On a Color Mission – Comparing two RA4 color papers

With FUJIFILM being the only remaining manufacturer of RA4 color printing paper in the West, one of the questions remains how their products compare. Let’s have a look at an entry-level product in their portfolio pitched against a one of their top-level products and see what real-world differences there are.

When I started printing color with the RA4 process, we still had Kodak papers as well as FUJIFILM. There was a real difference between printing the same negative on Kodak Endura or Fuji Crystal Archive – usually to the benefit of the former. However, part of my preference for Endura ended up being due to an unfortunate comparison with Fuji’s entry-level papers.

Later on, I learned about the other papers they have on offer, and I found that the higher-end papers really were nicer (simply and subjectively put) than the entry-level papers. You get what you pay for – or, perhaps, you get a lot more than what you pay for, because RA4 paper is still relatively (very) cheap.

I’ve written before about technical differences and similarities between the RA4 papers in the FUJIFILM lineup. Notably, I highlighted the fact that they share the same pedigree, and indeed, they really are variants of exactly the same product. But the subtle technical differences to translate into real-world performance differences that are meaningful to some printers – and certainly to me.

I learned about these differences mostly by buying increasingly ‘fancy’ papers and then going through an experience of “oh wow, this really does look better” every time. What I never did, until now, was systematically compare multiple papers. Well, let me tell you about one late night in the darkroom with a humming Durst RCP20, some negatives and two flavors of FUJIFILM RA4 paper.

This is going to be a long story, so I’ve compiled a table of contents for easier navigation:

• Materials and setup
• First glance and methods
• HD curves
• Black
• Mottling
• Crossover
• Dye density
• Conclusion
• Example photos
• Disclaimer

Materials and setup

To set the record straight right away – one of these papers is technically not really a FUJIFILM paper. Or rather, it really is – it’s just not being marketed under that brand. It’s actually ADOX Color Mission RA4 paper. But the public secret on this is that it is FUJIFILM Crystal Archive (I’ll call it ‘CA’) paper – with the notable characteristics that it has no backing print (which FUJIILM-brand CA does have), and that ADOX cuts this into convenient sheet sizes and puts it into (quite attractive) boxes.

I’ll be comparing this paper with my go-to paper: FUJIFILM DPII, which I buy in rolls and cut into sheets myself. DPII is Fuji’s standard ‘pro’ paper and one notch down from their super-duper-premium Maxima paper. But Maxima only comes in roll sizes that I find unwieldy, so I’m sticking to DPII instead.

The first and immediately apparent difference is that DPII is a thicker, heavier paper than CA. FUJIFILM lists the thickness of CA as 205um and DPII as 240um. This 20% difference may sound small, but it makes a real difference in how flexible/sturdy the paper feels.

First glance and methods

My comparison uses the glossy variant of both papers, and it also turns out that the image surface of the DPII paper has a more rubbery, resistant feel to it; CA feels slicker and smoother. The nature of the ‘glossiness’ is also subtly different; they’re unmistakably both a high-gloss surface, but the surface qualities differ slightly in a way that’s difficult to put in words. Specular reflections on DPII seem a little smaller and more concentrated than they do on CA.

As hinted in the example photo above, I printed a series of density steps on both papers to take some measurements on. My previous experience was all subjective, and I wanted to get a more solid feeling for the real (more objective) differences as well. So I whipped out the Stouffer T3110 and i1Pro photospectrometer.

Here’s the starting point for the density analysis that follows:

The methodology is simple: I used my LED enlarger to expose some contact-printed T3110 steps (0.10logD steps) with pure red (660nm), green (550nm) and blue (450nm) light. This yields pure cyan, magenta and yellow images with no crossover, except for the overexposed parts of the yellow step, which gets contaminated with magenta (this is by design).

For the greyscale, I determined a combination of red, green and blue that would yield a (sort of) neutral greyscale, which I did through simple visual iteration (“does this look grey? nahhh….looks a bit green; let’s add a little green…” etc.) The greyscale wedge was not so much intended to be a dead-neutral grey and it’s OK that it has a bit of a color cast to it; we’ll come back to this later.

