Kodak DCS Pro 14n Digital SLR Kodak's latest digital SLR brings full-frame, 13.7 megapixel resolution to market for under $5,000. (Review first posted 3/23/2003)

Page: "Davebox" Analysis

Basic color/tonal rendition - The "Davebox" target

Pro 14n	EOS-1Ds
Given the way the two cameras performed in some of my other test shots, I was surprised to see the 1Ds' colors come out as saturated here as they did. - It was looking like the 14n generally produced more saturated colors, but the 1Ds actually comes out with better saturation here in many instances. - But I still felt that the 14n's color looked a bit more natural on many real-world subjects: Just goes to show you can't judge everything from a test chart, I guess.

While the 14n I tested was still in its prototype state, the word from Kodak was that additional firmware changes probably wouldn't affect image quality significantly. While I'll reserve final judgement until I can test a full production model, I took the time to analyze the noise levels from both cameras, not only in the neutrals, but in the additive primary colors as well. The results are displayed in the table below. (NOTE that the 14n image used in this test had been processed with the noise reduction adjustment in Photo Desk set to "normal." The noise could be a good bit higher or lower than that shown here if different noise-reduction settings were used. Values higher than "normal" result in severe flattening of detail though, so I think the figures below could be considered to represent the best case you'd find in typical practice.)

In the chart above, the values for the 14n are at reflected in the top bar of each pair, while those for the 1Ds are shown in the lower one. What's interesting here is that the 14n shows less noise than the 1Ds in neutrals (shades of gray), but more noise in colors. The noise level is also rather color-dependent, really skyrocketing in the deep blue swatch of the MacBeth chart. I saw similar effects in the Sigma SD9 when I tested it - Good noise performance in neutrals, but much higher levels in certain colors. (This underscores the need to evaluate noise across the spectrum when evaluating digicam image quality.)

Although noise in the 14n is a strong function of color and saturation, it is nonetheless instructive to see how it varies across the camera's ISO range. The chart above shows the results of this analysis, once again using an image that had been processed with the "normal" noise reduction setting.

Here, the 14n starts out ahead of the 1Ds at ISO 100, but the 1Ds draws even at ISO 200, and pulls ahead at ISO 400. (It would be interesting to see how the two cameras compared across ISO values for different colors, but I just don't have the time to indulge in that deep an analysis.

For those of you interested in making your own analysis of the Davebox target with the two cameras, here are a set of links to all the relevant sample image files:

Lowlight Performance
Well, in a word, it's pretty miserable. While newer firmware might possibly improve the situation somewhat, the 14n clearly doesn't look like it will be anyone's first choice for after-dark photography. The links below connect to images shot at a light level of one foot-candle (roughly equivalent to typical city street lighting at night) at a range of ISOs. NOTE that the camera warned me that the prototype firmware wasn't qualified for exposures this long, so the noise levels are doubtless higher than production models will achieve.

ISO 100, one foot-candle (1.5 seconds, f/2.8)
ISO 200, one foot-candle (1.0 seconds, f/2.8)
ISO 400, one foot-candle (0.5 seconds, f/2.8)

Incandescent performance
Since this wasn't a production camera yet (and I had *precious* little time with the camera), I didn't take time to shoot all my standardized test images, and one set that I dropped was the "indoor portrait" shot. This is often a very tough shot for digicams, due to the strong yellowish cast of the household incandescent lighting it's shot under. Hearing that people were complaining of the 14n's behavior under incandescent lighting, I did take time to grab one shot of the Davebox under incandescent lighting (a pair of 100 watt household bulbs in reflectors). The results are shown below.

Pro 14n under household incandescent lighting
Auto white balance The 14n's auto white balance had a pretty hard time with the strong color cast of the incandescent light source.
"Click" white balance adjustment BUT, a "click" white balance correction really cleaned things up, producing beautiful colors with not a hint of color cast! (Keep in mind that you can do this adjustment right in the camera, with unusual precision in selecting the area you want to use for your neutral reference.)

Conclusion:
A bit of a mixed bag. The 14n's color in this test was pretty good, but not as saturated as I'd expected, after seeing the results on my other "real world" shots. Noise was pretty low at ISO 100 in neutrals, but colors show higher levels, particularly highly saturated blue hues. Looking less at this test than at shots like the Silo or Studio test, I'd say that the 14n has very nice-looking color. Even in the harsher test of the Davebox, the 14n does very well, just not quite into the topmost tier.

<<"Studio" Analysis :(Previous) | (Next): "Wide Angle" Analysis>>
<< Return to Main Test Images Page for the Pro 14n