I am writing this review in the form of a field report. I have used the D1 for over a year and now have used the D1x for about 3 months. In my photography, I mainly focus on scenic shots involving rivers and creeks in my native state of Arkansas. I also do some event work on the side, but this review is only being written from the point of view of the outdoor work that I do. In my work I only shoot raw. I use all of the three main software tools; Qimage, Bibble and Capture 2. With most of my shots, I work with water as my main subject matter preferring to slow the motion of the water down with time exposures.

With the former D1 and its minimum ISO of 200, I often had trouble with "blown out" highlights on my water. In my work, I use a circular polarizer (CL-PL) combined with at least one Neutral Density (ND) filter, sometimes two. I also tend to use the higher apertures starting around F11 all the way to F22 as I am constantly working to cut down the amount of light hitting the focal plane so that I won't "blow out" my highlights. I have come to realize that many times I can pull out details in the shadows but a blown out highlight is pure white and there is no recovery. I prefer to shoot in sunlight or partial sunlight, as the colors and overall contrast are much better. When shooting in cloudy conditions, I feel that your image will suffer considerably in the areas of saturation and color depth as the colors take on a flat look.

1. The ISO of 200 caused me to use too many filters i.e. CL-PL and ND even at the F16 or F22 stops.

2. The write to storage speed of the D1 was slow taking about 15 seconds to write one raw file to a 340MB microdrive.

3. You couldn't take advantage of the memory buffer on the D1 while in single mode, thus you had to wait till the image had written to the card before you could take the next image. Not a big issue but it did get rather tedious at times.

4. The max resolution of the D1 was an issue in regards to printing anything larger than 8x10. If I was shooting a close up or macro I had no problems taking the image up to 11 x 16, but any shot that included a lot of smaller details such as leaves, branches or small details on rock, would suffer at the 2000 x 1300 resolution of the D1

5. Viewing the LCD in any condition other than total darkness was almost impossible and I had made a separate hood to allow me to see the image.

6. The battery life on my D1 was very limited. I tend to preview many of my shots as I am constantly checking to see if the highlights were "blown out". This would really task my internal batteries and early on I modified an EN-4 so I could use external sources of power.

7. As most shooters of the D1 have noted, the issue of dust on the filter of the camera is a never-ending problem. I was hoping to see Nikon change the design on the D1x so that you could clean the filter with more ease.

8. Support for the IBM microdrive as Nikon had never offered official support.

There are some other issues but this really takes the majority of them into account.

Its minimum ISO is 125, which I was hoping would allow for me to shoot water without blowing out my highlights. The D1x supported a much higher final resolution of 5.74 mega pixels that allows a resolution of 3008 x 1960. Actually the math comes out to 5.89 but apparently you lose a few thousand pixels in the conversion process. However the camera doesn't do this straight off the chip, as the layout of the pixels is unique. You had a 4,024 x 1,324 basic CCD and Nikon has written a special interpolation to move approximately 1 million pixels from the horizontal, back to the vertical. This concerned me, but the only other camera out there that offered 6mp (3032 x 2008) was the Kodak 760. The price point of the 760 took it out of consideration.

Another issue that Nikon had addressed was the support for the 1Gb microdrive, which was a major issue for me. The price point of solid-state memory in the 512MB size is currently priced at a premium. I had used (2) 340MB microdrives on the D1 and never had a problem so the thought of using the 1Gb on the D1x was not an issue. Nikon was claiming that they were using a new LCD on the D1x that used white backlighting so I hoped that I would be able to view the images better in daylight. Hopefully the monitor would be similar to the LCD on the 990 Coolpix, which I have always felt was excellent.

Unfortunately Nikon made no change in design of the placement of the CCD/filter, so cleaning the filter is still a major hassle. Also since the CCD of the D1x is the same physical dimensions as the older D1, dust actually is twice as bad in that you now have a CCD with pixels approximately half the size of the D1. One piece of dust will cover more area of the image. They also had still not authorized the use of the sensor swab (From Photographic Solutions) as an officially approved method of cleaning. This disappointed me as I had used the swabs on my D1 with no real problems. In the field; dust is the true bane of a digital SLR.

I found the following areas as a positive improvement overall:

• Write speed to the microdrive was much faster 2.5x
• Option to use the LCD for setup and custom functions
• Much improved battery life of the EN-4
• Improved noise with the lower ISO of 125
• Better selection with the color modes, either sRGB or Adobe RGB (1998)

The most immediate impression was the overall speed with which the camera would write a full resolution raw image to the microdrive. I was using both the 340MB and 1GB sized drives and there didn't seem to be any differences in write times. I was able to write a file from the buffer to the card in around 4.5 seconds and I was now able to take full advantage of the buffer and shoot as many times as I wanted, not having to wait for the first image to complete the write. I realize that the buffer on the D1x is small and many people have complained about it, however I was never hindered by it at all. In the past on the D1, I had reverted to changing to continuous mode and setting the counter to 1. This allowed taking as many images as I wanted, but I could no longer view the images, I had to move to play mode. If you plan to shoot raw on the D1x, a 1GB microdrive would be a good investment. With a raw file taking 7.9MB, you can get about 128 images to a card.

