mobydick

Lives in Germany Germany
Joined on Oct 16, 2003

Comments

Total: 42, showing: 1 – 20
« First‹ Previous123Next ›Last »

I just tried the interactive demo, and although I am really impressed that it somehow works at all, the results are far from being a challenge to any human artist...

Link | Posted on Nov 22, 2021 at 19:23 UTC as 38th comment | 2 replies

Thank you for the nice coverage of the AIV launch. One question: At 10 fps cont. shooting, does the viewfinder show real "live view" (and not the last taken picture)?

Link | Posted on Oct 22, 2021 at 11:46 UTC as 15th comment
On photo Door to red in the red challenge challenge (10 comments in total)
In reply to:

Wildbegonia: Congratulations on your win, great photo. I personnally like very much diagonal compositions, they are way more dynamic.

Really great photo, one can imagine a lot of different stories about the "lady in red".
I also wondered how it would look like in a non-diagonal composition, but you are right, verybiglebowski, most likely it would be more conventional und less "catchy".
I wonder in which cases in general the diagonal twist works well; I think one can easily overdo and overuse this type of composition.

Link | Posted on Oct 9, 2021 at 10:49 UTC
In reply to:

mobydick: Dear Roger, thanks a lot for your very insightful articles. I know that this article is already quite a few months online, but I hope that this is still a good place to ask two questions.

I have the impression that your field curvature graphs show the measurements along a horizontal line through the image plane. If this impression is correct, wouldn't it be helpful to generate measurements along the orthogonal vertical line as well? At least reg. tilt of field, because the tilt axis could have any orientation. Lenses with a slight tilt along the horizontal direction could show a more severe tilt along another direction.

And a second question: You have written in the past that decentering and tilted field are something different. I can follow your argument reg. the observed effects. But I wonder if the causes for both phenomena are different, i.e.: Are different types of lens alignment issues responsible for decentering vs. field tilt?

@RCicala: Thank you for the clear explanation. Very well understandable that this is too expensive. I have recently done a casual test of two Tamron SP 35/1.4, following your "plane field of grass" methodology to get an idea about the field curvature. Both lenses had roughly the same amount of tilt (btw, quite large for a prime lens in my view), but one mainly in the horizontal direction and the other mainly in the vertical direction.

Link | Posted on Oct 6, 2021 at 18:46 UTC
In reply to:

Foskito: "Neutrino Coating"... reminds me an old cartoon characters lol

I also stumbled about this wording. Complete marketing nonsense... Nikon's "nano coating" sounds very reasonable in comparison.

Link | Posted on Sep 30, 2021 at 10:19 UTC
In reply to:

mobydick: Dear Roger, thanks a lot for your very insightful articles. I know that this article is already quite a few months online, but I hope that this is still a good place to ask two questions.

I have the impression that your field curvature graphs show the measurements along a horizontal line through the image plane. If this impression is correct, wouldn't it be helpful to generate measurements along the orthogonal vertical line as well? At least reg. tilt of field, because the tilt axis could have any orientation. Lenses with a slight tilt along the horizontal direction could show a more severe tilt along another direction.

And a second question: You have written in the past that decentering and tilted field are something different. I can follow your argument reg. the observed effects. But I wonder if the causes for both phenomena are different, i.e.: Are different types of lens alignment issues responsible for decentering vs. field tilt?

Meanwhile I have also found the answer to my first question in one of your previous writings: https://www.lensrentals.com/blog/2016/11/fun-with-field-of-focus-ii-copy-to-copy-variation-and-lens-testing/ (in the Addendum at the end)

Anyhow, I think that looking only at the horizontal direction to detect tilted fields nearly always underestimates the real tilt in a lens. But very clearly, the horizontal direction is the most important one, more so than the image diagonals for most types of subjects, and the image vertical is less relevant because it is shorter.

Link | Posted on Sep 26, 2021 at 17:05 UTC
In reply to:

mobydick: Dear Roger, thanks a lot for your very insightful articles. I know that this article is already quite a few months online, but I hope that this is still a good place to ask two questions.

I have the impression that your field curvature graphs show the measurements along a horizontal line through the image plane. If this impression is correct, wouldn't it be helpful to generate measurements along the orthogonal vertical line as well? At least reg. tilt of field, because the tilt axis could have any orientation. Lenses with a slight tilt along the horizontal direction could show a more severe tilt along another direction.

And a second question: You have written in the past that decentering and tilted field are something different. I can follow your argument reg. the observed effects. But I wonder if the causes for both phenomena are different, i.e.: Are different types of lens alignment issues responsible for decentering vs. field tilt?

I think I found the answer to my second question in the following article by you: https://www.dpreview.com/opinion/4042117089/roger-cicala-why-i-dont-use-an-mtf-bench-to-test-my-own-lenses

Nevertheless, if you have time at some point in the future to reply to my first question this would still be appreciated.

Link | Posted on Sep 25, 2021 at 12:40 UTC

Dear Roger, thanks a lot for your very insightful articles. I know that this article is already quite a few months online, but I hope that this is still a good place to ask two questions.

I have the impression that your field curvature graphs show the measurements along a horizontal line through the image plane. If this impression is correct, wouldn't it be helpful to generate measurements along the orthogonal vertical line as well? At least reg. tilt of field, because the tilt axis could have any orientation. Lenses with a slight tilt along the horizontal direction could show a more severe tilt along another direction.

