fetopher wrote:
Focus distance on the hard drive was 9 feet, or thereabouts. Thanks!
Not sure you'll need the raw since mine looks similar to yours. Here it is shot on a tripod, 3-second exposure delay, processed in CS6 with all defaults. My lighting was contrastier thus the fringing.
Snapsy, that looks very similar to what I'm looking at in my sample. Perhaps with the +11 fine-tune, I'm right in line with where I should be. I can't shake the feeling that my D800 and this lens just can't produce sharp images based on my real-world session with my wife and son (which prompted this thread), but I'll get out there and shoot in the real world at +11 and see how it goes.
fetopher wrote:
Fantastic shot! Your photo was one I found before purchasing this lens and I was really wowed by it.
I tried to look at the original size on Flickr, but it looks like all I can get to is the large size (2048px wide). Any chance you could take a screen cap of the focused area at 100% and post here?
Thanks!
The following are straight out of camera, minus adding a watermark.
D800e f1.8 sharpness at 3 full size
D700 f2 sharpness at 5 full size
D700 f1.8 sharpness at 5 full size
D700 f2.2 sharpness at 5 full size
Someone please tell us because I don't know.... when the D800 focuses in Live view, does it use contrast detection off the sensor (like a D3200 does), or does it flip up the mirror and use the regular AF phase detection system (like the D700 does). If contrast detection, I would assume that would be as accurate as most people's manual focus, but is that true?
jbouchard wrote:
Someone please tell us because I don't know.... when the D800 focuses in Live view, does it use contrast detection off the sensor (like a D3200 does), or does it flip up the mirror and use the regular AF phase detection system (like the D700 does). If contrast detection, I would assume that would be as accurate as most people's manual focus, but is that true?
As I understand it, live view uses contrast detect (CDAF) where was the view finder uses phase detect (PDAF). Here's a good writeup of the differences if you're interested (I was).
But, this got me thinking, the hard drive comparison between my photo and Snapsy's photo is using live view, or CDAF. Is it possible PDAF introduces my issues? I'm going to take the same hard drive shot again, but using the view finder (PDAF) and see it it compares to the live view shot. I'll post my results soon.
Here's a test of roughly the same test as yesterday. First shot is using live view (CDAF), second shot is using view finder (PDAF). To my eye, there's quite a difference and CDAF looks sharper.
Snapsy (or others), is that what you would expect? If so, how much of a variance is there between CDAF and PDAF? I assume fine tune adjusting only affects PDAF. Is that correct?
With the D3200 I believe it says in the manual that CDAF is generally inferior to PDAF. I expect is says the same in the D800 manual.
I think you are right, the fine tuning only applies to PDAF, not CDAF. It makes sense... CDAF is looking at actual data from the sensor and trying to optimize it by getting sharp edges between pixels. Throwing in a fudge factor should only make it worse from there.
I think you owe it to yourself to try manual focus in live view. First use CDAF, then mess up the focus, and then do in manually, zoomed all the way in. There are people who use a magnifying glass to look at the LCD.
If you bought this from a place where you can return it and ask for another copy, I think it's time to do that.
jbouchard wrote:
With the D3200 I believe it says in the manual that CDAF is generally inferior to PDAF. I expect is says the same in the D800 manual.
I think you are right, the fine tuning only applies to PDAF, not CDAF. It makes sense... CDAF is looking at actual data from the sensor and trying to optimize it by getting sharp edges between pixels. Throwing in a fudge factor should only make it worse from there.
I think you owe it to yourself to try manual focus in live view. First use CDAF, then mess up the focus, and then do in manually, zoomed all the way in. There are people who use a magnifying glass to look at the LCD. ...Show more →
Here's my manual focus attempt. Not sure it's any sharper than the PDAF or CDAF shots. Also, I can't manually focus on my moving target (portraits of children, families, etc.), so even if this sample is markedly better, it does me no good. Interesting nonetheless.
Snapsy, would you care to share your thoughts on my FoCal test result chart using at the same focus point distance used to come up with +10.5 with Dot Tune? You can see the quality jumps up at +10 and +11, decreases at +12, then gets even better at +16, +18 and +20. Based on this chart, do you suppose +10.5 is an anomaly and something closer to +16 or +18 is correct, or would you toss out these results and stick with +10 or +11 based on the Dot Tune method?
I think if your 85mm F1.8 is anything like mine, you'll have a hard time calibrating it. While mine seems decently sharp wide open, it suffers from inconsistent focus in differing lighting situations. I struggle to find the right fine tune number because it always changes.
