How do you tell you got a sharp lens?

Discussion in 'Lens Lust' started by ptchan5, Jul 7, 2007.

  1. ptchan5

    ptchan5

    47
    Apr 21, 2007
    California, USA
    I just received a 70-200mm VR I bought from another fellow cafe member and I took a few photos indoor and outdoor and I am somewhat disappointed. I've seen many photos from the crowd here using the 70-200mm and I was amazed at the sharpness and crispness of the photos.

    Here's one of my son outside. I think my 17-55 would have been sharper. I'm using the lens handheld like the rest of you guys posting your "wow" pictures.

    Is there that great a possibility that I got one of the bad ones and I need to send it back to the seller or to Nikon to have it recalibrated? How do you guys tell if you got a sharp lens, cause this lens is supposed to be the bomb?

    1/125, f5.6, ISO 560
    [​IMG]

    One more.
    1/100, f2.8, ISO 100
    [​IMG]

    100% Crop
    [​IMG]
     
    Last edited by a moderator: Jul 8, 2007
  2. ultimind

    ultimind

    990
    May 13, 2007
    Cleveland, OH
    One thing that you may want to take note of is when there are sample images posted here, Flickr, or anywhere... you need to find out whether it's a 100% crop or if it is scaled and post processed. I can make pictures from almost any lens look tack sharp if I scale them and add some unsharp masking. 100% crops is where a lens' true sharpness is shown.
     
  3. Chris101

    Chris101

    Feb 2, 2005
    Arizona
    These look sharp to me! You know that all images from digital cameras need to be sharpened. All Bayer filtered sensors have an anti-aliasing filter in fromt of the sensor to prevent jagged pixels. This filter takes some of a lens's sharpness, but a tiny bit of Unsharp Masking (or Smart Sharpening or in camera sharpening) will bring it back. Further, downsizing a photo for the web will also soften an image.

    I'll bet if you hit these two pictures with a bit of sharpening, they will jump right off the page (and if you use in camera sharpening, you can claim they are 'straight out of the camera'!)
     
  4. Those pictures look pretty good to me. As everyone mentioned, a bit of unsharp mask should do it.
     
  5. Where was the camera focusing? The first image seems to be focused on the v neck collar of the blue shirt.
     
  6. This was done in photoshop, any sharper?
    [​IMG]
     
  7. You have to post 100% crops, shrinking an image makes it sharper - notice how the images always seems sharp on your camera's LCD but not so much when you view it on your computer?

    Here's a 100% crop of a shot with my copy wide open (F2.8) on a D2h, no post processing, NEF processed with Picassa (60% jpg setting):
    [​IMG]
     
  8. yamo

    yamo

    Jun 28, 2007
    Santa Cruz, CA
    Peter,

    Greetings. So, you have a 18-200VR. Would you say that the 70-200 VR is less sharp? I would begin to wonder about the 70-200 if you think it isn't as sharp as the 18-200. (at which point I'd put them both on a tripod, VR off, and compare shots with the delayed release setting on your D80). If you still think the 18-200 is sharper, I'd have the 70-200 looked at.

    Cheers,

    -Yamo-
     
  9. Chris101

    Chris101

    Feb 2, 2005
    Arizona
    It may appear sharper on the tiny LCD, but that's because it's hard to see that small. Shrinking an image actually blurs it.

    Imagine a six by one pixel picture that is made of two white pixels, two black pixels followed by two more white pixels. The black region in the center is has very sharp edges; it's pure white, then suddenly pure black, then white again:
    W W B B W W​
    If this is then resized to five pixels wide, adjacent pixels are averaged, and the central black region becomes three pixels wide, with only the center pixel completely black, with gray pixels along the edge:
    W G B G W​
    Thus the image is less sharp as the contrast across the edges is more gradual. In a real image, it's much more complex, but the concept is the same.
     
  10. galupi20

    galupi20

    76
    Jun 12, 2006
  11. Makes sense, so why do images look sharper when they are displayed in reduced size to fit our screen and only when we blow it up to 100% view do we notice the blur/lack of sharpness?
     
