As some of you know I just got a brand new d7100 after drowning my d90. I shot a few images around the house, indoors, but couldn't get out and take it for a good spin until today. Got up early, headed out to a favorite spot to catch the morning light. It's a small pond and when we got there it was still dark.
As the sun came up I shot the tree line over the lake with my Sigma 10mm-20mm. It's an older HSM lens and I had emailed Sigma to learn that it would work fine with the d7100 but would not auto focus in "Live View" (that's what Nikon calls shooting with the LCD). I never compose a shot that way, so no concerns. The camera indicated a good focus. Both the green focus indicator and the little red bounding box in the view finder indicated good focus and appropriate focus points.
When I got home I was disappointed to see that the images where very soft... all of them. So, I hurried out into the yard and shot a number of shots both the "suspect lens" and a number of Nikon lens. One a prime lens I have always leaned on for my sharpest landscape images a Nikon 35mm f1.8. Some shots are focused, some are not. It would seem the camera focuses well on objects near by, but struggles with distant objects. Regardless of which lens I choose, focus is hit or miss.
I have contacted Nikon through their website and wait to hear what they have to say. In the mean time, I thought some of you might have advise:
How would I go about troubleshooting this issue? Is there a "scientific" method for eliminating variables when it comes to focus problems?
I have manually selected the point of focus, the focus light comes on, but no joy when I view the image on the computer. Any help thinking this out would be appreciated. I would like to work on this issue while I wait to hear from Nikon.
I will include one image with this posting. There are also examples of this lens performing perfectly with D90 if that will help narrow down the problem. Aside from the focus issue you will also note a large piece of crap on the sensor (egad!). Nikon's sensor cleaning mode took care of it.
Susan just wrote on another thread regarding this, I will post her comments here:
I don't usually let the camera choose my focus points and I rarely use full Auto. I usually use the continuous AF mode and then select from the 9-51 point setting or Single. From what I've read somewhere, the 51 points take longer to focus. I rely more on the aperture and use 11 or below (larger number) to avoid diffraction. The "good focus" light isn't a perfect judge on any camera, IMO.
Were you using Matrix or Center-weighted metering today?
Also, if using a zoom, I always try to avoid the absolute longest and shortest focal lengths on any zoom lens since under less than ideal conditions those can result in a softer photo.
When taking landscape shots, I use the live view and zoom in during live to fine tune the focus manually. I realize that this is now what you want to do, but with so many focus points available on the D7100, you might be best narrowing down what you want to hold in focus for the camera.
I do not own a D7100 yet, but I have used a friend's D7100 and it focuses great with my Nikon 35mm 1.8, so I am curious to find out what you come up with.
Initially I was letting the camera choose focus points. After I got home and saw the results I configured auto focus to work more like what I had been used to on with my D90. AF-S Matrix metering. I normally shoot landscape and macro shots so this worked well for me with the D90. I thought it might help troubleshot the problem.
Regardless of which mode I used, poor focus.
I understand that auto focus isn't perfect, and the light isn't always truthful, but I never had a problem like this with the D90.
David, that's a good point. And I will do just that.
So, for starters I will set up the 7100 on a tripod with the 35mm and see if I get a good focus. I will zoom in using Live View and see how auto focus did. For some reason it hadn't dawned on me use the zoom feature in live view. Like I said, I just don't use the LCD that much. And as I mentioned to Susan, under the same conditions this would not be happening with the D90.
Sadly I'm loosing light at the moment, so this experiment will have to wait until tomorrow. I would do some indoor stuff, but this problem only seems to occur with distant objects.
David, I've never used the live view on either the D90 or the D7100 except for videos, but I'll try that the next time I do a landscape. I've read that using Live View for a long time can overheat the camera. Have you heard anything about that? The Nikon lenses I have allow me to switch quickly from AF to focus by hand.
J Scott, I do see underexposure in your photo. What exposure settings did you use?
