[mst]

The border has objects over a large depth in the lenstip photo.... So where do you propose the focus in the border actually is?

"depth" on the left border maybe, but certainly not on the right. To me, the focus clearly seems to be in the image center, so the wall of the building pointing towards the photographer. That means that none of the corners and nothing on the right side of the image is near the focal plane. On the left, maybe, yes. Judging from that image, I'd assume a curved image field bent backwrads towards the corners.

Same for the second image you posted.

And if field curvature is going on, that should be reflected in the Sony MTF too?

Not necessarily, no. It's the maximum resolution the lens can deliver at a given spot, which requires it to be in focus (in that spot).

[thxbb12]

Exactly, those MTFs from Sony are a bit silly.
Wide open:
https://pliki.optyczne.pl/son14GM/son14_fot01.JPG
The borders are not that great (CA issues, astigmatism pulling things "apart", while you would not have guessed that looking at those OOTW MTFs.
No doubt a very good 14mm lens, but those MTFs are meaningless.

That photo from lenstip is not representative of border sharpness. It's clearly not focus in the right spot or there is some field curvature going on.

In comparison, look at these ones at f1.8, these are sharp: here

The border has objects over a large depth in the lenstip photo.... So where do you propose the focus in the border actually is?

And if field curvature is going on, that should be reflected in the Sony MTF too? The link you post does not work, but I see dpreview in the URL so my guess is that you link to post processed images, where the lenstip image is a plain conversion (as always) with no corrections and sharpening applied?
Another lenstip wide open sample, again with lesser borders than those MTFs would suggest.
https://pliki.optyczne.pl/son14GM/son14_fot13.JPG

Clearly, if in the photo you mention, the focus had been on the corners, they would be sharp enough.
The results you observe are related to where it was focused.

I'm still waiting for your explanation as to how the borders here (link) can be sharp (especially the bottom right).
You seem to conveniently ignore the photo I mention.

[Klaus]

Taken at f/1.8 - https://www.dpreview.com/sample-galleries/2673848171/sony-fe-14mm-f1-8-gm-sample-gallery/0556364263

looks plenty sharp in the corners.

[thxbb12]

Taken at f/1.8 - https://www.dpreview.com/sample-galleries/2673848171/sony-fe-14mm-f1-8-gm-sample-gallery/0556364263

looks plenty sharp in the corners.

Eheh, the same photo I keep mentioning over and over but people seem to ignore it somehow ;-)

[Brightcolours]

Taken at f/1.8 - https://www.dpreview.com/sample-galleries/2673848171/sony-fe-14mm-f1-8-gm-sample-gallery/0556364263

looks plenty sharp in the corners.

Eheh, the same photo I keep mentioning over and over but people seem to ignore it somehow ;-)

Maybe that is because the link you post does not actually work. It gives an XML file from the server.

Somehow you ignore that DPReview posts post processed images, and Lenstip does not? ;-)

---

Clearly, if in the photo you mention, the focus had been on the corners, they would be sharp enough.
The results you observe are related to where it was focused.

I'm still waiting for your explanation as to how the borders here (link) can be sharp (especially the bottom right).
You seem to conveniently ignore the photo I mention.

The photo is focussed on the building, which is not in the corners. Subjects on the borders (left/right) are in the same distance ballpark as the building, and not as sharp as the MTFs promise.

[mst]

Somehow you ignore that DPReview posts post processed images, and Lenstip does not? ;-)

That's not exactly a fair description of the different approaches DPR used the default sharpening parameters in ACR, while lenstip for ages claims that it's the best approach to not apply any sharpening during RAW-conversion at all... which is something no one would ever seriously consider in real life, and their sample images regularly show why

In any way: the different approaches definitely do not explain the very visible differences between the sample images. The DPR sample is obviously not focused on the objects in the image center, but maybe (manually?) focused to give good corners (the corners look better than the image center to me). The lens tip samples a focused in the image center obviously, with the corners not in the same distance and thus blurry.

The dpr sample image shows that the lens can deliver good corner sharpness when focused properly. The lenstip samples linked to in this thread actually tell nothing about the lens' potential in the corners.

[thxbb12]

The dpr sample image shows that the lens can deliver good corner sharpness when focused properly. The lenstip samples linked to in this thread actually tell nothing about the lens' potential in the corners.

Exactly my point, thanks Markus :-)

[borisbg]

We can split heirs all day long, but truth is Sony did a great job. And look the price point...
Speculation: if CaNikon would release such lens it will be $2,000 +. I doubt it will do much better on the test bench, let alone size and weight.

[thxbb12]

Maybe that is because the link you post does not actually work. It gives an XML file from the server.

Somehow you ignore that DPReview posts post processed images, and Lenstip does not? ;-)

The photo is focussed on the building, which is not in the corners. Subjects on the borders (left/right) are in the same distance ballpark as the building, and not as sharp as the MTFs promise.

The link works just fine.

Again, it's got nothing to do with sharpening and everything to do with focusing.

---

We can split heirs all day long, but truth is Sony did a great job. And look the price point...
Speculation: if CaNikon would release such lens it will be $2,000 +. I doubt it will do much better on the test bench, let alone size and weight.

Indeed, the lens is impressively compact and light for what it is and its performance level.

[Brightcolours]

Maybe that is because the link you post does not actually work. It gives an XML file from the server.

Somehow you ignore that DPReview posts post processed images, and Lenstip does not? ;-)

The photo is focussed on the building, which is not in the corners. Subjects on the borders (left/right) are in the same distance ballpark as the building, and not as sharp as the MTFs promise.

The link works just fine.

Again, it's got nothing to do with sharpening and everything to do with focusing.

---

We can split heirs all day long, but truth is Sony did a great job. And look the price point...
Speculation: if CaNikon would release such lens it will be $2,000 +. I doubt it will do much better on the test bench, let alone size and weight.

Indeed, the lens is impressively compact and light for what it is and its performance level.

Klaus' link works, yours does not. Close your dpreview tabs and try your link, it does not work.

---

The dpr sample image shows that the lens can deliver good corner sharpness when focused properly. The lenstip samples linked to in this thread actually tell nothing about the lens' potential in the corners.

Exactly my point, thanks Markus :-)

No one complained about the lens being good, no idea why you have that impression. We just complained about the optimistic Sony MTF calculations.