  12. ptchan5

    ptchan5

    47
    Apr 21, 2007
    California, USA
    Dave, that does look sharper.

    Ok, I'm going to post some 100% crops and see what you guys think. I'm still trying to decipher the focus test chart on the link above. It's very interesting.
     
  13. Chris101

    Chris101

    Feb 2, 2005
    Arizona
    When blur is large, say 10 pixels wide, reducing it by half would result in, say, 7 pixel wide blur. More than half as much, but less than the original.
     
  14. weiran

    weiran

    966
    Jan 2, 2007
    Nottingham, UK
    While the D80 has a strong AA filter, it should still be sharper than that with the 70-200VR, perhaps its not focused the spot you've cropped at?
     
  15. Peter,

    Keep in mind that these "wow" images that you've seen, are probably the very best the poster has generated.
    We all have more deletes than keepers.

    I agree with Dave that the focus on the first picture seems to be more on the blue shirt than on your son.
    Have you checked in NX for the camera's focus point?

    If you make your test shot NEFs available for download on yousendit.com or some similar free service,
    I'm sure the group here can offer more concrete opinions as to the sharpness of your new lens.

    From what I've seen so far, I don't think there's a problem.
     
  16. ptchan5

    ptchan5

    47
    Apr 21, 2007
    California, USA
    Ok, I did what you guys said; still think there is a problem

    Ok, I put the camera in manual mode and adjusted both the 70-200VR and my 18-200VR to 200mm. I then set the camera so that I would shoot both lenses at f5.6. I used the timer so I wouldn't introduce any jitter in the capture and set the camera on a table focusing at the same spot on a plant leaf across the room.

    The center leaf was the focus point. Surprisingly both images were not spectacularly sharp, but nice when viewed at less than 100%. I then cropped at 100% the two images below. Guess what? The 18-200VR looks noticeably sharper.

    Now do you guys think I should send it back to the seller or to Nikon for calibration? What is involved in getting it serviced if I'm not the original owner?

    70-200VR, 200mm, f5.6
    [​IMG]

    18-200VR, 200mm, f5.6
    [​IMG]
     
  17. yamo

    yamo

    Jun 28, 2007
    Santa Cruz, CA
    I played around with some test shots looking how my rig (D200, 70-200VR and TC17EII) was working in regards to back focus... It turned out to be a bit trickier than I thought. I discovered that focusing on an object that didn't cover the entire area in the focus rectangle gave variable results.

    Here's the set up I used shooting from 15 feet away:
    [​IMG]

    100% crop, 340mm (200+1.7TC), 1/125th, f4.8 (f2.8 + stops for TC), VR, handheld (different shot than the above shot)
    [​IMG]

    I focused on the book. When I tried focusing on the battery the spot focus rectangle extended on either side of the battery and I got results that looked like back focus (from 13-17 on the tape measure was in focus), but on examining several shots it looked like the focus zone wondered around somewhat. Then it occured to me that since the focus rectangle wasn't completely covered by the battery, I might inadvertently actually be focusing elsewhere. Hence the book, being flat and covering the entire focus rectangle. With this setup I got consistent results and was happy to find everything was working great and the AF with both the 70-200 alone and with the TC nailed the center of the focus zone.

    You might try a similar set up. Given the images it does look like there might be a problem, but shooting at a very moveable object like leaves might give variable results with the slightest bit of wind.

    Suggest you try the test again. Shoot something with black and white print to show sharp lines that is unlikely to be moved by the wind. Be sure the VR is off (shooting from resting on a table is kinda like a tripod). There is a setting on the D80 for a delayed shutter release (not the timer)... it snaps the mirror then after a delay (1/2 second?) it releases the shutter. Shoot with plenty of light on the focus subject (your test shots have the leaves back lit).

    Test shots are a bit tricky and one can be fooled by the results if they aren't set up properly. When it's all said and done, you might need to send the lens to have it checked out.

