Best of the best: The Phase One IQ3-100MP Trichromatic compared to the original IQ3-100MP
By Andrew Paquette – See his website HERE
I recently had an opportunity to test Phase One’s new IQ3-100 Trichromatic back, hereafter called “trichromatic”. It is a terrific digital back, but how different is it from the original IQ3-100? Both are excellent backs and have excellent color. A couple of recent reviews contain photos that have stunning color. One from Luminous Landscape is worth a look if you want to understand in what way the colors are different in the trichromatic. Based on these reviews, I looked into what it would cost to upgrade from the IQ3-100 to the trichromatic. The prices may vary depending on what region you are in, but suffice to say that I could have bought a small but new car for the upgrade price. With that in mind, I wanted to have a test to see if it would be worth it to me. Having said that, the trichromatic is substantially better in many ways that any other medium format digital back available. The question for me was, how much better than the next closest digital back, the IQ3-100?
I would have preferred to do the test outside with some models, but thanks to enforced convalescence after a recent operation, I had to do the test in my living room with some simple still lifes. I don’t do these very much, so I haven’t put together a proper product photography stage yet. All this means is that instead of a proper seamless background, I used colored construction paper for backgrounds. The main thing I wanted to test was color fidelity and smooth gradients, so that was enough for my purposes, even if it wasn’t optimal.
I used a Phase One XF camera body and a Schneider-Kreuznach LS 120 mm blue ring lens for all of the shots. For lighting, I had natural light coming in from a window on the left, and fill light bounced off the ceiling from a ProFoto B1 with an octobox. All of the shots used the same lighting, but the settings changed as the light shifted outside. This is most obvious in the shot with a green background. The shoot was done with a tether, so all images were available for review as soon as they were shot. Also, most are focus stacks. The reason is that when I shoot macro, I almost always use focus stacking. The settings varied within a narrow range. All images are shot at ISO 50, f-stop ranged from 4 to 4.5, and shutter speed varied between .8s and 1.3s, depending on available light.
All of the photos used the same elements, but the compositions are not identical. This is because I only had the trichromatic for a couple of hours, so all of the IQ3-100 shots were done before the trichromatic back arrived, meaning there were a number of changes made to the various elements to get the shots, and then due to limited time, similar but not identical compositions were made with the trichromatic.
There are two batches of images presented here. The first batch are color comparisons, the others are final edits. The color comparison images have had one post-processing edit applied to them: all have had highlights restored by adjusting the highlight recovery slider in Capture One. Apart from that, these are straight out of the camera. The second group of images have had too many edits applied to describe in a reasonable amount of space.
My impression during the shoot, as I saw the images appear in Capture One, was that the color from the trichromatic was more accurate than the IQ3-100. This was expected, primarily because of the article in Luminous Landscape. The areas of greatest difference were greens, which were noticeably different and more accurate, without much squinting or side-by-side comparison. Yellows were also quite different, though it was easier to see this if the images were close together. Blues were different, but unfortunately, I can’t show that here because I forgot to shoot the blue background with the IQ3-100. In all cases, the trichromatic colors were cleaner than the IQ3-100. With reds, color differentiation was improved to the point that image clarity was enhanced in addition to having more accurate color. These differences should not be taken to mean that the IQ3-100 produced weak color. To the contrary, it produces stunning color also, particularly in comparison to every other camera made, from Hasselblad, to Sony, Nikon, Leica, Canon, and others. However, even among the best of the best, there are differences.
Color comparisons (IQ3-100 on left, Trichromatic on right)
Figure 1Yellow and green comparison, trichromatic on right
Figure 2 yellow paper, trichromatic on right
Figure 3 green needles, trichromatic on right
Figure 4 cardboard, trichromatic on right
In all of the comparisons, yellows tended to be more red from the IQ3-100 than the trichromatic, as can be seen here. Of perhaps greater importance, a visual inspection of the subject with the photo beside it on a monitor, shows that the trichromatic is more accurate. That is, the extra red seen in the IQ3-100 did not belong. This conforms with the color model described in the Luminous Landscape article, that shows how certain colors have a tendency to bleed together on non trichromatic sensors.
Figure 5 Red green comparison, trichromatic on right
Figure 6 red paper, trichromatic on right
Figure 7 green gourd, trichromatic on right
In this example, greens are noticeably bluer from the IQ3-100 than the trichromatic. One might argue that the trichromatic has red reflected into the green from the backdrop, but if that was the reason, it would happen with the IQ3-100 also, which uses the same backdrop.
Figure 8 Pink and yellow comparison, trichromatic on right
Figure 9 pink paper, trichromatic on right
Figure 10 yellow green squash, trichromatic on right
This shot highlights one of the reasons a photographer might want to upgrade to the trichromatic. In the background, the pink is noticeably bluer than on the right, but the yellow of the squash is redder on the left. A simple color temperature adjustment won’t fix this. This shows up in other photos as well, where certain colors have highly specific differences that are not carried through the rest of the image, because of the way specific wavelengths are treated in the CFA profile. The trichromatic has less overlap of wavelengths than in a standard CFA, which results in cleaner colors. Also, from looking at the subject, I can verify that the trichromatic was very accurate in comparison to the IQ3-100. Another point is that the additional clarity also increased the apparent sharpness of the image.
Both backs have the same resolution and bit depth, so the major factors where a difference is visible have already been covered. There are other differences though, but they aren’t relevant to the kind of shooting I do. For instance, the ISO range starts at 35 instead of 50, and at high ISO, color noise is diminished. I did do a few shots at ISO 35, but these are very difficult to distinguish from ISO 50, so they aren’t presented. I don’t like to shoot above ISO 800 with the IQ3-100, but have gone as high as ISO 1600 with acceptable results. It can go much higher than that (ISO 12,800), but whenever I have shot at 3,200 and above, I’ve tossed the photos. There have been times when I needed to get more light in the camera but I didn’t have the option of bringing lights to the shoot. When that happens, I prefer to pull out my Nikon with an Otus or some other lens that has a max aperture of 1.4 or so.
For fun, here are a few of the final shots. These have all been edited, so they aren’t very good for making comparisons.
Figure 11 green gourd having a drink, IQ3-100
Figure 12 Desert mushroom, IQ3-100
Figure 13 Poltergeist chairs and pumpkin, IQ3-100
Figure 14 Grandpa squash, IQ3-100
Figure 15 Squash at the beach, trichromatic
Figure 16 Forest gourd, trichromatic