The exposure time and lens aperture for the cyan, magenta and black/grey steps was identical, but the yellow step wedge was exposed for three stops less (0.5 sec instead of 4.0 sec). This was to cut back excessive fogging; the yellow layer is also (by far) the fastest as well as the most prone to halation issues.

Exposure for the DPII and CA papers was identical, as was processing. I used my trusty/rusty old RCP20 for this, loaded with FUJIFILM MP90 replenished developer and FUJIFILM CP-RA replenished blix. Development time was around 45 seconds at 35C. The strips were then measured with an i1Pro and data were processed in Excel. The scans of the strips shown above were made with my Epson 4990, which I used in ‘raw’ mode (no color corrections) and then adjusted the color balance using the ICC profile I made using an IT8 target obtained from Wolfgang Faust. The resulting scan is quite close to the real-world appearance of the strips.

HD curves

Looking at the test strips, differences are already visible. The DPII strips just look more intense, so to speak. This difference becomes more apparent if we look at the plotted data.

To start off: one thing that’s virtually the same on these papers is the amount of exposure they need to start building density. There’s a very small difference on the yellow layer, which needs a little more exposure on CA than it does on DPII to get going. Magenta and cyan seem to take off at roughly the same point. Note that both papers also require roughly the same amount of exposure to hit their respective maximum densities (‘dmax’). If you want to draw a parallel to B&W variable contrast paper grades: they are about the same ‘contrast grade’.

The key difference is quite clear; DPII generates significantly more density in all three color layers than CA. One implication of this is that to get the same medium-density value in an absolute sense, the CA paper needs a little more exposure. Because the dmax of DPII is higher, the contrast appears to be greater, and in an absolute sense, it really is – as expressed as the difference between paper white and dmax. But within their own respective density spaces, both papers are of the same contrast in a ‘gradation’ sense.

Black

This difference in dmax for each of the color layers also translates to a deeper black, which is limited to around 2.0logD for CA and hits 2.5logD for DPII. Now, greyscale density differences above 1.8logD or so tend to become a little subtle to see, but the difference both quantitatively as well as qualitatively speaking is very real. Here are the black/grey strips again, photographed in (admittedly rather harsh) daylight:

What’s also apparent in the greyscale strip, is that the color balance is subtly different. I balanced the colors on the DPII test strip, but it came out a little on the blue side; it’s less pronounced in the real-life strip than it looks on the photo above. The scans shown earlier are more representative.

Still, it’s clear that the greyscale looks relatively warmer (less blue/more yellow) on the CA strip than on DPII. This is in fact consistent with the density plots shown earlier: these show a larger difference in dmax for the magenta and cyan channels (which combine into blue/violet), while the density difference in yellow is a little less.

There’s something more going on, too. If we boost the contrast on the black/grey set of strips, we get this:

Note that the top strips (1) shows a warmer shade of black that leans towards red/brown, and (2) exhibits a clear mottling pattern. Both of these aspects you will find in any print featuring large black areas (e.g. a low-key portrait against a black background) on CA paper.

Mottling

Coincidentally, this is one of the reasons why I started my ‘upgrade path’ in RA4 papers in the first place. Initially, I used a mix of papers, including Kodak Endura, some rather old but still printable Fuji Crystal Archive from around (I suspect) 2005 or 2010, and some (then) new Crystal Archive. Only the latter paper exhibited this mottling paper; the old Crystal Archive from 10+ years prior printed very satisfying, deep blacks, and so did the Endura. I found that as I upgraded papers, going from CA to Crystal Archive Supreme and then to DPII, the problem became less (Supreme) to the point of invisible (DPII).

It took me a heck of a long time to figure out what was going wrong, as I initially suspected a problem on my end. This turned out (for once) not to be the case. Here’s a hint – if we boost contrast on each of the strips, we get this:

You’ll notice if you look closely that the mottling problem only appears in one specific strip: the cyan CA strip. So apparently, the problem is related to that very specific layer.

And the plot thickens if I tell you that several years before I even started color printing at all, I was printing on an old version of Adox EasyPrint RC B&W paper, and I ran into…the exact same kind of mottling problem! What the h….?