    Good luck. The 70-200VR is an outstanding lens. You shouldn't have to work so hard to get the performance... so maybe there is some sort of problem.

    Cheers,

    -Yamo-
     
  18. weiran

    weiran

    966
    Jan 2, 2007
    Nottingham, UK
    That 70-200VR shot looks out of focus to me, rather than lack of sharpness. You don't say how far the subject is from the lens, the 70-200 is also optimised for distance shooting and will probably get softer towards the extreme short end of its focus.

    It could also be front or back focusing, in which case the best test would be to focus on a ruler at a certain point and see how far off the focus is. Also try a better test subject, something with small text like a newspaper is good as it shows off the contrast (black text on white paper) and also the fine detail.
     
  19. My Lens Testing--More Than You Probably Want to Know

    Peter,

    I think your plant target is causing the out of focus test shot with your 70-200vr. You might want to try other test targets before returning the lens or sending it in to Nikon.

    FWIW, I shoot staggered boxes (3 identical ones) at wide open aperture in NEF format with no in camera sharpening. Focus will be on the middle box. This will help reveal any back- or front-focus, and give a general sense of how sharp the lens is at its focus point. In Capture, I will go through the various sharpening levels to see at what point the image is acceptably sharp.

    All test shots are done on tripod with remote release. For long telephoto lenses, use mirror lockup if your camera has it. Try to have the lens at the same height as the center of your focus target and directly in line with the target. You want to avoid any up/down or left/right angle shooting. The plane of the front lens glass and the target should be parallel.

    I'll repeat the shots at around f/4, f/5.6, f/8, f/11, etc. to see how the lens performs at apertures I’m likely to use and to check that there is improvement as the lens is stopped down. It helps me identify the sweetspots for a lens.

    I normally take my test shots just beyond the minimum focus distance of each lens, so that the test target appears as large as possible in the image. Still, there is the possibility that your lens is fine at shorter distances, but is problematic at longer distances. I had a 200vr that was fine up close but had noticeable backfocus at longer distances. A quick visit to Nikon Service remedied that problem. However, I can only be so anal and would only test at a longer subject distance if I noticed a problem in actual shooting.

    I shoot a similar NEF wide open of a newspaper taped to the wall. This should help reveal if there is any distortion or varying sharpness in edges, corners, side-to-side, etc. Shoot at close to the minimum focus distance for your lens and try and keep the plane of the front glass perfectly parallel to the newspaper. Shoot at the same height level as well. This should reveal any variations in sharpness or distortion in different areas of the image.

    Here are some sample focus test shots using 2 different size boxes --staggered cassette boxes with newsprinted taped to them, and staggered bread crumb containers.

    70-200vr at 200mm & f/2.8:
    [​IMG]

    70-200vr at 200mm & f/2.8 (further away):
    [​IMG]

    70-200vr at 200mm & f/2.8 (closer):
    [​IMG]
     
  20. Peter,
    Have you spoken to the original owner? Have you seen any photos he/she took with this lens? Did they have any problems and was this lens ever serviced by Nikon? Any time left on the 5year warranty?

    I went through the same exact problem with this lens which I purchased from someone here at the Cafe. I was never impressed with this lens and chalked it up to operator error. I went so far as to do some of Tosh's testing and everyone here agreed it was within expected tolerances.

    I wound up keeping the lens but was never happy with how soft it was. I sent it to Nikon to have it checked. The lens underwent a "major" repair which cost me $540!:eek: It is now sharp and functions perfectly. I've now spent more for this lens than I would have for a new one. You may want to return yours if you still can.
     
Loading...
Similar Threads Forum Date
Got My 300mm Today!!! Lens Lust Apr 1, 2015
How to tell a U.S. or Gray 180 F/2.8? Lens Lust Dec 8, 2014
50mm and 30mm. What can you tell me about options? Lens Lust Apr 19, 2013
how can you tell if a 180/2.8 is a USA model Lens Lust Jan 14, 2012
Looking for a wide lens.. Tell me and show me some proof! Lens Lust Oct 24, 2011