Commons talk:Featured picture candidates

From Wikimedia Commons, the free media repository
Jump to: navigation, search

Portrait photography[edit]

Hi, I am a bit sick of people making silly comments about depth of field being too shallow in portrait photography. A shallow depth of field is feature, and that's how good portraits are made. See basic courses online: Normally f/2.8 and wider are the best, but for longer focal lengths, you can use up to f/4.0., [1], [2], [3], [4], etc. Please learn the basic techniques before voting. Thanks, Yann (talk) 13:40, 21 April 2018 (UTC)

Yann Whoi is commenting it?, it's a very basic rule. --The Photographer 13:48, 21 April 2018 (UTC)
Yann is angry with me for 'opposing' at this FPC nom and should say so up front intead of making insinuating comments. In another recent nom, f/4.0 was used and no one complained about the DoF being too long or against the rules of basic portrait photography then. (Or f/8.) --Cart (talk) 14:48, 21 April 2018 (UTC)
Well, the f-numer is only a part of the truth in terms of DoF. f/5.6 will have a different DoF depending on the focal length, 30 mm will have a poorer DoF than 80 mm, so that's also important. That's why the boy face and the soccer player look better than the painted face. Poco2 15:32, 21 April 2018 (UTC)
Thanks Poco for that info, I shall read up more on lenses. This little incident is exactly why people are more comfortable just avoiding to vote than speaking out in an 'oppose'. If you do so, you risk being dragged out here for a public flogging and be told that you don't know basic photography techniques. I could take offense for being told such a thing, but I'm pretty sure that Yann just said that in the heat of the moment. --Cart (talk) 17:09, 21 April 2018 (UTC)
Knowing him (I also had the pleasure in person) I am sure about that, too :) Poco2 17:15, 21 April 2018 (UTC)
I don't believe that a shallow DoF is necessary to make a good portrait. It always depends on the respective photograph. You can create a wonderful portrait at f/11 and a bad & messy looking one at f/1.4. So regarding the photo Yann is referring to I don't think it's worth being featured (I personally don't like the painting, the focus is not really on the eye and some other issues) but I wouldn't say DoF is an issue here. --Code (talk) 19:10, 21 April 2018 (UTC) P.S.: I think neither this nor that photo should have been promoted FP as both suffer from visible motion blur (while the DoF is perfectly ok IMO).
Code, what you say makes perfect sense. If applied at the right "starting point", a shallow DoF can be quite enough. Place it wrong, and it looks like you should have increased it. It is possible that the photographer of the face paint shot used auto focus instead of pinpoint focus and therefore it landed on the finger and nose. --Cart (talk) 19:33, 21 April 2018 (UTC)
Poco see Wikipedia DoF. "for the same subject magnification and the same f-number, all focal lengths for a given image format give approximately the same DOF." For example, on a FF camera, 100mm lens, f/4, subject at 3m gives DoF of 0.21m. You get the same DoF in a 50mm lens at 1.5m distance. And the same DoF in a 25mm lens at 0.75m distance. You have to move closer to get the same subject size. While the DoF is the same in all three, the telephoto will include less background than the wide-angle, so it will often appear less busy, but both are just as "out-of-focus". The problem that portrait photography is usually trying to solve with shallow DoF is to isolate the subject from the background and to avoid the background becoming a distraction. There is nothing in portrait photography that encourages the front of the nose or the ears to be out-of-focus. Nor, for a photo where the subject is facing the camera, is there anything that makes it desirable, never mind acceptable, for only one eye to be in focus. These things are all negative consequences of the photographer trying to isolate the background. For a studio photo, it is really up to the photographer to ensure their background is plain and neutral and far enough behind the model for it to be out-of-focus, and then they can use whatever aperture they want. Obviously, for a smaller aperture, they need a more powerful light, which could be a technical problem. For a photo of face painting, where the whole head and neck is painted, I can't understand why you wouldn't want all that to be in focus. The choice of focal length in that image is untypical (equivalent to 50mm full frame), and will exaggerate the model's nose and chin and head curve, compared to a more typical portrait focal length (85-135) -- and this is a result of having to be closer to the subject, rather than a property of the lens focal length or optics.
I agree that we often don't review portraits very well, as our pixel-peeping UI lets us too often focus on parts of the image rather than the whole. And sometimes the shallow DoF is just fine, and we don't need the sort of focus-stacking-front-to-back-sharpness obsession we see some voters apply everywhere. -- Colin (talk) 11:00, 23 April 2018 (UTC)
In my opinion the amount of DoF necessary in a portrait depends on the portrait itself. That painted face draws the viewers interest to every part of the face that has color painted on. The viewer wants to examine the whole painting on her face, so it needs to be sharp from at least her middle finger to her ears. When I shoot "business portraits" in conjunction with a studio setup and flash lights I mostly set my f-Stop to 11 to get the face and the visible part of the clothes worn completely sharp. The mentioned portrait of Kristina Inhof was shot with the sun just going down so I did not want to close the aperture too much and keep the ISO at a reasonable low level. Code, yes there is some motion blur. This sometimes becomes visible with my D850 even with the shutter speed being more than twice of the focal length, but you are the first one to pixel peep that much to forget the fact (25 pro-votes in five days are quite factious) that overall the image seems to be quite good. btw: You can always nominate that image for delisting. --Granada (talk) 11:42, 23 April 2018 (UTC)
  • The funny thing, given how this thread started, was that when I made my vote on the face paint photo, I had in mind an illustration from one of the very first photography books I've ever read. (Yes, I have read such! ;-) ) In the section on portraits there were two photos showing when a shallow DoF should be used in portraits and not. The "not-photo" was a woman in almost exactly the same position as the face-painted woman, only she had a microscope in front of her face instead of a hand. Of course the best DoF depends on the photo and situation. --Cart (talk) 13:07, 23 April 2018 (UTC)
  • @Granada: Well, my words above might have been a little bit harsh and I apologize for that. It's a good picture overall and you're right that one shouldn't pixelpeep that much. --Code (talk) 17:55, 24 April 2018 (UTC)
The discussion here is about portraits of people of course. I am a portrait photographer, but of animals, and here a large-as-sensibly-possible DoF is alwys better. And with macro, that is really tricky to manage of course as it is if I am using, say, a 400mm lens, quite close to the subject. Charles (talk) 15:51, 4 May 2018 (UTC)