And a second question: You have written in the past that decentering and tilted field are something different. I can follow your argument reg. the observed effects. But I wonder if the causes for both phenomena are different, i.e.: Are different types of lens alignment issues responsible for decentering vs. field tilt?

Link | Posted on Sep 24, 2021 at 08:48 UTC as 3rd comment | 4 replies
In reply to:

Mutovkin: I experimented a while back with Topaz Gigapixel AI to "upscale" scanned photos and results are definitely not impressive - barely producing any additional detail. So artificially generating low res images from high res digital originals is not the same as say upconverting old photographs. That's the area I would be excited to see having more research done. It would be great to have a good collection of algorithms to process old photos and enhance them automatically removing dust, scratches, increasing resolution and optionally applying user guided colorization.

That's interesting. I also tried Gigapixel AI lately and was not impressed at all. The results were slightly better than upsizing with standard image processing algorithms, but not by much. I actually work in this area (deep learning for various AI tasks) and current academic papers on superresolution show way more impressive results than the actual output of Gigapixel AI (of course, the example images for advertising on their website look really good...).

Link | Posted on Sep 7, 2021 at 16:32 UTC
In reply to:

Ruy Penalva: Suddenly everybody in China is making lenses. We know that is not true. There is a big maker (state owned?) distributing lenses to these pretense makers. I'm able to quote here at least 10 makers without optic tradition in making lenses, adapters, filters etc. In China everything is possible from retro-engineering to cosmetic changes in appearance.

@Jon555: Thank you for the link to the synopsis software. Highly interesting, especially the references to "manufacturing tolerances" and AI even on the entry page.

Link | Posted on Sep 7, 2021 at 10:28 UTC
On article Home at last: a travelogue with the GFX 50S II (123 comments in total)

Beautiful pictures, thanks a lot for sharing them here, wrapped in this thoughtful story.

Link | Posted on Sep 6, 2021 at 15:52 UTC as 44th comment

Thank you, this was entertaining and informative. I kind of expected this result because - when keeping print size constant - the number of collected photons by the whole sensor decides mainly over noise (with the same sensor technology and processing pipeline). Since BSI sensors seem to use most of their surface to collect photons and not much for other stuff, the result can be well explained.

But I have to admit that maybe, maybe at the smaller print size the A7S had a slight edge over the A7R reg. color noise in your video. But definitely not regarding overall appearance and definitely not at the large print size.

Link | Posted on Aug 14, 2021 at 15:05 UTC as 162nd comment | 1 reply
On article Sample gallery: Nikon Z fc (89 comments in total)

Since one cannot expect any big surprises from the well-known sensor, I think this gallery is more about assessing the quality of the kit lens and maybe the jpeg engine. Anyhow, the results look really good!

Link | Posted on Aug 14, 2021 at 10:20 UTC as 5th comment | 1 reply

Really very nice pictures, I enjoyed watching them a lot.

One question: There are some pictures with long shutter speed like 1/25sec with no flash use mentioned. How can this work out given all the movement in the insects and plants (by slight wind, etc.)? When I shoot macros without flash I usually try to achieve really short shutter speeds.

Link | Posted on Aug 14, 2021 at 10:13 UTC as 6th comment

Good news! I hope for Nikon and all other camera manufacturers that this positive trend will continue...

Link | Posted on Aug 5, 2021 at 15:59 UTC as 49th comment

I wonder if anyone has experience in combining the Techart adapter with the Sony LA-EA3/4/5 adapter to use A mount lenses with autofocus and electronic aperture control on a Z body...

Link | Posted on Aug 3, 2021 at 11:18 UTC as 23rd comment | 1 reply
In reply to:

Kirnbichler: It's somewhat strange, that this adapter tries to sell every attached lens as the same Nikon AF-S 50mm F1.4 G.

Other adapters are well abled to communicate different lenses, as for instance, the "Fringer" adapter that adapts Canon EF lenses to Z mount communicates more information.

Canon's "EF 70-300mm f/4-5.6 IS II USM" is reported in Exif as "70-300mm f/4-5.6G",

Sigma's old "70-210/4-5.6 UC-II" is reported in Exif as "70-210mm /f4-5.6 G",

the but slightly newer "Sigma 28-135mm 1:3.8-5.6 Macro" is reported in Exif as "28-135mm f/4-5.6 G".

None of these reported names are the actual lens names, but both focal length and aperture are reported correctly (with a slight exception in the last case).

I wonder what Techart's EF to Z adapter would report in these cases.

I also wonder how IBIS in the Z cameras can work properly if the current focal length is not communicated correctly to the camera.

Link | Posted on Aug 2, 2021 at 15:55 UTC

This was a really highly interesting and entertaining read! I am looking forward to purchase your planned ebook at some point in the future...

Link | Posted on Jul 12, 2021 at 10:38 UTC as 24th comment | 1 reply

As a long-term Nikon F-mount user, I considered to buy a fullframe Z-mount camera at some point in the future, but never took APS-C Z-mount seriously... until now! The Zfc really looks so cute and tempting...

Link | Posted on Jul 3, 2021 at 13:00 UTC as 6th comment

Really nice, I very much enjoyed watching this video. I definitely like the slight roller coaster effect. The beauty of the place is very well represented.

Link | Posted on May 16, 2021 at 11:49 UTC as 3rd comment
Total: 42, showing: 1 – 20
« First‹ Previous123Next ›Last »