takes up very little extra space in the wheelbarrow when you're carrying them around.
The corners cannot reasonably be criticized at f/4, although they may improve a bit closer to f/5.6. They are noticeably weaker wide open, although amply good for most purposes even on the R4. It is not going to be a perfect astro lens wide open, but maybe pretty good.
The distortion situation is a bit puzzling. There is abrupt pincushion distortion in the corners, and applying enough generic correction in either LR or C1 20 to straighten the short edges of the image leads to some barrel distortion along the long edges. There isn't a lot of distortion: the best compromise setting in C1 is "1" (fractions are not accepted, so this is the minimum correction), and around "-4" in LR. Toggling "disable default lens correction" in C1 seems to have no effect on the image, as if there is no built-in distortion correction in the raw file. One wonders if Sony shipped the lens that way. Maybe I'm misunderstanding something? Maybe Sony just hasn't gotten around to it?
I measured the vertical field of view at a distance of 6+ meters, and got an implied focal length of ~19.5mm, which should be correct to several tenths of a mm, so not far from the nominal FL.
It is discouraging that Fred got a crook copy: I'd thought Sony was really upping their QC game, and my copy seems fine.
p.20 #11 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
MAubrey wrote:
Spherical aberration happens when rays come through the periphery of the lens focus at a different point than rays coming through the center of the lens. That's why aspherical elements are so effective at correcting it. They change the focus point of those outer rays.
With spherical aberration, you often have the point of highest low frequency contrast being at a different focus position than the point of highest resolution.
p.20 #14 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Fred Miranda wrote:
Received a bad copy (likely element tilt) and requested a replacement. I prefer not posting samples and wait for an optimal copy to compare with my other lenses.
Was it way out of whack optically or just enough to make it noticeable during your tests?
Took my centering test images before dinner here in Seattle and need to look at them in ACR. 100% viewing in the viewfinder of the corners did not reveal any obvious difference in the 4 corners wide open focused at a mile away distance.
p.20 #19 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
AZSteve wrote:
The distortion situation is a bit puzzling. There is abrupt pincushion distortion in the corners, and applying enough generic correction in either LR or C1 20 to straighten the short edges of the image leads to some barrel distortion along the long edges. There isn't a lot of distortion: the best compromise setting in C1 is "1" (fractions are not accepted, so this is the minimum correction), and around "-4" in LR. Toggling "disable default lens correction" in C1 seems to have no effect on the image, as if there is no built-in distortion correction in the raw file. One wonders if Sony shipped the lens that way. Maybe I'm misunderstanding something? Maybe Sony just hasn't gotten around to it?...Show more →
Thank you for your observations. Just one thing with C1: I thought that to enable distortion correction one needs to dial the Lens Distortion dial to 100 ( or whatever amount one desires) irrespective of what lensprofile is applied. Is that not the case in C1 version 20?
p.20 #20 · Back On Sale: Sony FE 20mm f/1.8 G lens for only $798 ($100 off)
Just took delivery on my Sony 20mm F1.8. Tested it against a brick wall. Not de-centered, sharp in the corners. Some vingetting at F1.8, but gone by 2.8. Still waiting to test it in the field, particulary for coma in astro photography, but for now a keeper.