Counting FPs[edit]

There is a template which I and many others use on our user pages to count our FPs:

Cscr-featured.svg Charles has uploaded 121 featured pictures to Wikimedia Commons.




It seems to me that Commons should have a template to distinguish 'own work' uploads from uploads of other people's work. Both are valuable, but are different. One would recognise photographic skill, the other would recognize users' efforts in adding high quality media to Commons. Charles (talk) 16:00, 4 May 2018 (UTC)

Hi, Sure. There are also restorations, which is a significant work, but different than taking pictures. Personally, I only claim FPs as mine, those which I took or did important restoration. For me, among 150 successful nominations, 16 were taken by me, 15 restored by me. I also have one FP nominated by someone else. Regards, Yann (talk) 17:57, 4 May 2018 (UTC)
Yes, I like the way you have handled this. I think restorations or old images should be considered as 'own work', but not changes to living photographer's works. Computer-generated images would also be 'own work'. 'Own work' imgaes are the only ones to qualify for the 'Meet our photogrpahers' page so you should be there. An FP nominated by someone else would definitely count. By the way, what happened to POTY 2017? Charles (talk) 19:48, 4 May 2018 (UTC)

FP galleries -- unsorted[edit]

I post it here instead of Commons talk:Featured pictures, as this place seems to be more widely watched. In Commons:Featured pictures, we have the problem of the overflow of "Unsorted" sections in many of the galleries (especially in the popular themes like "Places" there are lots of unsorted; see Architecture, for example). It's lot of work to move it all and close to impossible to manage it alone for anyone. May I kindly request at least our most active FPC nominators, that they do it at least with their own pictures? Example (it's also useful to fix the descriptions below the pictures, e.g. "Pashkov House in Moscow" instead of "Casa Pashkov, Moscú, Rusia, 2016-10-03, DD 36-37 HDR"). This would be a big step towards some order, thanks. --A.Savin 13:29, 8 May 2018 (UTC)