Also, under Picture Control, I like the Landscape setting outdoors to help give a punch to outdoor colors. It's not as strong as Vivid, but a lot better than Standard.
The D90 was only 12MP, half of the D7100. Smaller images always look more clear and hide focus/exposure errors - just like a camera's LCD screen image looks so great sometimes until we zoom in on it! :))
EDIT: J Scott - Again, I highly recommend David Busch's Nikon D7100 guide. It even has a focus test chart under "Evaluating Current Focus." The book came out just recently.
Susan, Yup, the image may be a tad underexposed. I hadn't worked with the RAW yet. A moment in ACR and all would be well (regarding exposure). I find underexposed images yield better color saturation once processed. Given that I wasn't seeing a gob of color on this particular morning I leaned in that direction deliberately. That being said, it wouldn't have been a very good image regardless, I posted it for troubleshooting purposes only.
I always shoot manually (exposure). I use as little electronic hindrance as I can. Autofocus on the other hand is something different. With wider angles it made (note past tense) focusing much quicker. As I mentioned I have used this lens, under far more challenging conditions (dim lighting being difficult for many auto focus systems) and it always focused nicely. I have attached a image taken with the same lens on the D90.
Yes, it is true, a smaller image will be less likely to show small focus issues. But this is a large focus issue occurring with seven lens, Nikon and Sigma. Even magnified %100, images taken with the same lens/lenses and the D90 are acceptably sharp.
I have David Busch's book for the D90 and liked it. That focus chart would be helpful now. I'll look for it, thanks!
I have been using my Nikon D7100 for the last 5 months with Nikon lenses and have not experienced any focusing problems. Only once or twice so far when I accidentally bumped the focus switch on the side of the lens from auto to manual, and did not realize it.
Other than that, I have been very pleased with the camera and Nikon lenses.
Susan hit on an important point. The increased resolution. When you double the number of megapixels, you are doubling the resolution on the 100% zoom view of the image. When I upgraded from the D60 (10mp) to the D7000 (16mp), I found it much more difficult to get a sharp, handheld shot. After some time and practice, my technique got better and I was able to get sharp handheld shots. Then, I upgraded to the D800 (36mp) and again I had the same problem. Again, I had to hone my skills in order to produce sharp images with the higher resolution camera. It's possible that the autofocus is not working properly, yet it's also possible that a small amount of camera shake is causing the image to be blurry--an amount that was not noticeable with the lower resolution camera. Testing with a tripod and remote release will either confirm or eliminate this as the issue. I've also found that manually choosing a single focus point is much more accurate that allowing the camera to choose. :-)
I have no doubt that the camera is fine. Nikon has always been great anyhow so I'm not worried (much).
It may be that I am noticing things at 24mp that I didn't see at 12 (as Susan suggested). If that is the case, WOW! I would have never believed it! In the case of the Sigma 10-20 it would be the difference between tolerably sharp and unusable. It's third party glass, maybe a matter of you get what you pay for. But the Nikon 35mm has a good reputation and I have scores of tack sharp images taken with the old body.
So not worried, just a little irritated and somewhat mystified.
I was hoping someone might have some ideas for a down and dirty testing regimen I could occupy myself with until I hear from Nikon. I will report back again tomorrow after I sort it out.
Okay, so that eliminates camera shake. Long exposures can be tricky. Was there a breeze? If not, then I would look at the differences between multiple focus points vs. single focus point, combined with aperture. When I owned the Sigma 10-20, I did notice that it was much sharper when used at f8 and higher. Also, sharper in the "middle" focal range. (i.e. not all the way wide or all the way zoomed) But, I didn't have any major issues with that lens and for the most part, I was very pleased with the image quality.
Loree, would you please share how you honed your skills for sharper hand-held shots? When I use a heavier lens, I can sometimes feel my arms begin to "protest!" And I can't take many of my images with a tripod! :))
The Sigma 10-20 is known for its softness... I hope its not a problem with the high-pass filter in the Nikon - as there have been some in the past!