The 85mm 1.8G on my D3 is hellish sharp, it lacks pro glass micro contrast and isn't great on moving subjects.
For focus issues I try two things:
1) Download a printable focus chart, snap it wide open on a tripod. If you get loads of CA and blurry lines there's an issue.
2) Do the battery line up test, line up a series of batteries, cascade them adding a cm to each battery. Focus on one near the middle and see which one appears to be in more focus. One in front, one behind or the one you focussed on, very simple.
If your other lenses are fine then get the lens serviced, if it's part of a bigger problem get the body serviced. These types of problems drive you crazy, grrhhhhhh!!!!
I read an article once by Roger C. from LensRentals where he talked about the fact that very small manufacturing variations in the body (lens mount, etc.) and lenses can produce exceptionally sharp pairings or bad pairings. I wonder if that is at work here. Honestly, your samples don't look very good to me, and at F8 on a tripod you shouldn't see much to complain about.
My 85 1.8G is really sweet on my D300s. (a handheld pic that isn't up to this thread, but is super-sharp when looking at the file - not sure Flickr does it full justice...) LEE_2433-2.jpg by e24mpwr, on Flickr
M635_Guy wrote:
I read an article once by Roger C. from LensRentals where he talked about the fact that very small manufacturing variations in the body (lens mount, etc.) and lenses can produce exceptionally sharp pairings or bad pairings. I wonder if that is at work here. Honestly, your samples don't look very good to me, and at F8 on a tripod you shouldn't see much to complain about.
Thanks for the feedback. I've read Roger's article as well. Actually, I've read dozens of articles (perhaps 50 or more) on focusing, fine tuning, etc. At this point, I feel I have a pretty good understanding of where things could be going wrong for my D800 and this 85mm 1.8 lens. None of it seems to apply to me exactly as my results seem to be inconsistent.
I would happily accept that this copy of this lens doesn't work with my D800, except this is actually the second copy of this lens I've had. The first went back right away as it was measurably bad and could not be fine-tuned into acceptable focus. Is it possible I could have received two "bad" copies in a row? Yes, but it's highly unlikely.
This second copy is better--much better even--but I'm unsure if it's "good enough." I've fine tuned using different methods, tested it in a plethora of conditions and reviewed my photos at 100%. The results leave me wanting more. I can get a sharp image with this lens, but my "hit" rate is really low (perhaps 2%).
I find myself frustrated and fighting to find sharp photos with this lens. To give myself a break from all my testing, I popped on my Nikon 50mm f/1.8D for a few minutes: 10 shots, 8 sharp. Felt good.
Here's a photo I took yesterday with the 85mm f/1.8. It is probably the sharpest I've taken with it yet. When you consider I've taken more than 1,000 frames with this lens, it's not really the best news. All in all, I'd say my 85mm 1.8 can achieve sharp focus, but it's extremely difficult to get it there and often inconsistent.
At this point my period for evaluation is almost up. My return by date is a couple weeks, but I don't know what else I can do to test it. I left deciding if I want the focal length at the cost of a low hit rate, or simply accept my D800 cannot handle longer focal lengths due to its front focusing and return it for a refund.
Focus on subject's right eye, no sharpening applied
I realize this is a fairly old thread, but I've pointed a number of people here to show my not-so-great experiences with two copies of the Nikon 85mm 1.8G, and I need to update it with my new experience with a third copy.
If you'll recall, I posted a number of real-world shots that were lacking sharpness. I also ran the lens through Focal and DotTune testing and nothing helped me "nail" focus consistently.
Since then (about a year ago), I switched bodies (to a D810) and just in the last week purchased a third copy of the 85mm 1.8G. As you can imagine, purchasing the third copy felt like a complete shot in the dark. However, I'm happy to report this third copy is stellar! It hits focus at an acceptable rate and my focus point is sharp and can be!
Here a few samples (and 100% crops) with my third copy of the lens.
So, if anyone finds this thread and shies away from this lens based on my early experiences, hopefully my third experience and these example photos will help complete the picture.
Congratulations to your new sharp copy of the lens and new baby!
I read from the first post, looks like today's images have much better lighting, they're indeed sharp.
Noticed you removed all EXIF info from the images.
Steve Park wrote:
Congratulations to your new sharp copy of the lens and new baby!
I read from the first post, looks like today's images have much better lighting, they're indeed sharp.
Noticed you removed all EXIF info from the images.
Thanks much! EXIF should be intact on Flickr (links after each photo) for those who are interested. Most are f/1.8 or f/2.