I usually sort a bit if I have business at one of th galleries, I'll try do do some more of it in the upcomming evenings. --Cart (talk) 13:54, 8 May 2018 (UTC)
+5 countries done -- Basile Morin (talk) 01:31, 9 May 2018 (UTC)
Thanks Basile, I did the rest. Whoever is sorting, please remember to remove the sorted photos from "Unsorted". I came across a few that were still there and this caused some doubles. Easy to fix of course, but better if we don't have to check for that too. --Cart (talk) 21:31, 9 May 2018 (UTC)
Cart, are you sure ? Here are my 5 edits done yesterday [1],[2], [3], [4], [5] and I don't see any mistake -- Basile Morin (talk) 23:02, 9 May 2018 (UTC)
Apologies Basile, that came out wrong. I was tired when I wrote it. That part of the post was not directed at you personally. I only saw that it had happened, I don't know who made the mistakes and we should try to avoid such doubles. --Cart (talk) 08:02, 10 May 2018 (UTC)
No worry, now I understand. Huge work you did, yes ! Thanks for the clarification -- Basile Morin (talk) 08:29, 10 May 2018 (UTC)

Thanks to both who responded, but I wonder if no further colleagues are reading it... --A.Savin 12:47, 9 May 2018 (UTC)

They are probably waiting for someone to construct a Bot to do this. :-) --Cart (talk) 21:34, 9 May 2018 (UTC)

Resolution and 100 percent views[edit]

We have a serious problem. It's a combination of resolution, lenses and views at 100 percent. I've been watching it since switching to a higher resolution camera some months ago. New cameras with a resolution of 30 MP and more needs lenses for this resolution. A good lens for an amateur photographer is nearly never good enough to respect the full resolution of the image sensor - or it is very expensive. If a reviewer take a look to the photograph at 100 percent level, the image looks unsharp - more or less, respecting the aperture too. The enlargement of only a few square centimetre or millimetre is enormous. IMO it's wrong to evaluate sharpness at 100 percent level without respecting the resolution. I found a really good article about this human problem at gwegner.de (3 Ursachen für unscharfe Fotos und wie ihr sie in den Griff bekommt!). Sorry, it's only in german. But it's a problem of reviews of FPCs with photographs in a high resolution. --XRay talk 07:33, 9 May 2018 (UTC)

An English translation here. I've also written an essay at User:Colin/PixelPeeping.
I think there is a trend with photographic gear towards higher prices for lenses, and technology has improved their sharpness considerably. This has encouraged reviews of gear which glorify pixel peeping or over-emphasise optical flaws by testing wide-open. Btw, my old 14MP Sony A33 from 2010 has a pixel pitch of 5.16 which is very slightly smaller (and thus more demanding of lenses when viewed 100%) than your 30MP Canon 5DIV at 5.36. The highest resolution 50MP Canon 5DS has a pixel pitch of 4.14, which is still larger (and thus less demanding of lenses when viewed at 100%) than the pixel pitch of my 24MP APS-C Sony A77ii at 3.92. So us APS-C shooters have been pushing the demands of lenses for a while. I have some cheap primes that are just great, so I don't feel encouraged to spend the $1000 manufacturers seem to want these days. There are also other factors than optics that mean a 100% pixel peep experience is disappointing. I think the main problem is that on nearly every other website, people view downsized images, and so come to expect pixel sharpness. -- Colin (talk) 09:03, 9 May 2018 (UTC)
FPC section should focus on compositions quality rather than the perfection of each pixels and I think that on FPC there is a competition to nominate the most gigantic and impossible to print picture. And we can see users buying falling on the comercial game called consumerism and censor camera mansturbation that kill the creative activity. Additionally we have a mediawiki that makes it difficult to see the photo in full size, and yes only a small people percentage can see the image description page thanks to the excellent decision of WMF to create the wall Media viewer. At this moment I am receiving much more feedback on flickr than on wikimedia commons to use my photos in educational books. --The Photographer 05:01, 12 May 2018 (UTC)
Yes, pixel peeping is getting irrational here, in spite of all the warnings not to do it. :( Yann (talk) 05:30, 12 May 2018 (UTC)
I think it important we remember Commons is a media repository so the failings of our UI to display images well should not be used as a criticism of large images. We have zoom viewer but I agree it would be much better if the UI was improved. We do require a significant number of pixels for many publishing formats. See User:Colin/PixelPeeping for some figures. To be used in a 4K TV video, one of our 3:2 images needs to be 9.8MP. To fill a two-page spread in a glossy magazine like Vogue requires 16.7MP. In the future, an 8K video will require a still from an 39.3MP 3:2 format camera. My hope is that as the DPI of displays has finally doubled to 200DPI that we will start appreciating more that the tiniest sliver of purple on a high contrast edge, or the faint dots of noise in a sky become totally invisible when viewing reasonably, just as they would if printed. Whereas if your image lacks wow and strong composition, nobody will even give it a second glance. -- Colin (talk) 14:58, 12 May 2018 (UTC)