As it turns out, the problem is related to the paper base. The present paper base used for these FUJIFILM papers is not perfectly even – it’s very slightly wavy, like a gently rolling landscape, with dips and valleys on a micrometer scale. Since the cyan/red layer is the bottom emulsion layer, it’s this layer that conforms to the surface of the actual paper base. If that surface happens to be slightly wavy, this translates into slight thickness variations of that emulsion layer. Since emulsion layer thickness relates directly to dye density, this explains why we’re seeing cyan density fluctuations – a.k.a. mottling.

Wait, hold on – so why doesn’t the DPII paper suffer from this issue, and a mid-range paper like CA Supreme less so than plain CA? This must have something (or a lot) to do with the thickness of the layers – I suspect that the actual emulsion layers on CA are on the thin side, and so marginally thin that the mottling just starts to become a problem. The emulsion layers on the premium/professional papers like DPII and Maxima are significantly thicker, and thus the surface unevenness of the paper is less of a factor. It’s also conceivable that the paper base of the more premium papers is actually smoother – but I don’t know for sure. The layer thickness differences I do know for a fact are real.

OK, and what about the Adox EasyPrint B&W paper story? And the old Fuji CA paper that gave such nice blacks? The latter is simple – FUJIFILM changes their papers from time to time. The old CA that I got those nice blacks from was in many was more comparable to modern DPII than to present-day CA. Put differently: present-day CA is a slimmed-down version (literally!) of its old namesake. As to the EasyPrint parallel…well, I’m speculating a bit here, but I have reasons to suspect that the paper base of that old Adox EasyPrint was the same or very similar to the present-day paper base of FUJIFILM CA. If you look into who presently makes photographic paper bases in Europe, you’ll realize why I may think this is the case. Again, this is speculation on my behalf.

Crossover

Earlier, I’ve created a bit of a fuss with stories about crossover in today’s FUJIFILM RA4 paper. To touch/expand upon that topic a little, let’s have a look at two sets of chart. The first is a set of curves for each paper normalized within their full scale (as opposed to showing absolute density) and superimposed so that the toe region overlaps:

Note that both sets look pretty much identical and show no severe crossover problems. Note also that these plots are made on the basis of pure cyan, magenta and yellow test strips – so I’ve effectively isolated the colors, so to speak.

Things look a little differently if I use the makeshift greyscale test strips and extract cyan, magenta and yellow data from that to make the same kind of crossover plot:

In this case, it becomes apparent that the CA paper has a significant cyan crossover, resulting in relatively speaking red highlights and cyan shadows. How come this effect is present on the CA paper, but not (or barely) on the DPII paper? I think what we’re looking at here is the effect of the interlayers, one of whose functions is to prevent interactions between the image layers during development.

The dyes after all are formed by oxidized developer molecules linking up with dye coupler molecules embedded in the image layers. If these oxidized developer molecules are allowed to wander from one image layer (where the development took place) to an adjacent layer, they will create dyes that do not correspond to the actual exposure. Apparently, this result mostly manifests itself on the cyan layer, which may have to something to do with the fact that this is the bottom layer of the paper. Oxidized developer that starts wandering about ends up in part at the bottom layer, from where it has nowhere to go (except back up through the whole emulsion stack). Think of it as a dead-end street with cars driving in and piling up in a traffic jam at the dead end.

Saturation and gamut

Anyway, let’s go back to the colors proper for a bit. The curves and visual inspection shows a difference between CA and DPII, and that difference is also visible if we look at the color space both papers can cough up:

What becomes apparent here is that DPII is capable of higher saturation on all channels, with the difference being biggest on yellow, followed by magenta and then cyan. The implication of this is that DPII can print more saturated colors for not just the CMY primaries, but also all mixed hues in-between. In short: DPII can produce more vivid prints.

But why is the yellow saturation higher, while when we were looking at the HD curves earlier on, the relative maximum density difference was biggest on the cyan and magenta colors? This has everything to do with hue purity. Both papers suffer to an extent from crosstalk between the color layers. Put differently: if you print pure yellow, you always get a little magenta (and ultimately cyan) with it as well. The degree to which this happens differs between these papers, and apparently FUJIFILM manages to control it fairly well in CA on the cyan and magenta layers, with the yellow layer being more challenging. I think much of this, again, has to do with the interlayers between the actual image-forming layers.

