The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Fun with the RX-1

ptomsu

Workshop Member
What? It's longitudinal chromatic aberration. If the colors all focused the same, the demosaicing wouldn't produce those results.

Meanwhile, more examples on Huff's site, this one shows some more CA under admittedly tough conditions

http://www.stevehuffphoto.com/wp-content/uploads/2012/12/web.jpg
Fully agree this is CA, although it is far less than lenses like the Nikkor or Sony/Zeiss 1.4/85 or the 1.4/35 Zeiss etc.

Easy to correct BTW, but a bit strange as normally one would not expect that from a 2/35 - at least my 2/35 M ASPH does not show this.

Maybe this lens is not as great as most expect?
 

Taylor Sherman

New member
Sigh. So, if the lens is focusing the colors perfectly, how could the color-mixing processing be consistently shifting the color one way for parts in front of the focus plane, and differently for parts behind it? It can't. It doesn't have that information and therefore there's no way for that to be possible.

I'm happy to be wrong about things. Because when I'm wrong, I've just learned something. But what have I learned so far? that you disagree with me, that the picture shows "nothing" despite having purple and green artifacts, that purple and green artifacts aren't "typical" of CA yet when I link a page that describes LoCA by demonstrating purple and green artifacts due to front- and back-focus you dismiss this without any reasoning as to why this is "the definition" yet I don't "recognize" it.

Are you trying to be helpful? because it's not coming through.

I didn't mean to say that Steve's picture showed LoCA, just CA in general.
 

ptomsu

Workshop Member
Taylor, I am glad you looked up the definition of CA. Now hopefully you can learn to recognize it. Steve's example shows Lateral or Transverse CA, not longitudinal as you claim (and your link shows the difference). And Matt's images shows nothing.

But I agree, more images.
I could not care less if either LCA or TCA is creating wrong colors, I would expect a specialized high grade lens which is optimized for a sensor and comes from Zeiss would do better. Or at least would be corrected in camera.

So something is wrong ....
 

douglasf13

New member
Then why a red and green shift? Not typical CA behavior, especially on such low contrast targets. Given the fact this is not a text book setup for copy work and placed in complex environment, to jump to longitudinal CA as a cause (and an unusual one where it is increasing with a loss of focus) is a bit of a jump, especially when it simply can be the influence of the surroundings.

Steve's images are showing lateral CA (but not longitudinal CA). And, as you pointed out, under very high-contrast conditions.

Sorry, but there has been no real evidence that this lens has such a serious CA problem.
Longitudinal CA is CA on different focus planes behind and in front of the focus point, and it is usually red and green. Steve Huff's example shows this, as the red CA on the left side of the pic is in front of the focus plane, and the green CA on the right side of the pic is behind the focus plane.

Longitudinal CA is pretty common in most fast-ish lenses to some degree, and most Sony fast lenses that I've owned exhibit it at least a little. The ZA 85/1.4 has the most longitudinal CA that I've ever seen in a lens, although LR4 makes this pretty easy to correct, for the most part.

That certainly looks like longitudinal CA on the title of the book pictured above.
 
Last edited:

ausemmao

New member
Fully agree this is CA, although it is far less than lenses like the Nikkor or Sony/Zeiss 1.4/85 or the 1.4/35 Zeiss etc.

Easy to correct BTW, but a bit strange as normally one would not expect that from a 2/35 - at least my 2/35 M ASPH does not show this.

Maybe this lens is not as great as most expect?
Dunno how well I'm remembering my physics and optics. But for there to be no LoCA in a fast lens, you'd either need lens material with a wavelength invariant refractive index (doesn't exist) or made almost entirely from ultra low dispersion (so reduced wavelength invariance) elements (hello $$$$$ and size), or lose the internal focusing capability so that an apochromatic design can be used and IIRC, that's goodbye f/2.

Then again you're determined to dislike this camera, so :shrug:
 

douglasf13

New member
Dunno how well I'm remembering my physics and optics. But for there to be no LoCA in a fast lens, you'd either need lens material with a wavelength invariant refractive index (doesn't exist) or made almost entirely from ultra low dispersion (so reduced wavelength invariance) elements (hello $$$$$ and size), or lose the internal focusing capability so that an apochromatic design can be used and IIRC, that's goodbye f/2.

Then again you're determined to dislike this camera, so :shrug:
Yep, and even APO designs don't necessarily correct all CA outside of the focus plane.

Either way, like you said, most fast-ish lenses have LOCA. It's the amount of LOCA that varies, and I can't tell if the LOCA in the above pic is above normal.
 

Shashin

Well-known member
Longitudinal CA is CA on different focus planes behind and in front of the focus point, and it is usually red and green. Steve Huff's example shows this, as the red CA on the left side of the pic is in front of the focus plane, and the green CA on the right side of the pic is behind the focus plane.

Longitudinal CA is pretty common in most fast-ish lenses to some degree, and most Sony fast lenses that I've owned exhibit it at least a little. The ZA 85/1.4 has the most longitudinal CA that I've ever seen in a lens, although LR4 makes this pretty easy to correct, for the most part.