Another positive impression was that even while the camera is writing the file, you can opt to view it and you see the complete image. On the D1, this wasn't true as you had to view an incomplete version of the image until it had fully written to the card. Of course this meant that your LCD stayed on much longer and the battery went dead fairly quickly. Also the image play back on the D1x is very fast. You can now zip through the captured images on the card as quickly as you want.

I found that I had to still use the CL-PL and most often a 4x ND filter for most of the shots. This kept the highlights in line for most of the shooting but I still had problems and really had to watch my metering. In general, I used the center weighted metering with the smallest circle for the metering sensitivity. I shot most of the images in the full resolution mode and then the medium jpg mode to see if there is any difference in the results with highlights.

The camera did allow me to work effectively as I moved around my subjects. I used the LCD to step through the menu screens as I changed the various settings I needed. As with the D1 the tripod mount location is excellent and very stable with the rubber base. I made the decision to set the in camera sharpening to "none". I found that using the LCD was easier than having to remember all the custom settings of the older D1. While navigating the menus the screen seemed to have plenty of brightness to read and make your choice. However, when reviewing a captured image the same is not true.

I found that the LCD displayed a colder image than that which had been taken. Many of the images seemed to have a blue tint so that the effect was a colder looking image. Also the screen seems to display the image darker than the captured image making you feel that you have underexposed the image. However as you look over the image you can check the basics such as focus, correct framing, highlights etc. I did not use the zoom feature, as it cuts down on the total buffer memory available by 1/3. With any light at all on the LCD then you will find it hard to view your image, as it will be washed out. I used the both a Hoodman LCD cover and the Hoodman rubber hood and still had trouble viewing the images.

The D1x's viewfinder is the same as the D1 so you get the same amount of excellent information in regards to shooting. These include the exposure compensation setting, the shooting mode selected, the number of your shot, and the like. I shoot much of my work in M mode (manual mode) so having the command dials available to adjust the shutter speed and aperture are a great help. The real major change in information display is the use of the camera's LCD for changing your shooting, playback and custom function setting. This greatly expedites your changes, as you don't have to remember all of the various numbers and settings for the custom functions. I made many changes throughout the day to my shooting settings all of them while using the LCD and still had great battery life.

The battery life of the EN-4 with D1x over the D1 is one of the most important improvements in my opinion. On this trip I was able to shoot the entire day on one battery and I ended up taking over 250 images. I previewed over half of them and since I was testing the medium jpg mode my use of the LCD was very heavy. I was able to shoot the whole afternoon on one EN-4 battery. With my older D1, I could have never done this; best-case maybe 70 images and these images were half the size of the files I was writing now. I used a new EN-4 that had been conditioned per the instructions by Rob Galbraith and these seem to really make a difference.

The noise at ISO 125 was very low. Shadows also held good detail also. This was a pleasant surprise for me. Also what noise that does show is not the pattern noise that was so common with the D1. During the shoot I increased my ISO to both 160 and 200. I did notice some increase in overall noise in my water at the 160 and 200 setting, but everything else was very clean.

I found the following areas to be of concern to me with the D1x

• A lack of the medium resolution setting for raw images
• No white balance support for jpgs with Capture 2
• Interpolation Aliasing (the jaggies)
• Still very easy to blow out highlights
• Overall image quality improvements of about 40-45% over the D1

On the day I had chosen I had mixed sunlight. I arrived on the site right at 12:00 p.m. and the storm clouds from the night before were still moving out. I set the camera to ISO 125 and put my 17-35 AF-S lens on. I started out with only the CL-PL hoping that I would have better control on the highlights. My average exposure would be a 1/3 sec to 1 second and right away I noticed that the camera was still going to be plagued with blown highlights. Spots where the water was concentrated like the bottom of waterfalls or cascades were blowing out and I didn't have total sunlight.

Without a doubt to me the best file format to use is raw. This is mainly in part for the white balance offerings that you have in Capture 2. Capture 2 allows you to totally change the white balance that was selected in camera. You can also fine-tune this setting in degrees. This only applies however to raw files and not to jpg. After comparing many of the jpgs I shot to the raw files, I realized that the color retained by my jpgs was not as rich as the raw images. On both I used the Adobe RGB (1998) for camera color mode. The difference between raw and jpg in color was really amazing to me. The jpg images just seemed flat when compared to the raw files. This was true no matter what software I used for image conversion, Bibble, Capture 2 or Qimage. Bibble will allow for a white balance correction on jpgs. I left the in camera white balance set to A for auto. The loss of the white balance correction with the medium jpgs in Capture 2 is huge hit for me. I am finding that most of the images I take with the D1x need some form of white balance correction and this is very hard to do in Photoshop.

