Obviously this doesn't mean much, but just for interest I actually did not touch the saturation at all in terms of using the Hue/Saturation tool in Photoshop. I achieved that effect through a few rather unconventional techniques which worked well on this image because of the blue sky (enabled me to get an almost perfect cut out through the blue channel). Obviously I can't tell what the pic looks like on your monitor or what color of the sky you think is "normal", but personally I think it's just the prefect hue for that kind of sky (high above the horizon). Anyway that's just for interest; as obviously saturation is a percieved strength of the colors, not just adjustment through the hue/saturation tool, but through any means of boosting the colors. --Fir000212:20, 2 October 2006 (UTC)[reply]
Although I do sometimes disagree with the sentiment and results of some of your edits, I have to admit that you do yourself no favours by uploading an edit without any explanation of what you have done (except in circumstances like this when you feel the need to explain it). Sometimes it is obvious, but other times, subtle adjustments that you have made can be missed because we aren't aware we should be looking for them. In this case though, I'm not sure exactly what you are trying to say you did to the image, but it almost looks to me like you painted the sky - There is a distinct lack of contrast around the aerial (it 'fades' into the sky compared to the original and other edits) above the R and around the canopy, and a strange halo around the blurred propellor blades. Diliff | (Talk)(Contribs)13:24, 2 October 2006 (UTC)[reply]
Well see this is why I don't usually explain what I do in my edits because it is usually a complex process. In this particular image, I extracted the blue channel as a seperate layer. I then applied levels until I had the sky a pure white and the plane a dark black. I painted over the blue markings on the plane in black. I then inverted that layer. Then I used levels to reduce the white point to a gray (about 1/3 down the slider by memory). Then I put that layer into softlight ontop of the original, and bang! It looks good. --Fir000222:22, 2 October 2006 (UTC)[reply]
Well fair enough but bang! It has also introduced haloes around the propellers and removed detail from the areas I mentioned above. I don't think it is as good as you think it is - too many side-effects. Diliff | (Talk)(Contribs)07:00, 3 October 2006 (UTC)[reply]
Fair enough about the aerial (minor detail), but I really can't see where you get your haloes from. Several images you've mentioned haloes where I can't seen any. Oh well I guess it's impossible to determine what the average person would see on an average monitor. You're free to think it's not very good, but I hope I'm also extended the right to think it is good. The only side effect I'm getting is the less prominent aerial - something I personally am quite willing to sacrifice in the face of the huge improvements elsewhere. --Fir000208:33, 3 October 2006 (UTC)[reply]
Well, I urge you to compare the propellers in your image and the others. There are definitely artifacts, particularly on the far left side. The problem I see is that you didn't need to sacrifice anything if only you had masked the plane better when you were working your black magic. And as for average people seeing things on average monitors, I can't imagine how anyone could miss it on any monitor if only they lookd closely, as it isn't at the light or dark end of the spectrum here - as long as you can see mid-range detail, you should be able to see the faults I mentioned. If you still can't see it, I will go ahead and create an image with the faults magnified side-by-side later, but I can't at the moment as I'm at work. And yes, I extend you the right to think whatever you like, but I'll try to point out the flaws in your logic if they're presented in FPC ;-). This is about creating the best final image, when it comes to it. Diliff | (Talk)(Contribs)09:58, 3 October 2006 (UTC)[reply]
It's all very well to say you'll point out flaws in my logic, but since I'm not presenting logic and all I'm trying to do is help, I'd rather you didn't point flaws like "I don't think it is as good as you think it is" - not overly helpful really. And I'm all for making the best final image, that's why I'm here. But if you don't think my edit is any good than why are you so worried about it? And if you do think it is good but there are flaws, why not try fix them? --Fir000208:23, 4 October 2006 (UTC)[reply]
I've just added some elargements to highlight the haloes and strange effects that your edit has had, as well as strong noise visible. I feel that the canopy above the pilots head lacks definition in Edit 4. Diliff | (Talk)(Contribs)18:29, 3 October 2006 (UTC)[reply]
I think you're a little hung up on these haloes, I can see what you're trying to show but it's such an imperceptible (at least to me) detail that I'm surprised you've brought it up. True my edit has more noise, but come on! You can't even see it at 100%! But to oblige you, here's another edit. --Fir000208:23, 4 October 2006 (UTC)[reply]
