- Are you sure about the lack of sharnpess? For the edge of a 12 MPix wide angle shot, that's great sharpness. To be honest, it would even be fine for that area to be out of focus imo. — Julian H.✈ 07:55, 6 March 2015 (UTC)[reply]
- He's kind of right. The borders, particularly the left one, are not that sharp. It's something that I've observed on many of Poco's photos taken with the 24-105mm f/4L lens, and it's also something that happened on my 24-105mm lens. But yeah, I think the problem is not too bad in this photo. In other photos taken with the lens it is much more of a problem. Diliff (talk) 12:40, 6 March 2015 (UTC)[reply]
- Some pictures get promoted here being much softer in the centre (centre!) of the image. I see it quite sharp.
 Support ¡Muy bonita! --Kadellar (talk) 12:55, 6 March 2015 (UTC)[reply]
- Knowing the image passed through Lightroom, that issue at the left could have been corrected. The 24-105mm is an excellent lens for its versatility but crap for its optical quality (except around 70mm) imo, even more for landscape or FPC where quality at 100% matters.
- Out of focus borders? It would be ok (= natural) if their distance were closer than the divers, but here it's clearly a lack of sharpness due to the lens.
- "...it's more than good enough for FP for me...": that's the "problem" with the FPC page as it depends on the greatly different levels of appreciation of the voters. Imho FP shouldn't be "good enough" but among the best on the composition and technically, something special, because there's also QI. Sting (talk) 17:38, 6 March 2015 (UTC)[reply]
- True, the 24-105mm f/4L is not the best optically, but it is usually acceptable. What I found in my experience with the lens is that it occasionally will severely blur one side of the image. It was very difficult to replicate the problem every time, which makes me think it is related to the image stabilisation. A perfect example of how bad it can be is one of Poco's recent nominations. This is well outside of what would be considered acceptable corner sharpness and is a fault with the lens IMO. Although this image's edge sharpness is a lot better than that one I linked to, it is a manifestation of the same problem, I think. Diliff (talk) 18:28, 6 March 2015 (UTC)[reply]
- The linked example doesn't look like a sharpness problem to me though, it's almost exclusively vertical, the vertical lines are quite sharp even on the left side. In some areas, you can even discern two reproductions of some detail on top of each other. I can't think of a way that a lens can do such a thing and would therefore expect that to be camera shake. — Julian H.✈ 22:29, 6 March 2015 (UTC)[reply]
- If the image isn't sharp, then it is a sharpness problem, regardless of what is the actual cause of it. That's what I was saying earlier though, I believe it is the image stabilisation of the lens that is causing the problem rather than camera shake. If it was camera shake, the entire image would be blurry equally, not just the left side. But when IS shifts the lens elements, it does so in three dimensions. This the only way you could get motion blur only on one side. I suspect the IS has pivoted one of the lens elements such that the middle and right side remained relatively stable, but the left side shifted considerably. Diliff (talk) 22:53, 6 March 2015 (UTC)[reply]
- Sorry, I didn't mean it's not a sharpness problem, I meant it's not a lens sharpness problem. Rotational camera shake can easily cause blurryness in a part of the frame. If the IS corrects the translation of the camera shake, what is left is the rotation. So if you rotate around your right hand, the result can look somewhat like this. — Julian H.✈ 23:15, 6 March 2015 (UTC)[reply]
- Wow Julian! You should change monitor or wear glasses (no offence)! That Montenegro picture is an optical quality masterpiece! @ Poco a poco: I'm not happy with my 24-105 but in your case you should process Canon because you got a lemon! In that image, the left 2/3 are totally unusable. No post-processing will recover it. I hate advertisement (and the company) selling you at gold price the heaven while they push you into hell. Imagine showing that photograph to your Montenegro's gov client: you'll get burned on public place at once, with some oil to help. If you allow me a friendly advice: next time you change your camera, sell that lens together as a bundle, as fast as possible and get an other one if you want it again (I'm already sorry for the buyer). Sting (talk) 23:36, 6 March 2015 (UTC)[reply]
- PS: I've got a similar and even worse problem on a fixed lens, a TS-E (sold at gold price, remember?): setting it at 0°, one side is blurred, more than in the Montenegros's one. Setting it at 1/4 or 1/3 of a degree, it's perfect. I know it and I can change these settings on that lens, so that's ok for me, but not acceptable as a customer. That 24-105 may have a lens paralellism problem.
