[Christine Tham]

here's a full-size example of the fdn 24, look at the left side, those houses across the lake are fairly clear:
http://www.devine.us/img/s9/v92/p240976313.jpg

Wow that is a fantastic picture - excellent micro contrast on the house and as you say good sharpness right throughout.

[Digitalis]

Wow that is a fantastic picture - excellent micro contrast on the house and as you say good sharpness right throughout.

I took a look at the sample image and I disagree, the lens is stopped down way too much(f/11) to give any meaningful assessment of sharpness because of diffraction blurring, micro-contrast is also lacking as well. Most SLR lenses perform at their best two stops down from their widest apertures, most RF lenses perform best at one stop down.

[osv]

I took a look at the sample image and I disagree, the lens is stopped down way too much(f/11) to give any meaningful assessment of sharpness because of diffraction blurring, micro-contrast is also lacking as well. Most SLR lenses perform at their best two stops down from their widest apertures, most RF lenses perform best at one stop down.

the fdn 24mm 2.8 lens has it's highest resolution at f8, and it doesn't lose much by f11: Canon FDn 24mm f/2.8 on Sony A7 : ERPhotoReview

however, i've compared apertures with this lens, and with extreme pixel peeping, the slight loss of resolution in the center is indeed noticeable at f11... i've been trying to shoot it at f10 or wider for landscapes.

i agree with christine about the quality of the micro-contrast, but it's really a rather meaningless point to bicker over, because it's so easy to change in photoshop.

with 36mp full-frame sensors and wide-angle glass, getting clean sides is a real challenge... you have to experience it to understand it, and even then, a lot of people still don't get it.

---------- Post added 03-18-2014 at 01:18 PM ----------

another a7r/fdn 24 2.8 pic, from the same guy:

http://www.devine.us/img/s5/v117/p311687025.jpg

he has more a7r shots at Devine | Photo

[Christine Tham]

the fdn 24mm 2.8 lens has it's highest resolution at f8, and it doesn't lose much by f11: Canon FDn 24mm f/2.8 on Sony A7 : ERPhotoReview
with 36mp full-frame sensors and wide-angle glass, getting clean sides is a real challenge... you have to experience it to understand it, and even then, a lot of people still don't get it.

I've read that Leica does special processing to get cleaner sides - but only if you use Leica lenses with correct 6-bit coding. Something the Sony doesn't do except for native lenses. Although there is the lens compensation app which can correct corner distortion, colour casting and vignetting which I have been told is rather good.

I understand why you needed to stop down to f11 - in general, micro contrast improves as you stop down wide angle lenses and you reduce smearing at the edges, but at some point diffraction takes over. f11 is not too bad, although I generally stop at f8.

PS - don't worry too much about Digitalis comments - I have him on my ignore list because I don't really like his attitude and his comments are often off the mark, I didn't see his posts but judging from your reply I didn't miss much.

[wombat2go]

I disagree, the lens is stopped down way too much(f/11) to give any meaningful assessment of sharpness because of diffraction blurring, micro-contrast is also lacking as well.

Hi Digital,

That image is just an ordinary 8 bit jpg in srgb space coming down here as 1551 by 1036 pix. / 7368 x 4920
Looks fairly good on the Eizo here.
Can you give more detail about how you assessed "microcontrast" on it and "diffraction blurring "?

[osv]

PS - don't worry too much about Digitalis comments - I have him on my ignore list because I don't really like his attitude and his comments are often off the mark, I didn't see his posts but judging from your reply I didn't miss much.

he's clearly off the mark here, lol

i like dissenting opinions, but only when they can be backed up.

we are going a bit off-topic from light leaks, but if anyone is interested, here is a 42mb dng file that i shot this morning... a7r with the fdn 24mm 2.8, at f10, it's a wide landscape type of shot.

it needs ca removal and sharpening, but i'm posting it like this so that any of you who want to investigate this combo in a raw file, can do it on your own platform... i converted the sony arw to dng with the free adobe converter, and i haven't even opened the dng to look at it.

https://www.dropbox.com/s/r7ztau6s023o2qy/DSC01127.dng

[wombat2go]

i haven't even opened the dng to look at it.

ufraw here can't decode it . But it decodes the .dng ' s from the K-01 OK

edit: I don't think it is a true dng because my ordinary jpg viewer can open it - if only the thumb maybe

[osv]

i downloaded it from the link, and it opened right up in photoshop cs5, which won't open up the native sony arw files.

another option might be: RawTherapee Blog

i've never used it, but i understand that it's free?

i could also upload the sony file, but i don't know if that would help?

i just looked up that ufraw, it's been nearly a year since it's been updated?