Terabite: The World’s First Terapixel Macro Image[edit]

"To get some idea of size: if we printed the Terabite image at at high resolution, it would be higher than One World Trade Center!". Sounds familiar. You can pixel peep here. -- Colin (talk) 19:48, 9 May 2018 (UTC)

Meh, DOF too shallow, they should have focus-stacked the whole thing. Scnr, --El Grafo (talk) 08:37, 10 May 2018 (UTC)
Exactly what I thought. --Code (talk) 09:00, 10 May 2018 (UTC)
+1 ;-) --XRay talk 09:11, 10 May 2018 (UTC)
There is also an awful lot of dust that should have been cloned out " [food] photography like this demands pefect cleanliness". :-) --Cart (talk) 09:18, 10 May 2018 (UTC)
Oh, and it's a terrible portrait as well: the eyes are out of focus. :-P --El Grafo (talk) 10:05, 10 May 2018 (UTC)
Apparently they did focus stack, taking six frames, each with about 1mm DoF to give 5mm DoF overall. For the eyes, I'm guessing a bit of olive on top of a cherry tomato on top of a slice of courgette (zucchini), which probably exceeds the 5mm limit. I think they should have got children to do the artwork. If you are going to all that effort, the image could have been a bit more imaginative. -- Colin (talk) 10:29, 10 May 2018 (UTC)
Also problems joining the images, I can do something better --The Photographer 04:46, 12 May 2018 (UTC)

Commons:Featured_picture_candidates/File:700 years Old Baltit Fort.jpg[edit]

Post copied/moved from User talk:W.carter:

"Hi, I reverted this nomination, as I expected it to be FPXed. I see no point in helping someone adding an incomplete nomination, just to be removed one day later. Regards, Yann (talk) 05:51, 13 May 2018 (UTC)"
Hi Yann, the normal use for reverting an edit is when something is really wrong or in case of vandalism. Reverting a nom instead of FPX it would mean a change in policy and discussed by the community. In this case the nom was not incomplete, the user has created a nomination page, they simply made a fumble in placing the transclusion in the right place (the edit you reverted and I fixed). So the nom would have been open, just not visible on the FPC page, until the nine days were up and after that it would have popped up on Category:Featured picture candidates awaiting closure review and questions would have been raised as to what had happened.
In any case, I think that every user has the right to at least make an FPC nomination if they want to. An FPX is also a kind of review, at least the user will know what happened and maybe think more carefully about their next nom. Simply reverting noms from FPC will only leave them confused as to what happened. An FPX can also be contested in some way, something you yourself did with this. It is harder to contest a revert. --Cart (talk) 08:38, 13 May 2018 (UTC)
Please see my answer in your talk page. Regards, Yann (talk) 08:48, 13 May 2018 (UTC)
Of course I can, I just don't see why this can't be discussed here. --Cart (talk) 08:49, 13 May 2018 (UTC)

Galleries style[edit]

@Colin: I remember you complained about the style of some galleries, do you think this kind of thing is ok/better Commons:Featured pictures/Places/Interiors? Christian Ferrer (talk) 17:26, 13 May 2018 (UTC)

I think at the time the "packed-hover" wasn't working for me on the browser I was using. It showed the title over the image at all times, which was just too much. But nobody else saw that bug, and it seems to be working for me now. I prefer the packed style because the default thumb size and layout (which you link above) is too small and wastes too much space. The pictures need to be big for people to appreciate how good they are. -- Colin (talk) 07:17, 14 May 2018 (UTC)
+1 --Cart (talk) 10:00, 14 May 2018 (UTC)

FPC criteria[edit]

We should not be judging single-shot photos against the same criteria as multiple-image compositions. I don't think the criteria are clear enough and voters will inevitiably prefer the impact of a stiched panorama or focus-stacked image. I propose that FPC nominations should have some sort of 'multiple shot' flag to allow both photographic skills and post-processing skills to be judged on appropriate criteria. Charles (talk) 13:08, 16 May 2018 (UTC)