Gentlemen, one question (maybe dumb, certainly impertinent): Why are you so "hot" on making (often unrequested) edits, many of them rather pointless (example: Zanzibar)? Sure, you've both done a lot of good edit work, but I think edits should be held to a minimum number, since too many make the closing of a nom really hard - remember all the previous talk, and the re-voting requests by closers. Of course, when an edit is requested (such as in this nomination), then it's time to act, but do we really need four edits, thus five versions to vote on, for a simple picture like this? BTW, Fir, I must agree with Diliff, especially after seeing the enlarged example - the cockpit cover has a "cookie-cutter" edge in edit 4, and there are indeed artifacts around the prop. --Janke | Talk07:34, 4 October 2006 (UTC)[reply]
Well Janke, I don't usually make edits that are unrequested. If there is an obvious reason for doing so (unnecessary noise, bad colour balance etc) then I will attempt to salvage it without a request to do so, but usually I'm happy to leave an image as-is. In the case of this image, it was requested that the shadows were lifted. First, Pharaoh Hound had a go, and I felt I could achieve the same result only with a bit more finesse, so I did. As mine was basically an objective improvement over Pharaoh Hound's (obviously that is subjective but I think its an improvement as far as anyone can be sure of such a thing), and nobody has supported her edits, I think its fair to say we could probably eliminate her edit for simplicity's sake. Likewise, my crop edit could easily be removed as it was proven to be relatively unpopular. At the time that I submitted those two edits, I didn't expect Fir0002 to also add another two edits, making a total of 6 images to compare. I agree that it becomes difficult to determine the best image, and particularly difficult to determine which is the more popular. Diliff | (Talk)(Contribs)09:24, 4 October 2006 (UTC)[reply]
Going back to left field... ;-) Friends, this is not about collaboration. It's about five (now six) very similar versions of a single image, which discombobulates the voters, to say the least. In most edits you both have done (almost all of them good, of course, and you know that) the differences are so small that only experts like you (and me? ;-) see any so marked a difference that they can definitely decide. Why else would people vote differently! Here, some prefer edit 2, some 3, some 4, and now there's a new one! They cannot all be the "best" one. That's what bothers me - so many versions that the voters can't agree upon any one of them... I hope you understand my concern. --Janke | Talk13:01, 4 October 2006 (UTC)[reply]
The editing thing is something that bugs me slightly too. I do like some of the edits, and I appreciate that people are only trying to improve things. But on the other hand, I dislike some of the edits so much that I will really regret uploading the image to wikipedia in the first place if they become a featured pic since I consider them far inferior to the original, and I don't particularly want my name associated with them. It puts me off uploading more of my pics. chowells14:13, 4 October 2006 (UTC)[reply]
I do understand Chowells. Its quite a different feeling being on the receiving end of what you consider to be inferior edits. I've had that experience with a number of my FPC images being edited by people such as Fir0002 (and I'm not singling him out for reasons of malice, it just happens that he does a lot of the edits!). For example, he consistently lightens the shadows of the image and increases saturation/contrast on just about every image he edits. Sometimes it just isn't natural looking and completely unnecessary. I completely appreciate that in the case of this image, you're not a beginner photographer who happened to get lucky or be in the right place at the right time (although lets face it, thats what photography is sometimes!) and needs some help with the image he snapped. I personally would have been quite happy to leave the edits alone and support the original but just responded to the numerous requests and to improve on Pharaoh Hound's copy. In the end, everyone has their own opinion, and everyone has a different monitor with different calibration so we're never even seeing the same image! Its possible that Fir0002's monitor is too dark and too undersaturated ;-). Anyway, thanks for your opinion on the matter. Diliff | (Talk)(Contribs)14:33, 4 October 2006 (UTC)[reply]
I must say that I agree with Diliff. Many edits I see are oversaturated and oversharpened, making maybe a slightly "flahsier" image, but one thats dimished quality can bee noticed without too much scrutiny. Anyway, maybe this should be moved to the additional input from voters section, considering it's rather obvious we've got a support here, it's just which edit we have to decide on now. Anyway, I think I'll remove Pharaoh's edit (no offense to her) and edit 3, just for simplicity's sake. Any objections? NauticaShades(talk)19:13, 4 October 2006 (UTC)[reply]
Side comment, related to this: "Its possible that Fir0002's monitor is too dark and too undersaturated ;-)."
