Your thoughts?

I took my new D7100 to the Talladega superspeedway last weekend. I had just upgraded from a D90. Overall I was pleased with the results. However, at extreme zoom (200mm) with my NIKKOR AF-S 18-200mm 1:3.5 -5.6 GII ED DX with VR there is a definite degradation of sharpness when the center of the picture is compared to the 4 corners. I did not notice this with the D90, but then again my expectations were diminished with the D90 vs. the D7100. Perhaps I’m expecting too much? This is why I’m soliciting your thoughts about the picture. When you download the picture (warning 12.5 MB) right click on it and then click details to see the camera’s settings the moment the picture was taken. Basically it was auto everything. Of course for photographing the cars at speed I changed the shutter speed to 1/8000.

http://www.davidjohnsonpage.com/D7100.JPG

kind of hard to tell with this photo. different focal plane on the corners.

Also this is the first picture at Talladega where it was not a girl flashing the crowd.

Higher resolution cameras do tend to show weaknesses in lenses more than lower resolution cameras. And the 18-200 is a decent lens I'm sure but maybe not up to the standards needed for the D7100.

Tracsoft - Don Kuykendall said:

kind of hard to tell with this photo. different focal plane on the corners.

Also this is the first picture at Talladega where it was not a girl flashing the crowd.

Click to expand...

I cropped her out. Plus she was near a corner and blurry.

At 1/640th of a second, VR shouldn't be turned on. All that will do is help blur the image as the lens is trying to compensate for shake that just is not happening at that fast a speed. Over 1/250th, you should just turn it off (in my experience anyway). Try again without VR, assuming you can get a quick shutter speed and that will tell you if you have an issue with that crazy sensor out-doing the glass you have.

Shooting even a race car at 1/8000th seems a bit overkill. Unless you were in mid-day sun, shooting at f/1.4 or something.

Also, where is that picture of the girl flashing the crowd ??

eurotrash said:

Also, where is that picture of the girl flashing the crowd ??

Click to expand...

LOL! Don siad, this is the first picture at Talladega where it was NOT a girl flashing the crowd.

eurotrash said:

At 1/640th of a second, VR shouldn't be turned on. All that will do is help blur the image as the lens is trying to compensate for shake that just is not happening at that fast a speed. Over 1/250th, you should just turn it off (in my experience anyway). Try again without VR, assuming you can get a quick shutter speed and that will tell you if you have an issue with that crazy sensor out-doing the glass you have.

Shooting even a race car at 1/8000th seems a bit overkill. Unless you were in mid-day sun, shooting at f/1.4 or something.

Click to expand...

No complaints at all with the shutter speed. Stop action at 190 miles per hour plus with 1/8000 was better than expected! In both bright light and heavy overcast.

certain people claim that having vr on either when on a tripos or shooting at high enough shutter speed theoretically can impact the picture in a negative way.

Take a pic with the d90 and the 7100 of the same object and view at the same size on screen to compare. That will. Highlight the difference.

Geoffc said:

Take a pic with the d90 and the 7100 of the same object and view at the same size on screen to compare. That will. Highlight the difference.

Click to expand...

I wish I had that option. I sold the D90 before I bought the D7100. Of course I have 1000s of D90 pictures with the same lens as well as at 200mm zoom. In comparing 200mm zoomed pictures I see less degradation from center to edge with the D90, but the D7100 is a better picture overall.

OK, well then take a picture of the same object at a high shutter speed with a tripod both with vr on and off. Also do this handheld. Mebby that will showcase weather it is a lens problem or a camera issue or something entirely different.

The improved sensor in the 7100 may very well be showing you flaws you may not otherwise have seen with older technology.

eurotrash said:

OK, well then take a picture of the same object at a high shutter speed with a tripod both with vr on and off. Also do this handheld. Mebby that will showcase weather it is a lens problem or a camera issue or something entirely different.

The improved sensor in the 7100 may very well be showing you flaws you may not otherwise have seen with older technology.

Click to expand...

Sounds reasonable as well as prudent. I'll do that first chance I get.

You've got twice as many pixels, so there's a good chance that you're seeing stuff that was masked before. I'm seeing some chromatic aberation, which would not be impacted by the VR, but it's also not unexpected at extremes on a zoom. If you're shooting RAW then applying lens profiling in Lightroom or Adobe Camera RAW may clean some of that up. Not sure how much you believe in DxOMark scores, but what I'm seeing here isn't out of line with what they measured with the lens on a D300. It scored only slightly better on a D90.

DxOMark - Nikon AF-S DX VR Zoom-Nikkor 18-200mm f/3.5-5.6G IF-ED

Just an update here to this thread. I downloaded the free trial version of DXO Optics 8 and was very impressed with the output. DXO O8 corrected most of the issues I had with the photo. And some I didn't even know where there! With my former D90 I used to use Photoshop Elements to correct a lot of issues, but it was not as effective with my D7100. As of today I'll be using DXO O8 with my D7100 photos. I'm not sure I believe in the so called specific modules for my camera and lens, but I like knowing they're there anyway.