That certainly looks like longitudinal CA on the title of the book pictured above.
First, Steve's example show lateral or transverse CA, not longitudinal CA--you are seeing the effect off axis. And longitudinal CA is blue to red--blue being the shortest wavelength to red, the longest. But I doubt Sony is using a simple lens. With a achromat, that has two wavelength corrected (blue/red), and so the classic signal is green fringing, but not red. An apochromat would correct for three wavelength and you would think that red and green would be in there somewhere. Sorry, there is nothing here in this single image to point to that conclusion that it is CA. It could simply be the influence of the environment.

BTW, software corrects for lateral CA not longitudinal CA. But the book image is not showing longitudinal CA because it simply does not look like it. The other problem is that CA tends to be optics specific and this is not a problem showing up in other images. And this is really low contrast.
 

douglasf13

New member
Nope. Longitudinal CA, which some call fringing or bokeh fringing, is magenta in front of the focus point and green behind it, and the image of the book above shows this. Scroll to the bottom of this link to see Photozone's test for it: Sony E 50mm f/1.8 OSS (SEL-50F18) - Review / Lens Test - Analysis

Lateral CA has easily fixable for a while, but longitudinal CA has become somewhat fixable only more recently with LR4's new defringe tool.
 

jonoslack

Active member
Ah Well
Interesting discussion . . I've seen it before (lots of times) the Walker Evans book at an angle is an obvious culprit (like Photozone's excellent tests).
I don't believe it's usually much of a problem in real life (unless you're taking pictures of text at 45 degrees).

. ... me ? I'd rather shoot with an M and a 35 FLE, but there's no accounting for taste! Added to which it really is a lot more expensive in the UK
 

Shashin

Well-known member
Nope. Longitudinal CA, which some call fringing or bokeh fringing, is magenta in front of the focus point and green behind it, and the image of the book above shows this. Scroll to the bottom of this link to see Photozone's test for it: Sony E 50mm f/1.8 OSS (SEL-50F18) - Review / Lens Test - Analysis

Lateral CA has easily fixable for a while, but longitudinal CA has become somewhat fixable only more recently with LR4's new defringe tool.
Thank you for that. It has been my first time with the effect.
 

douglasf13

New member
No problemo, Shashin. I can't say that seeing one example of it in the RX1 indicates whether the camera has a problem with it or not. It's pretty common, and I've never found it all that objectionable in use...except with that ZA 85 that I mentioned.

I just went back and looked at more pics up close with the ZA 85 on my A900, and it has LOCA at wide apertures like I've never seen before. My Sony 50/1.4 had it noticeably, too, but not like the ZA 85, where the LOCA bothered me quite often. Pics of my wife often look like she is surrounded in green/magenta, angelic halos with that lens. :) LR4's new corrections still don't fix it all that well. Really great lens, otherwise.

Can't wait to see more RX1 pics. Looks like a cool camera.
 
Can someone please refund the last 10 minutes of my life spent reading that CA discussion ;)

What is the consensus on JPEG detail/sharpness? Is it me or does it look less than impressive*

*comparing to X-PRO1 here.

Chad
 

douglasf13

New member
Can someone please refund the last 10 minutes of my life spent reading that CA discussion ;)

What is the consensus on JPEG detail/sharpness? Is it me or does it look less than impressive*

*comparing to X-PRO1 here.

Chad
I'm not sure if there's a consensus, but, if the jpeg engine is anything like prior Sony cameras, detail and sharpness often lags compared to the competitions, but color is usually very good.
 

jonoslack

Active member
Thank you for that. It has been my first time with the effect.
If you look at photozone lens reviews, there is hardly a lens which does not exhibit it - and often much worse than this.

I'd say that it's a serious non-issue!

all the best
 

Michiel Schierbeek

Well-known member
Can someone please refund the last 10 minutes of my life spent reading that CA discussion ;)

What is the consensus on JPEG detail/sharpness? Is it me or does it look less than impressive*

*comparing to X-PRO1 here.

Chad
Isn't it a bit early for consensus?
Tjeezz the camera hardly came out of the box.:)

I see a real sharp nice cat's head popping out, in the second photo, with a pleasant background Zeiss blur.

Let's give the camera a chance :watch:
Time for a RAW file to study first, I think

Michiel
 
I'm not sure if there's a consensus, but, if the jpeg engine is anything like prior Sony cameras, detail and sharpness often lags compared to the competitions, but color is usually very good.
That's been my experience as well. And as a NEX, Alpha and X-Pro shooter, I'm not trying to disparage the RX-1 or the Sony brand. This is an exciting camera and Sony should be rewarded for their bold move here.

I had one on pre-order and got the call that it was waiting for me at Precision but after falling hard for the X-Pro I can't justify the purchase.

Chad
 
Isn't it a bit early for consensus?
Tjeezz the camera hardly came out of the box.:)


Michiel
Baring a FW update, I wasn't aware that JPEGs got better with age Michiel ;)

Don't bother with a defensive reply, I'm jesting and of course we need to see more but the early indications here are not overly impressive to me personally (and I'm only talking about jpeg detail and sharpness - the lens renders nicely).

Best,
Chad
 
Top