I realize this was intended as a joke, but I actually contacted Fir0002 about the differences between his edits and most other people's edits a while back, suggesting that perhaps his monitor should be calibrated. I find most of his edits (and some of his own photos) to be overbright and/or oversaturated. For most images it's only a subtle difference, but occasionally it's a big difference -- usually on images with lots of blue sky in them. I have a hard time believing Fir0002 would actually prefer his edits over most others if he saw them as displayed on my screen. I honestly don't know if it's a question of calibration or if his monitor simply isn't delivering the brightness and contrast ratio that mine does.
Geez, sure makes me wish I wasn't in a different time zone, that's one helluva discussion that's been going on while I was sleeping. I will try address each comment in order.
Re Janke: Can you see a definite difference b/w mine and Diliff's edits here? Can you point to a nomination where the edits were near identical? And it pays to remember that we are supposed to have some idea of image differences, as we are judging the best photographic content on wiki here. I mean would you expect voters on FAC to be illiterate?
Re Chowells: With all due respect, if the community thinks they are better than the original that's all that matters. As this is a collaboritive wiki, community consensus is law.
Re Diliff: There may be another reason for my edits - they may just look better? And for my part I would say a lot of your images are too dark and can do with saturation/contrast. Perhaps your monitor is too bright and oversaturated ;-) I'm sorry if you don't appreciate my edits because they take time and effort. It's not like I just upload an edit for the heck or it. And I continue to make edits because it seems that many people do appreciate them.
Re NauticaShades: Is a flashier image a bad thing?
Re Moondigger: I can only say that my monitor passes the circles test just fine. It also is very close to the image that it prints to. Furthermore I doubt I would prefer other edits over mine (unless they looked better on this monitor) as I have access to a variety of monitors at school and elsewhere and on all of them my edits look better than the alternatives. And moondigger, you have mentioned you were part of photography forums like Fredmiranda - can you seriously claim my images are oversaturated compared to the standard there (which I personally thing is exemplary)?
Anyway bottom line is, I don't think too many edits is a problem (this is the first case were people have complained of too much to choose from - you guys are spoilt!), and if my edits are not to a reasonable standard, than why are people supporting them? If they didn't I'd obviously have to admit that I have a very different conception of what a good image looks like. As a side note, one of the main inspirations (that sounds lame - perhaps influences would be better) would be the photos on Webshots. If you look through their galleries, you'll hopefully see the style I try emulate. --Fir000222:13, 4 October 2006 (UTC)[reply]
Well, I'll try to respond to a few points.. I agree that sometimes my images are dark, sometimes a little too dark, but on other occasions if they weren't dark then the ambience of the scene would be ruined. I think back to the Notre Dame de Montreal Basilica image where I was adamant that it really just was that dark in the church and brightening it was not as encyclopaedic. But that said, I never said I didn't appreciate any of your edits. I just think that sometimes they are unnecessary. And regarding Nauticashades' comment, I agree with him - sometimes flashier images are a bad thing, if they start to deviate from visual reality for the purpose of flashiness. And sometimes you do get a lot of support for your edits but people can sometimes be idiots, too. ;-) Not everyone is a good judge of what a natural looking photo should be. I'm not saying I'm the expert, but I think that those of us commenting here tend to be at the higher end of the spectrum than most. Lets be honest though, Webshots photos would not necessarily pass the encyclopaedic value test either. They're often oversaturated, warm-n-fuzzy photos designed to make people say "Wow, so beautiful...". We're programmed as humans to appreciate bright, saturated colours, but that doesn't mean we should do it just to please the masses. This isn't webshots! Diliff | (Talk)(Contribs)07:08, 5 October 2006 (UTC)[reply]
Well I would disagree, an image can be flashy and encylopedic. In fact that is what we should be looking for here on FPC. Sure an image can be encyclopedic, I mean for a random example this is a perfectly encyclopedic image but it hasn't got any "flashiness" and so wouldn't fare all that well on FPC. By "those of us commenting here" I assume you mean on this talk page? Well either way, I think you're treading a pretty fine line there Diliff, everyone is equal here. Sure it'd be nice to have a formal review with more "upper spectrum" reviewers, but that's not how it works on wiki. Remember wiki isn't going to be only for "upper spectrum" end users. And yes, lets be honest - do you seriously think that images from Webshots, images taken by professionals, wouldn't sail through FPC? And true this isn't webshots, but that doesn't mean we have to desert art.--Fir000213:12, 5 October 2006 (UTC)[reply]