Personally for wide-open angles I use the Nikkor 14-24 - as good as any prime lens...
The reason for switching VR off when using a tripod is because VR actually "jiggles" the camera to counteract the movement of the photographer. When there is no movement, the VR can cause softness because it is moving the camera slightly.
As far as getting better at handheld shots, it is mostly a matter of practice, which it sounds like you get a lot of anyway! :-) No big secrets, just practicing the known techniques of rolling the finger over the shutter instead of pressing or squeezing it, taking a deep breath and shooting on the exhale, making sure the shutter speed is fast enough for the focal length, putting one foot slightly in front of the other instead of side by side, finding something to lean on if possible, etc. Also, if I am shooting handheld, I use burst mode and shoot 4 or 5 shots. That way, at least one of them is usually sharp.
As Susan and Loree already noted, increasing the megapixel resolution has a distinct affect, I noticed that immediately when I went from a 12MP K-x to a 16MP K30. It took me some practice to get better focusing accuracy, and I use all manual so that makes a difference too.
I also went through the camera settings and increased the in camera sharpening, that helped.
What you might need to do is take some controlled shots at different distances, with a tripod, (and shake reduction turned off) and see how it does at close, medium and long range. I like to use something like a coffee can with some text printed on it so I can see what the text looks like. I do this every time I get a new lens to see how it performs.
I'm not sure how Nikon does their shake reduction, or image stabilization, Pentax builds it into the body rather than the lens, so all it does is move the sensor in response to the camera movement. If Nikon has it in the lens it's probably moving something inside the lens, not the camera itself but I may be wrong. I don't see how it could move the camera though...if it's built into the lens.
The only other thing I can think of is lots of practice. That's mostly what it took before I got accustomed to the higher megapixel camera and started getting consistently better focus. I don't know how your viewfinder is, but mine on the K30 is great, loads better than the K-x, but I still have to really look it over close to be sure I'm getting good focus with the higher MP camera. I keep mine set on center spot focus, center weighted and in manual focus mode it still gives me a focus confirmation beep and LED, which I use extensively, and it's very accurate. With the one auto focus lens I have, the kit lens, I still use it in those settings, and it works nicely. I found that trying to use the 5, 9 and 11 points it did too much searching, so I went to one center spot and center weighted. For what I do I don't need those anyway.
Susan - to get better stabilization especially with heavier lenses, it ain't easy, but I still seem to have pretty steady hands to begin with, and if I'm worried about camera shake I'll try to keep my elbows in against my sides, or use something like a tree or fence post for extra bracing. I also use the middle section of my finger to squeeze the shutter rather than my fingertip. Similar to what Loree said. Right behind the first knuckle. That seems to get a better squeeze of the trigger rather than a jerky motion like when I use the fingertip. I had problems with that when I first got into DSLR, and I find it more comfortable now too. And same as to the original poster, the main thing is lots of practice...You can also try holding your hand under the lens, rather than from the side, with your thumb on the left, as if you're handing someone a hot dog, and resting your elbow on your stomach, that may help but I always found it uncomfortable. That's the way my photography books said was the "correct" way to hold a lens, I never liked it. I keep my thumb on bottom, as if I'm drinking a soda. That's wrong according to the books...but I could never get comfortable with the other way.
I don't own your specific body/lens combination but I am a professional Nikon user.
I cannot recall what your aperture setting was for your landscape photographs (if I missed my apologies) but I am thinking since you are using a quality UWA 10-20mm lens the actual issue may be a common lens miscue known as DIFFRACTION. BTW, both Sigma and Tokina make pro quality lenses that work terrific with Nikon & Canon bodies.
