Sensor in the D80/D200

[michael_pye]

Hi,

Is the sensor on the D200 alot better than the D80, so if I used the same lens

on both cameras the results on the D200 would be alot sharper e.t.c. because

the sensor is far superior and I guess collects more information, even though

it's the same MP size as the D80?

 

Thanks.

[Matt Laur]

Essentially the same. What most people perceive to be differences in the results are usually the result of in-camera image processing as the cameras produce JPG files directly. If you shoot in RAW, and then do the same things in post production, you'll get very similar results. Now, all of that being said, there are substantial differences in the experience of using each camera, and the D200's quicker performance may make the difference in getting, or not, a given shot (especially shooting sports, etc). But sharper, per se? Not in a nice, controlled, side-by-side test, not so much.

[mawz]

The D200's sensor is faster than the D80's, but is susceptible to some issues that the D80's sensor is not (Banding if the readouts are unbalanced, as it uses 4 channel readouts rather than the 2 channel design in the D80 and D40X). Also the sensor runs a bit hotter, which leads to more high ISO noise.

 

RAW results should be almost identical. The D80 has better in-camera processing and will produce marginally better JPEG's (The D40X produces even better JPEG's). But the differences are quite minor except at ISO 1600 or 3200 where the D80 and D40X's in-camera processing advantage is more noticable.

 

The D200 however is a higher performance body which will get you the shot in conditions where the D80 and D40X will fail you.

[tonybeach]

The D200 sensor collects more data in that it will allow shooting at 5 fps versus 3 fps. The D200 as a camera also delivers more data because it has an option for uncompressed RAW while the D80 does not (12 bits versus approximately 10 bits difference). Also, the D200 has the potential to deliver more data because of its larger buffer. It is arguable that increased noise and banding are also "more information"; naturally that's the kind of data we can do without -- but the differences in noise are negligible and banding is only infrequently encountered (I only see it where the brightest regions of the scene have been severely overexposed and are adjacent to darker regions).

[mawz]

@Anthony Beach: The only difference in the actual data delivered is the Uncompressed RAW vs Compressed RAW issue, as Nikon's compressed NEF format is lossy so id does deliver less data than uncompressed RAW (or the new lossless compressed raw on the D300 and D3). The D200 delivers no extra date per shot off the sensor to the processing than the D80 does. It simply delivers that data faster. No matter what, you get 12 bits of data per pixel with the exact same number of pixels delivered per shot as the D80 does. The buffer size is irrelevant to the amount of usable data, as is the shooting rate (they simply determine how many shots you have).

 

And noise and banding result in LESS information delivered, due to worsening the signal/noise ratio. noise is not information, it degrades the information you do receive.

[tonybeach]

For Adam:

 

"No matter what, you get 12 bits of data per pixel with the exact same number of pixels delivered per shot as the D80 does."

 

No, it is more like 10 bits of luminance data per pixel (pre-interpolation) when it is compressed. That is why the files are smaller.

 

"noise is not information, it degrades the information you do receive."

 

Noise is data, just not useful data. Take a file and add noise to it and it will be larger than the original; conversely, the file will be smaller than the original if you reduce the noise.