To go on a slightly related tangent, how would you define reality anyway? I mean would you say that Image:Bangkok skytrain sunset.jpg didn't show reality because my reality is 12 O'clock noon and there is now red tint to the scene? Of course not. Similarly under different lighting conditions virtually any "edit" could be a reality. And I suppose you could argue that this doesn't prove anything because it wasn't the reality at the time the image was taken. But so what? How is that "reality" going to be any more encyclopedic than a "reality" of a different photo under different light? Also would you say a sharp image doesn't show reality because I'm partially blind and my reality is all fuzzy? These are obviously pretty ridiculous examples, but shadows are a real example. The eye can capture so much more dynamic range than a camera, and I personally think that the Notre Dame Basillica could look how my edit looked. And maybe it does look like that to a six year old child. But to you it may look darker, flatter? I don't know but I think it's an intersting line of thought. --Fir000213:12, 5 October 2006 (UTC)[reply]
Ooh, dear, dear Fir, did my original comments get your feathers all furred up... that was not my intention. What prompted my original remark was both this nom, as well as the Zanzibar one, which I felt was pointless - oppose votes due to a non-encyclopedic subject, and suddenly we get a "splashy edit", which doesn't affect the main issue. As I've said, you most often do a very good job (and you know it), but sometimes I feel there's a little too much of the good - here, six different versions of the plane (OK, only two are yours). Re. your direct question to me: Yes, I see clear differences here, but do all voters? The main difference in your edit(s) is the increased saturation of the sky (the only thing visible in the thumbnails, which many peolpe in fact vote on, they don't bother to look at the full sized images! There are numerous past examples of that - borders on "illiteracy", doesn't it?). Then only in full-size, you see the halo around the plane, and the loss of detail in the antenna and the cockpit roof contour. You always loose something as soon as you edit a jpg image, due to de-/recompression, but in your edits, even more is lost. I originally abstained from voting, because I wanted to make a point (yes, I break the "rules" when I feel like it... ;-), but here, Diliff's edit(s) are definitely better, since they don't introduce artifacts (halo), nor do they remove detail (antenna, roof). I believe you see these clear differences too, and that's why I wondered: Why did you make an edit which looks "splashier", but is clearly inferior in losing detail and introducing artifacts? This is not meant offensively, rather as friendly advice - we're collaborating, remember! --Janke | Talk07:38, 5 October 2006 (UTC)[reply]
Well I'm glad you've remembered about the collaborating part, for a whlie there I thought you'd forgotten. And I'd like to assume good faith and all, but what's with the patronzing "Ooh, dear, dear Fir, did my original comments get your feathers all furred up"? In future I would much rather not to be addressed like a 6 yr old. Well can you blame me for making edits which people can't tell the difference from if they only look at the thumbnails? It's the full res image that we are judging here, and as such I reserve the right to make edits which do look significantly different at full size. Even at full res you can't see the halo (at least I can't) - it's only really visible when you go to 200% (or is it 300%) like in Diliff's crop comparison. Likewise with the noise, it's flawless at 100%. Furthermore since you guys were so concerned about it I made another edit which fixed those concerns. Remember? And I don't see an edit which is clearly inferior in losing detail and introducing artefacts. The only detail lost that I can see (and it's not really lost - if you have a correctly calibrated monitor ;-) is the aerial. And as I said before that's not something I'm wiling to trade for the sake of losing all the improvements in my image. I mean the original I feel is inaccurate in capturing the color of the sky. That kind of pale cyan looks pretty unnatural to me, and looks more like the camera's inability to capture the dynamic range. As a side point to show you the generally small image quality loss between progressive jpegs saves have a look at this: Image:Original crop.jpgSaved for Web 80 quality 5 times, Saved for Web 80 qualitySaved for Web 60 quality 5 times, Saved for Web 60 quality. Of course there will be a loss, but generally it's not very visible. --Fir000213:12, 5 October 2006 (UTC)[reply]
Hey, I didn't mean to be condescending, just lighten up the discussion a bit! I'm sorry if it felt patronizing, that was not my intention! I like a little ;-) now and then. (Yes, my monitor is calibrated... ;-)