  • Could be a good idea, but I don't know if it will make any difference. Folks here tend to compare the two ways of making photos even when it is clearly stated in the file decription how the image is made. --Cart (talk) 13:21, 16 May 2018 (UTC)
  • This seems to have arisen from this FP candidate where a focus stacked image here is compared against a standard short here (shot close-up at the FF equivalent of 75mm, f/7). There have also been a couple of focus-stacked insect photos.
I'm not sure separating technique or equipment is the key rather than the intended result. You can shoot a panorama of the river Thames with a wide-angle lens and some vertical cropping, or by stitching together any number of frames. The result at small size may well appear identical, with the difference in quality only apparent when zooming in. So I'm not sure why we'd not want to directly compare the two photos and consider the single shot one weaker (unless it had better light, etc). Similarly an HDR technique, if done well, can produce natural looking results that fix issues with blown highlights and crushed blacks, and should be compared against a basic shot that has these defects.
For focus stacking, I think it more important that we don't always consider that front-to-back sharpness is necessary/desirable, and that we don't get so wowed by technique that we forget the artistry and beauty of a photograph. Sometimes the stacking provides very impressive results, and it is a fun technique to try out (I've done a few). But it can create a rather artificial look, like something computer generated, with fewer depth clues for the eye. And it can produce rather extreme transitions to out-of-focus which can look strange. There's also the risk, with complex shapes, of weird artefacts where the stacking algorithm hasn't worked. Much like HDR has the risk of bad results if there is movement in the scene.
These techniques are just other tools in the box, like flash lighting, a polarising filter, or having a macro lens. Sometimes they enhance a photo a little, sometimes make a great photo possible that wouldn't otherwise, and sometimes they ruin things. I'd rather we differentiated between a specimen photo for Wikipedia and a more artistic photo and also less common photos like those showing animal behaviour. If someone manages to grab a great image of two animals fighting, we shouldn't demand the same resolution/sharpness/lighting that we might demand for the "boring" photo of the animal minding its own business and obediently staring at the camera. And if someone takes a portrait of a mushroom the woods, it will have a different character to if they setup a flash and tripod and aim to capture a specimen photo of just the mushroom. -- Colin (talk) 14:55, 16 May 2018 (UTC)
All I'm suggesting is that nominations flag up the use of sophisticated digital manipulation tools, so that voters know what they are looking at. What about a requirement that a nomination says (in the Info space). Panorama made from 34 images or Photo-stacked from 10 images? Charles (talk) 16:01, 16 May 2018 (UTC)
That's of course fine. I don't think anyone so far is holding photos of animal behavior to the same standards of sharpness as an FPC-nominated picture of a watch, though, and all these matters of technique are great for photographers to be aware of, but the viewing experience is most important in judging the result. -- Ikan Kekek (talk) 18:56, 20 May 2018 (UTC)
Agree with Ikan that the viewing experience is the most important. I understand the problem, and sometimes find my own landscapes not as detailed or not as sharp as others nominated here using stretching techniques. However I'm not in favor of setting special flags for each kind of photography. We present here any kind of images, sometimes paintings, portraits, animations, also professional pictures with important post-processing. The result is the result. We should judge the photo as it is, independently to the technique. If a photo succeeds because it's stretched or focus-stacked, then it's all good for the photographer. Means the technique led to a good result. Reviews from voters must be reliable. If a single shot works better because the bokeh is great, then well done, this choice of a low DoF was productive. It depends on each image. Also difficult to define the limit and say which small area was exceptionally pasted here from an extra image. Such flags would also encourage the users in hiding the reality. For example by stretching and then reducing the size to get more sharpness. If all the images were better focus-stacked (and I don't think so), then we should universally adopt the technique. But in reality, images depend mostly on non-technical aspects. More a situation, a lighting, some colors, a great action, etc, making the picture special. To finish, we have also to consider the fact that post-processing of composed images always require more work from the creator than single shots, and are more likely to fail for this reason of weaknesses visible in the technical part -- Basile Morin (talk) 03:50, 21 May 2018 (UTC)

FYI FPCBot[edit]

FYI, I've asked about FPCBot. Hopefully something can be done. --Cart (talk) 09:09, 19 May 2018 (UTC)