I'm going to assume you used the minimum aperture setting most likely F/22? (If I'm wrong-disregard!) Your D7100 is a cropped sensor - and not an FX full format. Since by nature "ultra wides" UWA lenses provide incredible DOF even at wide F-stops such as F/5.6 and F/8 it is not necessary to stop way down to F/20 or F/22. It's nearly impossible to even create a nice bokeh effect with an ultra wide angle lens even at F/4. It has been accurately pointed out previously you have gone to an extreme high resolution sensor which will exaggerate any minor flaws.
The actual aperture diameter at F/20 or F/22 is awfully tiny. (cropped sensor) That is the cause for diffraction, or lack of apparent sharpness. I would try to experiment with similar landscape compositions and step your F stop up to F/16 or even 1 step wider!! Try to look up some lens reviews, perhaps by Thom Hogan or DP online for that specific Sigma lens and research their findings for that lens' ultimate sharpness "sweet spot" .
Loree - thanks very much for the info on VR and your other tips. I like using the bursts as well, especially with anything moving!
I became curious about other reasons NOT to use VR (in addition to the tripod use) and saw Thom Hogan's article http://www.bythom.com/nikon-vr.htm The following heading caught my eye: "Rule #2: VR should normally be off if your shutter speed is over 1/500." Of course, "it depends on...." scenarios follow that statement!
However, information from Nikon itself about VR at http://www.nikonusa.com/en_US/IMG/Images/Micro-Sites/VR/index.htm doesn't mention anything about VR and high shutter speeds. It does have good info about VR in general and also gives an overview of when to use the Normal/Active setting that is a separate switch below the VR switch on some Nikon lenses. I have it on my 70-300 VR II.
Anyone have any experience with VR and high shutter speeds?
EDIT: The Thom Hogan article link above is now correct. I had accidentally put a period at the end of it when I originally wrote it.
First off, WOW, thank you all so much for coming to the rescue. I spent another day with new 7100 and it is true, my issues are due in large part to the increased resaloution. I am seeing things at 24mp that I did not at 12. I would have never believed it would make such a huge difference!
I will have to sharpen my skill set and pay more attention to details than I did.
@ Susan, thank you for being the first to point this out!
@ Jack, the reviews I read respecting the Sigma 10-20 were mixed, many complaints were resolved with an exchange. I decided to take a chance, and at 12mp was very happy. Today, taking more care with camera movement, scrutinizing focus as David suggested last night, and choosing my aperture more carefully I was happier with the Sigma then I was yesterday. True it's not tack sharp, but it was substantially cheaper than the Nikon 12-24.
@Dean, excellent suggestion! I will look and see how far the nearest Nikon shop is! Regardless of how this all works out it would be good to know there was a place close by that could help with the right tools and knowledge.
@Billy, thank you. I have been looking over the differences between my settings on the old camera and the new one. I will fool with the in camera sharpening. I noted today that even with the tripod, that images were sharper at faster shutter speeds. I think that would mean that even on the tripod there is some camera vibration occurring, this surprised me as I had never noticed this before.. I did turn Vibration Reduction off btw
@Thomas, you are right. Here again, this something I had never noticed at 12mp! f22 is not as sharp as f14 and as you mentioned there is ample DOF without stopping down that far.
@Paul, WOW, that is very helpful. I like this chart a great deal. This will give me good settings to start with and help me consider settings I didn't know about (there is a lot more going on with 7100 then there was with 90).
Today was much more successful. I am still learning the new camera and my keeper rate wasn't as high as it was with the D90 but I am convinced there is nothing wrong with the camera. I just have to be much more exacting. It is exciting to think that there will be so much improvement in my work once I buckle down and acclimate to my new camera. I am posting a few images today. I'll come back after the fact and add a few images to this topic. Nothing special, but proof that the camera works well!
Happy to see you're getting it sorted out. When I got my D7000, at first I thought there was something wrong with it, too! So, when I got the D800, I kind of expected a learning curve. You will master it in no time!
Susan, Thank you... I had missed that link (I had a lot of responses to catch up on). Great info, well assembled! For me it's remembering to turn it on. I almost always shoot on tripod so VR is almost always off. Now I will be paying more attention to it and using it more often.
