uhoh7, have you seen any color shift on the CV 35 1.4? I realize it's not hard to fix, but I would much prefer to find a small 35 that doesn't exhibit it. Thanks.
turnstyle wrote:
Are you referring to what I posted just above?
It doesn't seem to be about the lens -- but rather the different behavior of A7 vs A7r, no?
And what's with that crazy arc in the A7r photos?
The A7r was overexposed by at least 1.5 stops, probably even more. Look at the yellow walls, pushed into pure blown white on the -R but close to 2 Ev down on the A7. User errors should not be blamed on lens/camera.
I do see a larger shift towards the absolute corner though, but you'd do best to ignore the rest of the image.
Gary Clennan wrote:
That color shift is pretty gross... It is a shame though as that lens is such a great performer.
uhoh7 wrote:
Color shift is meaningless.
Why, because it can be (with varying degrees of effort/success) correctable in post?
By that logic ... vignetting, CA, distortion, color cast, clarity, bokeh, micro-contrast, etc. that are also aspects of the optically projected image's drawing style are meaningless as well, since they too can be manipulated/corrected in post.
Jeff Kott wrote:
Maybe to you, but some people don't want to deal with it.
agree. I sold my NEX-7 right after I noticed it. I was hoping it wouldn't be an issue,
but I just hated the extra steps to correct it and it never really looked good anyway...
p.130 #10 · A7/A7r - performance with WA RF lenses
So my a7r showed up right before sunset, only had the chance to shoot a couple images off the back deck with a 18 zm and the cv 28/2.
Voigtlander 28/2 F2 shows very very strong vignetting at close and far distances. A lot of smearing in corners F4 starts getting pretty good at all distances for vignetting, but still strong smearing, particularly in further distances . Outstanding central sharpness at all apertures, stopped down to f8 it seems pretty good (better than expected for me at least) across the field. From initial (seriously, 5 minutes) run I think it will be a quite decent all around lens for close-mid distances, people, and when needed stopped way down on a tripod for landscape. Far better on the a7r than an nex7
Zeiss ZM 18. Sigh. Strong color and vignetting at all apertures. Outstanding sharpness over center 2/3 of frame at all apertures, with the area sharpness growing as you stop down. Corners get pretty good at f/8, f/11 very good but not critically sharp. With a slight crop on a tripod I think this will be a quite good ~21mm equiv landscape lens.
Sorry, just initial thoughts, no photos. Ill post some tomorrow. Have a whole lot of other m mount glass (cv 35/1.2 II, Leica 24/3.8, Leica 35/2 asph, etc etc) to run through it. Also will shoot with my ZF.2 21/2.8 and 25/2.8 to see what will stay and what will go...
p.130 #11 · A7/A7r - performance with WA RF lenses
theSuede wrote:
The A7r was overexposed by at least 1.5 stops, probably even more. Look at the yellow walls, pushed into pure blown white on the -R but close to 2 Ev down on the A7. User errors should not be blamed on lens/camera.
I do see a larger shift towards the absolute corner though, but you'd do best to ignore the rest of the image.
I see that the link I posted stopped working, I think this will:
So not only is the color difference emphasized, there seems to be some odd arc in the A7r photos -- do you see what I mean? It should be pretty obvious. I find it very curious!
So not only is the color difference emphasized, there seems to be some odd arc in the A7r photos -- do you see what I mean? It should be pretty obvious. I find it very curious!