- Julian, perhaps but remember the blur direction seems vertical so a horizontal rotation around the right hand couldn't cause that. Also, I can confirm that I've had similar problems with my Canon 24-105mm and there was no camera movement/shake at the time. It is the lens itself malfunctioning IMO. Sting, you don't have to worry about Poco's lens now though, he sadly had all his camera gear stolen in Argentina a few months ago, and these photos were from August 2014. I hope for the sake of the photos that he doesn't get the 24-105mm again though. Sigma make a lens with exactly the same focal length range and their lens is supposed to be sharper. If I had the chance again, I would buy the Sigma. Diliff (talk) 00:03, 7 March 2015 (UTC)[reply]
- I'm really sorry for Poco a poco. Harsh and sad times. Sting (talk) 00:26, 7 March 2015 (UTC)[reply]
- Vertical is the only blur direction that could be caused by a rotation around the direction of light. — Julian H.✈ 13:30, 7 March 2015 (UTC)[reply]
- I'm not really what you mean by that. Vertical blur usually occurs when the camera rotates around a horizontal axis vertically, perpendicular to the direction of the camera. I originally understood "rotate around your right hand" to mean a horizontal rotation around the vertical axis, but I now think you mean a vertical rotation around the optical axis of the lens. Horizontal blur usually occurs when the camera rotates around a vertical axis horizontally, perpendicular to the direction of the camera. Are you saying that you think Poco twisted the camera along the axis of the lens as he took the photo, pivoting around his right hand so much that he had blur only on one side of the frame?? I suppose it's possible, but that would be very difficult to achieve unless you were trying to do it deliberately. I can't imagine Poco could do it by accident as it would require quite a significant movement in 1/80th of a second. I think it's far more likely that the IS shifted the lens element incorrectly in that moment than Poco physically rotating the entire camera. Diliff (talk) 14:17, 7 March 2015 (UTC)[reply]
- Yes, that's exactly what I mean. And yes, it's not the most common kind of camera shake. But the image is rotated. A vertical blur on the left and none on the right means a rotation along the optical axis, no matter how you look at it. And I can't come up with a way that an IS can rotate the image, ever. I don't think that's possible. At least not with spherical lenses. — Julian H.✈ 15:06, 7 March 2015 (UTC)[reply]
- I agree that it appears to be vertical blur and the most logical explanation for that is a rotation along the optical axis, but it would require an extreme rotational movement in a fairly awkward way and I'm still not convinced that's what actually happened here. This fault occurred in my 24-105mm lens fairly regularly and I am positive that I didn't rotate it, and to the best of my knowledge, it only happened when IS was enabled. It has never happened in any of my photos with any other lens. In any case, it could be that the perceived vertical blur is a red herring and not what we think it is. Diliff (talk) 15:48, 7 March 2015 (UTC)[reply]
- Hm, maybe it's extreme coma. I've never really seen coma half as strong as this, but I hadn't thought of that before and I guess that would be an optical way of achieving verticals in a part of the frame. — Julian H.✈ 17:00, 7 March 2015 (UTC)[reply]
- I've never seen motion blur resulting from a rotation around the optical axis as strong as that either... Perhaps Poco can give us some comments about it. I've tried to mention it a few times in previous nominations but he hasn't commented on it so I'm not sure what his experience of the problem is, or even if he noticed it before, given his nomination of the image in question... I suppose we should collapse this discussion though as it's quite a digression from the nomination here. Diliff (talk) 19:50, 7 March 2015 (UTC)[reply]
- Although all this is off-topic to this FPC I can just tell you that I haven't intentionally tried out any twist / rotation or similar things. My experience with that camera is that most of the time the quality was good and sometimes it wasn't. I cannot explain why, and I wouldn' call it a lemon, but is gone anywhow. Poco2 21:14, 7 March 2015 (UTC)[reply]
- Thanks Poco. Well I have to say, I would consider the lens a lemon if even one of my photos turned out like the one taken in Montenegro (if it wasn't a user issue anyway). But thanks for your point of view. I guess we won't get to the bottom of the problem since the lens is gone. I'll collapse the off-topic discussion. Sorry about that. Diliff (talk) 12:10, 8 March 2015 (UTC)[reply]
|
Info Snorkeling in the Silfra canyon, a continental drift between the tectonic plates (North American and Eurasian), Þingvellir National Park, Southern Region, Iceland. Poco2 20:21, 4 March 2015 (UTC)
Strong support -- David C. S. 21:57, 4 March 2015 (UTC)
Neutral great image below the horizon, fantastic quality, annoying clipping in the clouds. I wouldn't mind a little bit of clipping here, as it's at the very edge of the frame, but the area is quite large. — Julian H.✈ 11:30, 5 March 2015 (UTC)
Oppose --Nice but not a fantastic shot: the look gets a path at the middle but is closed at 3/4 of the way. And the lens quality of the left border is so-so (lack of sharpness). Sting (talk) 01:37, 6 March 2015 (UTC)