[Christine Tham]

https://www.dropbox.com/s/r7ztau6s023o2qy/DSC01127.dng

When I tried downloading it, it only gave me a 25KB file, so something is wrong somewhere.

[Digitalis]

Has sony acknowledged this issue and offered a solution to it for those who have already bought the affected cameras?

I know this thread has been a bit de-railed enough already but here goes.

i converted the sony arw to dng with the free adobe converter, and i haven't even opened the dng to look at it.

The converter did a good job, the DNG looks fine, reviewing the image on my Eizo monitor (calibrated this morning) optically the lens itself reminds me of the SMC-K 24mm f/2.8, though the macro and micro contrast is lower*. Again, sharpness is difficult to accurately asses with the lens stopped down so much. There is a possibility your lens may be a bit de-centred as the astigmatism isn't equal on the edges of the frame - there is more of it on the right side of the frame than the left. There is a very slight magenta colour cast in the extreme corners of the image along with Blue/Red CA and PF being apparent, but with the correction tools in Lightroom 5.3.6 these issues can be cleaned up nicely.

Looks fairly good on the Eizo here. Can you give more detail about how you assessed "microcontrast" on it and "diffraction blurring "?

in general, micro contrast improves as you stop down wide angle lenses...

To clarify : micro contrast is one of the first things to be affected by even slight diffraction blurring - Micro contrast is defined as the ability to resolve close boundaries between similar colours clearly - diffraction blurring can severely reduce a lenses ability to render these micro scale differences, and if diffraction is high enough even macro contrast can be affected. Micro contrast cannot be photoshopped into an image, only macro contrast can be affected by common sharpening tools. Deconvolution sharpening is a promising method of enhancing micro contrast but it is computationally demanding as the image output size increases.


100% un-sharpened crop from the central portion of the image osv provided.

If this lens had a high degree of micro contrast, the power lines would have greater clarity in the way they were resolved against the blue ocean. Diffraction has reduced the ability of the lens to resolve of such fine levels of contrast.

Here is a rather extreme example of diffraction:


100% un-sharpened crop from Pentax K5IIs sigma 8-16mm f/4.5-5.6. @ 16mm

don't worry too much about Digitalis comments - I have him on my ignore list because I don't really like his attitude and his comments are often off the mark, I didn't see his posts but judging from your reply I didn't miss much.

Communities don't grow from ignorance, you of all people should know better Christine Tham. I don't have you on my ignore list.

*though to be completely honest, the SMC-K 24mm f/2.8 isn't exactly what I would call the greatest 24mm lens ever. It has a ridiculously short focus throw for a wide angle lens and you really have to stop this lens down to get the corners to look good even on APS-C cameras.

[pinholecam]

the fdn 24mm 2.8 lens has it's highest resolution at f8, and it doesn't lose much by f11: Canon FDn 24mm f/2.8 on Sony A7 : ERPhotoReview

however, i've compared apertures with this lens, and with extreme pixel peeping, the slight loss of resolution in the center is indeed noticeable at f11... i've been trying to shoot it at f10 or wider for landscapes.

i agree with christine about the quality of the micro-contrast, but it's really a rather meaningless point to bicker over, because it's so easy to change in photoshop.

with 36mp full-frame sensors and wide-angle glass, getting clean sides is a real challenge... you have to experience it to understand it, and even then, a lot of people still don't get it.

---------- Post added 03-18-2014 at 01:18 PM ----------

another a7r/fdn 24 2.8 pic, from the same guy:

http://www.devine.us/img/s5/v117/p311687025.jpg

he has more a7r shots at Devine | Photo

O/p seems quite similar to K24/2.8
Lens seems to be sharpest in the center and falls off in sharpness a little immediately after that, holds up pretty well thereafter until the final 1/9 of the frame (still ok, just not as sharp)
I am surprised that some areas seem to exhibit 'nervousness' (eg. clock tower), when it should just be 'stark sharp' since its not at the edges or corners, esp when its a downsampled A7r o/p.
I agree a bit with what Digitalis says about microcontrast.
I would have expected the stone texture of the buildings to jump out at me as its a 36mp no AA-filter o/p (or maybe I'm expecting too much )

For the size of these 24/2.8 lenses, I think its a fair compromise.