You defend your edits well, but one thing still not addressed is the canopy top - don't you see how it is almost lost in your edit, where the canopy looks like a soap bubble, not like acrylic. The difference is visible even in the small size on the image page, and is much more marked in full size. I really don't approve of any edits that lose detail in this way. Sure, the aerial is still there, but it is much weakened. You also say: "the original I feel is inaccurate in capturing the color of the sky" - we don't know how the sky actually looked like that day, on that hour, do we? So, let's agree to disagree - you like your edit(s), I like Diliff's which hasn't lost those details, and doesn't have such a strong sky color... --Janke | Talk15:32, 5 October 2006 (UTC)[reply]
How is a reduction in contrast a lose of detail worthy of disapproval? Any image editing is a loss of detail if you want to be like that. What exactly is missing? Have I cloned something out? No. It just that as a result of my processing you no longer see the dark border seperating sky from cockpit. And complaining about that is very inconsistant. I mean you complain that a certain area has lost darkness, yet are happy with the majority of the plane from losing it's darkness in Diliff's edits. True we don't know what the sky looked like, but it's pretty easy to imagine. It's not like there is a significant shift in color of the sky from day to day :-). And I'm fine with you not liking my edit, that's OK by me. --Fir000207:33, 6 October 2006 (UTC)[reply]
As far as I can see, the only thing that WAS addressed was the halo around the propeller, but even then, it appears that the only part of the halo issue that was removed was the left side of a single propeller. The other side still has the same problem. He may have also reduced noise in the sky, I haven't had a good opportunity to compare side by side, but the sky looks a bit less noisy.
Fir, as I see it, the following issues still haven't been addressed in Edit 5:
1. The aerial is still faded.
2. As janke said, the canopy is still perculiar looking, and definitely not like the original looks.
3. The haloes are still visible around other parts of the propellers (ie it was only partially fixed - I assume by patching the halo up rather than by re-processing from the original).
4. I didn't raise this one earlier but I noticed that the sharpening is a little strong. It doesn't seem to have increased apparent sharpness, all it seems to have done is left light/dark borders around contrasty areas. Look, this is all getting too hard. ;-) If after all this, you can't be bothered fixing these issues, it doesn't really matter. I'm not asking for you to create the perfect edit. I'm just letting you/others know what I feel are issues, if you disagree, thats ok too. Diliff | (Talk)(Contribs)15:55, 5 October 2006 (UTC)[reply]
Well all those "issues" you raised are non existant. The aerial hasn't faded, it's just been lightened along with the plane. The cockpit does look different from the original, but then again so does the whole plane - and if it didn't then the edit would be pretty useless wouldn't it? I can't see any haloes! If as Janke says most people don't even look at 100%, then I'd harldy be worrying about details you can't see until you blow up to 300%. Show me a 100% crop which you can see haloes in. Personally I see significant sharpness improvements, particularly inside the cockpit on the pilot. And yes I do disagree, but I'm glad that's OK. This may sound a bit perverse but I've actually enjoyed this discussion, and so here's a token of thanks (feel free to move them to your respective userpages). --Fir000207:34, 6 October 2006 (UTC)[reply]
I really wouldn't say that these issues are nonexistant. Diliff's crop clearly shows the cockpit detail differences, which I would say weren't addressed in edit 5. The cockpit in edits 4/5 has clearly been sharpened (I thik excessively) and much of the detail of the glass is lost. With edits 2/3, you can clearly see the glass, and it i much more defined, as opposed to edits 4/5, in which the glass almost is part of the sky. On a side note, I thought I might leave you with a barnstar myself. NauticaShades(talk)08:25, 7 October 2006 (UTC)[reply]
Regarding jpg losses, here you can see the difference between two of Fir's examples above, the original and the 60_5x version - note how the compression gives artifacts especially around edges, i.e. the flower in the hair and the forehead and eylashes. This example was very quickly made: two levels, on bottom the 5x saved, on top the original with difference filter, then flatten, do auto level and invert to accentuate the difference. Even though you can hardly see the atrifacts when looking at the images themselves, this shows it all. Interesting, don't you think? --Janke | Talk17:39, 5 October 2006 (UTC)[reply]
Re. monitor calibration: I added a new calibration image to the FPC page, which will show errors in color gamma. Please have a look! --Janke | Talk07:26, 6 October 2006 (UTC)[reply]
Ok, I'm removing edit 1 and 3 (turns out 3 is actually quite popular), (and I'll add a small note indicating it exists). Feel free to undo this if you feel they should be left. I simply haven't had any objections to my suggestion earlier. NauticaShades(talk)10:33, 5 October 2006 (UTC)[reply]