Nikon D2HNikon introduces an 8 frame/second speed demon, with WiFi connectivity and an amazing new flash system to boot!
<<Video, Power, Software :(Previous) | (Next): Test Results: Noise>>
Page 12:Test Results: OverviewReview First Posted: 12/18/2003
Test Results: Overview
This is going to be a somewhat more involved "test results" section. As always, check out the D2H's Sample Pictures page for a detailed look at the standardized test photos I shot with the D2H. - And also be sure to check out the Gallery page for a set of random photos (that is, less boring, non-standardized ones), including shots captured by pro photographers Gibbs Frazeur and Stanley Leary at local college sporting events.
Given the furor that was raised over early sample images from the D2H though, I'm going to look at issues of image noise, color rendition, and resolution in quite a bit more depth than usual. Regular readers will recall that I and a number of other journalists had an opportunity to shoot with what were supposed to be late-model preproduction samples of the D2H. I unfortunately was the first to post sample photos from that shoot. - I say "unfortunately," because Nikon subsequently determined that those cameras had image-quality problems, particularly in the area of image noise, and therefore asked me to take down the images I'd posted. (By that time though, something on the order of 100,000 people had read my preview and downloaded multiple image files, putting my bandwidth costs through the roof for the month.) Rob Galbraith later discovered that at least part of the problem appeared to be caused by the prototype cameras shooting at higher ISO levels than indicated: A setting of ISO 800 on the preproduction cameras apparently corresponded to an actual ISO level closer to 1250. (Read Rob's comments on his experience with the D2H to date here.) Rob and others attributed the noise levels seen in the preproduction samples of the D2H to largely be the result of the errant ISO settings, and I'm sure it was at least a contributing factor.
Upon close examination though, it's now clear that false ISO readings accounted for only part of the elevated image noise, and in fact, the photos Nikon had me pull down may actually have been more fairly representative of production-model noise levels than was supposed. (Rob Galbraith has also come to the conclusion that the D2H has a noise problem, calling noise levels in the midrange of ISOs from 400-640 "unacceptable" in an update to his article, dated 12/9/2003.) Given the level of foment that has come to surround the D2H's image noise levels, I wanted to take particular pains to be as carefully analytic in my approach to it as possible, so there'd be no question of the objective validity of my results. There also appear to be issues with the D2H's color, particularly when the ISO is cranked up. - So I've likewise devoted a fair bit of attention to that parameter as well. Finally, in their presentations to the journalists at the D2H rollout event, Nikon seemed to make a particular point about the D2H's resolution, suggesting that characteristics of the LBCAST sensor would translate into greater ability to resolve fine detail than conventional technologies. I therefore expended a bit more effort in the resolution/detail area as well. (Happily, there are no negative results to report in the resolution area, but there also don't seem to be any anomalously positive ones either: The D2H behaves very much as a 4.1 megapixel camera, delivering very equivalent resolution to competing models with similar pixel counts.)
So, on the next few pages, we'll take a closer look at the D2H's noise, color, and resolution characteristics, followed by a brief summary of other camera characteristics I commonly comment on.