Thomas made one of the most important points yet, one I forgot...I never shoot smaller aperture than f16, usually f8 and occasionally f11 if I have enough light and that's why. f22 is usually not as sharp with most lenses, f8 to f11 is the best range in general, and with lenses that will get really wide, like my 50mm f1.4, I get excellent results at f5.6 and sometimes f4. And yeah, f22 is tiny, about the size of a pinhead. I have several older manual lenses so I can stop them down off the camera and see it, f1.4 is wide open, as big as the inside of the lens basically, and f22 is pinhead...I have one lens that does f32, not sure if I've ever looked but that's really tiny...
Anyway try to keep the aperture around f8 to f11, you'll most likely get the sharpest shots in that range. If you have any lenses that open up to f2 or larger, they may still be good in the f4 to f5.6 range too.
:) I'm glad this is working out for you J Scott. No matter enthusiast or pro, trying to diagnose any unusual issues with separate components like camera bodies/lenses can cause an anxious moment or 3. I'm sure Thom Hogan or Digital preview (online magazine) has figured out the sweet spots for all your lenses for maximum sharpness.
Full frame FX bodies don't encounter this issue (diffraction) however they deliver buttery smooooooth bokeh :)
This is all good info for me too. I just got a D7100 (from D76s) and once I got over the CMOS shock I am noticing some odd things about focusing, or I think maybe the VR thing. Plus I got a used lens from a fellow club member and I'm disappointed with it, or with the seller - long story there but I believe I need to go back to my trusty old 18-70 for the testing phase and learning curve.
@Kathi, I am glad this is helping you. Looks like you're in the right place! ;-)
@ Susan, I can not find much respecting VR and why you should turn it off while on a pod. But both Sigma and Nikon mention it. However, I don't think it featured in this problem.
In doing further research, on Nikon's site and others I learned that most high end DSLR manufactures acknowledge auto focus issues. Nikon has offered to examine my photos and help me diagnose this issue further but after reading about AF Fine Tune (located on the settings menu of the D7100) I feel much more confident that these smaller focus issues can be addressed here. There are a number of sites offering advise and printable charts to test your AF system with each lens in your arsenal. The D7100 will keep track of each lens you "tweak" and apply these little AF adjustments each time that particular lens is used. If I remember correctly it can store the focusing adjustments made for ten lenses. I don't think that option would be available if there wasn't an issue to address.
It would seem, on quick review yesterday, that the Sigma 10-20 is front focusing. I will shoot a series of test images today, and make the adjustment if necessary.
What is most important:
After seeing the results while shooting through that one lens (10-20) my good judgement regarding the other lenses was chucked in the basket. I began to chase a problem that seemed to affect all of my lens when in fact it was one lens that had a problem. That made troubleshooting the issue much more difficult. With twice the resolution, images taken with the D7100 will emphasize some of the negative issues in your images. Both with your equipment and technique. I was not prepared for this.
I had never noticed things like camera movement... on a tripod... caused by the wind! And though I had read that the smallest aperture on any one lens may not in fact be the sharpest, I had never seen it my own images until the jump in resolution. While shooting at the moon last night I took photos using mirror lock up. This is a feature I am enjoying for the first time, but never thought it would yield a noticeable improvement in image quality... WRONG! With a 500mm, a relatively slow shutter speed the mirror was causing enough movement in the camera to be noticed.... WOW!
I am energized by the challenge of "upping my game" to match the new tool in my bag! I'll be back with more as it becomes relevant.
EDIT: I just received an email from Sigma regarding my issues with the 10-20. Just one sentence, and in it's brevity... clarity. "Your Nikon D7100 has an auto focus fine tune adjustment". That was it... short and to the point. I'll let you all know how it goes.