[jogiba]

I would have expected the stone texture of the buildings to jump out at me as its a 36mp no AA-filter o/p (or maybe I'm expecting too much )

The 40mp Pentax 645D vs 36mp Sony A7r test looks like it shows more texture with the A7r.

[philbaum]

I suppose the bayonet mounts are kept with the shiny finish for appearance so the wear marks don't upset consumers.
However in my opinion black anodized will better attenuate any light that bounces along the microscopic gaps, due to slight warpage or uneven spring pressure. between the mount surfaces ,

Chrome plating has been used on numerous sliding surfaces over the years because it has a low coefficient of friction, also reduces galling actions on threads and reduces corrosion. People have come to recognize and associate chrome plating with quality products. Are there alternatives - sure, but it would be nice to test them before turning them loose on consumers, e.g. would thread wear products end up on the sensor and result in more numerous sensor cleanings such as the D600?

Plastic composites might be a good substitute for chrome plating and could also be used to reduce light leaks and reduce weight/cost. Also doesn't corrode :-)

By the way, how does the observed light leaks at 30sec and 25600 iso compare to the light leaking in from the OVF? would be interesting to know.

[osv]

O/p seems quite similar to K24/2.8
I would have expected the stone texture of the buildings to jump out at me as its a 36mp no AA-filter o/p (or maybe I'm expecting too much )
For the size of these 24/2.8 lenses, I think its a fair compromise.

i agree with that... at some level it's a bit disappointing to see with a $2,300 camera body atmospheric conditions were poor, due to haze and a fog bank next to the coast, but it's probably due more to pixel density limitations and 24mm wide angles.

putting it into perspective... 24mm lens on a full-frame sensor is equal to 15mm lens on a crop sensor?? i'd really like to see how those shots compare.

micro contrast is a fine gradation of tonal values, which can't be created from scratch in photoshop... however, there are lots of tricks with masking and plugins that can seriously improve the perception of micro contrast... what matters most to me is what i can see on prints, and from what i've seen, commonly available printing technologies are not able to reproduce the extremely fine tonal values that true micro contrast represents... so imho, microcontrast in a lens is way down the scale in importance... i'm a lot more concerned with flat fields, sharpness, resolution loss due to distortion correction, etc.

---------- Post added 03-19-2014 at 02:08 PM ----------

If this lens had a high degree of micro contrast, the power lines would have greater clarity in the way they were resolved against the blue ocean. Diffraction has reduced the ability of the lens to resolve of such fine levels of contrast.

you won't find the term "micro contrast" anywhere on imatest.com:
https://www.google.com/search?q=site%3Aimatest.com+microcontrast&oq=site%3Ai...+microcontrast

the holy grail for test charts is to shoot the edge of a razor blade, because there is no contrast gradation in the source... zero micro contrast gives the sharpest image acquisition results, and zero contrast gradation would give the sharpest looking prints, if it could be reproduced.

beyond that.. those 100% crops showed cranes of a similar size, but those similar sizes came from pictures that had vastly different frame sizes.

i'm guessing that it's probably a similar situation to when you swap a lens between an aps-c sensor and a full-frame sensor?

[Christine Tham]

micro contrast is a fine gradation of tonal values, which can't be created from scratch in photoshop...

On recent versions of Camera Raw, the "Clarity" control can be used to manipulate micro contrast - it was deliberately created for this purpose. I didn't realise it until Russell Preston Brown explained it to me.

Problem is - micro contrast manipulation is so rife and overused these days some of us may not even realise what natural contrast looks like. Russell tells me he is not afraid to push the control all the way to 90 or 95 in some images. That may be valid for artistic reasons, but I see a lot of photos these days where both macro and micro contrast are just too high to be natural - even in a lot of films.

I fell into the same trap myself - I used to routinely push Clarity to 25 or 30 on a lot of my photos (a lot of Lightroom presets has that sort of range included in the setting). Sometimes I have gone as high as 50. These days I consciously try and leave the control alone - sometimes I might take it to 15 but seldom more.

I liked your photo not because it "jumped at me" - I would have groaned if it did. I also liked your photo because it wasn't "perfect" - that told me you were honest. Sometimes people post photos claiming how good their camera/lens is and I can tell the image has been manipulated, because I can see tell tale signs of manipulation (there's a phenomenon called Gibb's Effect that gets emphasised after certain types of manipulation).

Don't worry too much about other people's opinions (including mine) - at the end of the day the only person who needs to be satisfied by your results are yourself. By all means manipulate, or don't manipulate, as long as you are happy.