After looking at some of my images, I feel that the AA filter on the D1x is considerably stronger than on the D1 making the images overall softer. This means that post conversion sharpening is even more a critical factor. I also feel that the improvement on image quality is about 40 to 45% over the D1. I had hoped to see a larger percentage, but after review of many images taken on this day I feel that's the best. The vast majority of my shots were taken on a tripod with weight added to allow for no movement. With any macro shot, you can easily crop your original image by 1/10 and still pull the crop back to around a 6.5 x 10 and have great image detail. This doesn't hold true with a big landscape or one of my typical outdoor scenic shots. The smaller leaves and such are still hurting. This means that after your conversion from raw to tiff your sharpening algorithm is very important.

I think it is important to not consider the D1x as a 6mp camera but instead more in the range of a 4mp camera. The D1x is interpolating within the camera to get to the final 5.74mp image. This is not to be confused with the normal Bayer pattern color interpolation, which any Bayer CCD will do. Anytime you interpolate for size you are going to create a softer final output. I don't think that the D1x is any different in this respect. The D1x is a nice resolution upgrade over the D1 but realistically it's not up to a true 6mp imager as on the Kodak 760. However by using it in the D1 mode, or 2000 x 1300 you might be able to get really decent color gradations out of it since you are getting one 1 green with every blue and red pixel. The larger imager has twice as many green pixels and using the older resolution setting of 2000 x 1300 means you can now have a 1:1 relationship. This leaves only one color to be interpolated instead of two per pixel. This is one reason I am still hoping to see Nikon address a medium raw format.

I was shooting both full resolution raw and medium jpg at the fine compression setting. The reason for the medium jpg was that I had discovered from my early testing that the D1x is introducing a new strange problem, interpolation aliasing or the "jaggies". You can see this effect most clearly when you shoot a red subject against a blue background. The red subject's lines will take on a stair step or jagged look. From talking to other photographers using the D1x, the medium jpg format doesn't seem to create this problem, since you get twice as many green pixels. However the jpg setting just doesn't have the rich color of the raw file. The jaggies are not to be confused with the normal color interpolation that has to occur within every Bayer pattern CCD. I have attached some crops of images that will show this effect. Currently I have not read much in the way of complaints on this issue however you can get the effect very easily. I also noticed it when I shot a dark subject like a tree limb against a light background. The problem is not a showstopper but it does create a problem with certain images. I have yet to find a way to remove it.

On the creek shot, I used the Nikkor AFS 17-35 and on the macro the Nikkor 60mm macro. On the day I shot the creek there was a slight wind blowing in some areas of the image but these show up as motion blur. However when you begin resample the image to 11 x 16.5, you will see that the details of the trees in the background don't hold up. They are better than the D1 could do as you would expect, but I was hoping for more like a 60% improvement on the D1. The D1x has problems with the smaller details. On the shot of the maple leaf, not the same issue. Here I was able to take an approximate 1/10 of the original image and then resample that to 8x10 with very little digitized effect. This is an example of the situation where all of your pixels were able to concentrate on one small area thus maximizing image quality. Also macro or close up work images don't seem to need as much sharpening after the conversion as images of the scenic type do.

My method for testing an image is to take the raw file and convert it to either a 3008 x 1960 sized file in Capture 2 or a 4018 x 2624 sized file with either Bibble or Qimage. The larger file sizes from the later two programs are nice to have but they again require additional sharpening. I have found that the ideal size from the D1x's sensor is the 3008 x 1960 or about a 6.5 x 10 sized print at 300 dpi. Here you can tell exactly just what you have with your image also you have a large enough print to read out your smaller details. In regards to the interpolation aliasing, I am finding that Capture 2 seems to bring out the effect worse than Qimage or Bibble. As far as resampling an image up past these sizes, the final results are very dependent on the engine being used for the resampling. I prefer the Lanzcos filter within Qimage for my resampling.

In conclusion, I am not writing with any intents or purposes other than to pass on my thoughts with a hope to increase discussion on the field use of the D1x. The camera is positive step in the right direction for Nikon and hopefully Nikon will step up to the interpolation aliasing issues in future firmware releases or a modification to the Capture 2 software. Right now I am still glad that I sold the D1 to purchase the D1x, however I am disappointed with my final image work in the larger print sizes unless I am shooting a macro or close up work.

I would also like to thank Ron Reznick and Stan Disbrow for their help.
Rob Galbraiths article on EN-4 conditioning can be found on his website, robgalbraith.com
Last but not least Uwe, for having this great site.