Auto focus...OK it starts to make more sense now. I almost never use auto focus, when I first got my Pentax K-x, I noticed I got better results using manual focus with the auto capable kit lens, started doing some research and found that virtually every camera manufacturer out there had less than stellar results with auto focus. That's one of the biggest complaints it seems, on the Pentax forums, auto focus just ain't what it's cracked up to be. And it's mostly useless for what I do, especially birds in flight and macro.
I was pretty sure Nikon should have focus adjustments, I know Canon does because I ran across a couple on Louisiana trying to adjust their back focusing EOS. Pentax has that function too but I've never needed it, both my kit lenses do pretty well, and it's not an issue with manual lenses.
I think that may be the biggest part of your problem now that you bring that issue up, I don't remember but I think my K30 has the option to set values for around 6 or 8 different lenses, your Nikon should be able to do something similar. OK I just scrolled up and saw you said 10 lenses, that's even better. One thing I have thought about tinkering with though, and you might too if your Nikon has it, the K30 has 2 user defined settings, so you can set everything specifically for one setup and just turn the dial to that setup for that particular lens. That means not only your focus fine tuning, but focal length, ISO, image saturation, contrast, sharpness, white balance, even black and white, basically everything for one specific purpose. SO if you have one lens you like for landscapes, you can set everything the way you want it, save all that to the user setting and when you shoot landscapes all you have to do it dial in that on the "Mode" dial. Same dial that selects Manual, Auto, Aperture or Shutter priority...
I've never tried the mirror lockup function, need to play with it some too I guess because the mirror does cause a minor amount of camera shake. That's the biggest selling point of mirrorless cameras. I always use the 2 or 12 second times when shooting moon or star shots on a tripod, which seems to work pretty well, but mirror lockup sounds good too. My Minolta film camera has that but it's so easy to trip it by mistake it's a nuisance. Every time I turn around I look through the viewfinder and see nothing. Took me months to finally remember to check that little switch...
Billy, that's a good idea, setting things up for THAT lens and it's uses. There are a number of customizable menus that would work for just that. It would help my results be more consistent. I have always appreciated your birds in flight. I find it very difficult to capture moving objects with worthwhile results. My portfolio speaks to this, not one moving object in the mix! I have even greater respect for your work now that I know you are focusing as you pan.
Perhaps I rely on auto focus (or more accurately the "in focus indicator") too much. My favorite camera of all time was a simple Pentax K1000. It focused using a prism I think, there was split image in the view finder you lined up to assure focus was correct. It worked flawlessly! Today I use the electronic focus aids to do the same. For me, with a 10mm lens it is difficult to judge focus by eye, the objects in the viewfinder are simply too small. If it weren't for the little green light that came on to tell me I was focused... I probably wouldn't be.
IMPORTANT: I just did the aforementioned test (auto focus fine tune) and the suspect lens was back focusing. After tweaking the adjustment I made substantial improvements judging from the test images I took of the chart. Now out to see what this means in the real world.
J Scott, I don't know if it's been mentioned above, but when using a tripod, you can use the little Nikon ML-L3 remote control (around $20) to further avoid any camera movement.
Re turning VR off on a tripod:
In its manual, Nikon just says turn it off while using the tripod - no explanation. I really appreciate Loree's non-technical explanation above: "The reason for switching VR off when using a tripod is because VR actually "jiggles" the camera to counteract the movement of the photographer. When there is no movement, the VR can cause softness because it is moving the camera slightly."
Re fine-tuning lenses yourself:
I've read that unless you really know what you're doing, " fixing" focus by yourself for one type of focus issue can create a focus issue at another focal length...?
Does anyone here send lenses in to Nikon/Canon/etc. for regular "tune-ups?"
I didn't find the adjustment to be at intimidating and the results are spot on.
In short, and to put this thread to rest: the D7100 back focused with the Sigam 10-20 Lens. That was the problem that started this thread. The process of making this adjustment is much easier than taking a good photo. If you have a focus problem with some of your lens and your camera has a auto focus fine tune feature like this one, I highly recommend the procedure for both it's easy implementation and it's fantastic results.