I have seen that posterized effect in a few A7r + adapted wide rangefinder shots, particularly when pushed to extremes as here. With extreme processing, that effect has also shown up with other Sony lossy raw images and I have wondered if it's at least partially due to the lossy compression going on. There is a lot of data being thrown out with the Sony lossy raws (non lossy sony raws from 24MP's are the same size as the current lossy 36MP raws for instance). Sony claims it's not visible (the difference between Sony lossy vs non-lossy raws) but perhaps in extreme processing, it shows up where it might not with non lossy raws.
p.130 #13 · A7/A7r - performance with WA RF lenses
Tariq Gibran wrote:
I have seen that posterized effect in a few A7r + adapted wide rangefinder shots, particularly when pushed to extremes as here. With extreme processing, that effect has also shown up with other Sony lossy raw images and I have wondered if it's at least partially due to the lossy compression going on. There is a lot of data being thrown out with the Sony lossy raws (non lossy sony raws from 24MP's are the same size as the current lossy 36MP raws for instance). Sony claims it's not visible but perhaps in extreme processing, it shows up where it might not with non lossy raws....Show more →
Thanks, I feel less crazy. But just a tiny bit less crazy.
But this leads me to wonder:
1) why don't I see a similar arc on the A7 shots? Sure, it might be a different hue, but I'd expect to see it on both sides.
2) doesn't this mean there is "less to recover" in the A7r shots, so to speak?
p.130 #14 · A7/A7r - performance with WA RF lenses
turnstyle wrote:
The left side is the A7, the right is the A7r -- they don't seem so terribly differently exposed to me.
I see the yellow (shadow area) and the satellite dish both about 10-13 points difference between the A7 vs. A7R. Some diff, but not radically overexposed. Looking at the A7R "blown" area it is only 229,229,229, so not really blown, but very neutral compared to the other @ A7 with a much lower blue channel in the A7 @ 235, 221, 192 shifting us from warm to neutral.
The same variance can be seen on the left side buildings so the two (A7 vs. A7R) seem to be at different WB's (or responding differently to a given WB setting with a particular sensitivity diff @ the blue channel).
p.130 #15 · A7/A7r - performance with WA RF lenses
turnstyle wrote:
Thanks, I feel less crazy. But just a tiny bit less crazy.
But this leads me to wonder:
1) why don't I see a similar arc on the A7 shots? Sure, it might be a different hue, but I'd expect to see it on both sides.
2) doesn't this mean there is "less to recover" in the A7r shots, so to speak?
The A7 plays better with rangefinder wides compared to the A7r so you will see more dramatic vignetting and color shifts with the A7r. That very well could be all we are seeing with over exposure in the center (meter compensating for the more extreme vignetting on the A7r)...but then there is definitely that arc of posterized color on the A7r. If it has nothing to do with the image processing (that's just a wild guess on my part regarding lossy raw), perhaps it's just an effect created by the major incompatibility between the lens and sensor on the A7r. If that's the reason, yes, it would be less recoverable using post processing with the A7r in this instance.
Edit, well Kent measured the blown area so over exposure may not be playing a role as much as simple lens/ sensor incompatibility.
Edit 2.. I do believe that the original A7r exposure was overexposed in the center as I originally hypothesized. Even though those ares do not measure close to 255, there is no detail in that area that is recoverable. That suggest that highlight recovery was likely used on the image initially which brought it's numerical value down but was unable to really do any recovery of detail as it was so blown out to begin with. We would need to know what editing was performed on the original file to confirm but that's my guess.
p.130 #16 · A7/A7r - performance with WA RF lenses
Looking at the cranked color versions, we can see the purple increase in the corners as well as the arc. I can't help but wonder if there is some AWB/software correction overcompensating to make the highlight areas assumed to be neutral, when they really aren't.
p.130 #18 · A7/A7r - performance with WA RF lenses
RustyBug wrote:
The same variance can be seen on the left side buildings so the two (A7 vs. A7R) seem to be at different WB's (or responding differently to a given WB setting with a particular sensitivity diff @ the blue channel).
Yes, so assuming slightly different WBs and slightly different exposures...
If I crank the color, it makes sense that we see an exaggerated difference in color -- so I'm not even griping about magenta here (though clearly that happens to be the direction the A7r takes).
What has me baffled is the "monochomatic rainbow" in all of the cranked A7r photos -- I don't get why I wouldn't see a similar monochomatic rainbow in the A7 photos -- it seems to me as if there is some sort of substantive difference in the underlying images, like less ability to recover, in the A7r side.