Dye density

OK, one more chart and then we’ll look at a few photos, promise.

Shown above is the spectral reflectivity of the dye images, which correlates to this plot from the datasheet (taken from that of DPII):

Note that in the datasheet, dye density is plotted while I show the inverse (reflectance). So the valleys in my plot should be compared to the peaks in the datasheet plot. There’s another key difference, which is that the datasheet plot shows the density of the actual dyes, while I’ve measured the reflectivity of the colors as they are on the paper, embedded in the emulsion. This means that in my measurements, the paper base, gelatin emulsion and adjuvants like optical brighteners also affect color response.

For my plot, I took the dmax for cyan and magenta, but I took the plot from step 11 in the step tablet for yellow. This is because with overexposure, yellow gets contaminated with magenta; an inherent design feature of color RA4 paper.

The main reason I show the dye reflectance plot is to illustrate that yes, it’s really the same dyes in both paper. The physical coloring agents are the same: the spectral plots are effectively the same, save for a difference in intensity. The way the colors are implemented in the paper in terms of layer thicknesses etc. is what accounts for the differences that we’ve seen.

Conclusion

What to make of all this? First and foremost: there are differences between these papers, but in the grand scheme of things, they’re subtle. From an objective quality viewpoint, DPII is the better paper: it offers higher saturation, a larger gamut, deeper blacks and has no mottling issues. It also is coated on a thicker base, but whether that’s better, of course depends on what you’re after.

This is also true in general: while the objective advantage is to DPII, subjectively, you may prefer CA over DPII for certain purposes. The higher saturation of DPII also means that it’ll more readily emphasize color casts, and it’s overall a more ‘punchy’ look. So if you’re looking to print something in very neutral or more subdued hues, CA is an easier paper to work with. On the other hand, if you’re printing something that relies on areas of solid black or otherwise very deep or saturated tones, DPII is the better choice.

From a practical viewpoint, there’s the fact that ADOX Color Mission paper comes neatly cut to standard sizes in attractive boxes, and out of the box, it lays perfectly flat. DPII is available in cut-sheet from some sources as well, but I personally cut it from rolls. In that case, it generally tends to preserve its curl unless it’s stored weighted down for an extended period of time. I have the means to conveniently cut sheets from rolls in my darkroom, but if you don’t, having access to papers from a reputable manufacturer (and in this case, confectioner) such as ADOX is a benefit.

Last but not least: the printed examples below demonstrate in my view very clearly that both of these papers (and the in-between variants like CA Supreme) are by all means suitable for optical/analog enlargement of C41 color negatives as well as ECN2 negatives.

Example photos

OK, gimme the darn pictures already. The proof is in the printing, after all, isn’t it? All prints below were made on a fairly small size (13x18cm to 15x20cm image area or so) for easy scanning and processing. Color filtration for the DPII vs CA prints was identical; please bear this in mind! Exposure of the CA prints was 15-20% longer to account for the on average lower density of the CA paper. This means that the highlights on the CA prints tend to be slightly darker, while the shadows end up slightly lighter. This is a compromise, but the papers just are different, so some kind of middle ground needs to be found – I opted to keep the midtones approximately the same in terms of lightness.

CA prints are top or left, DPII prints are right or bottom, always. Some of the prints show patches of severe mottling in light areas; ignore this as it’s a scanning artefact (dirty platen).

First up some shots on Kodak Ektar 100 negative film, exposed in Italy in September 2024. The film was home-processed in FUJIFILM C41 chemistry.

Now for some inherently much more muted tones. The following shots are on Kodak Vision3 250D film, developed in ECN2 developer with a slight push (3m30 development time) to compensate for the inherently lower contrast of ECN2 film. The light and subject matter are also fundamentally different from the Italian job above.

Disclaimer

This test was performed independently. None of the companies whose products were used have had control over the methods or outcomes of this test, nor have they been presented the results for review prior to publication. ADOX FotoImpex kindly gifted me several boxes of their Color Mission RA4 paper, but without any specific instruction, expectation or suggestion for use. FUJIFILM provided me at an earlier stage, far in advance of these tests, with some technical information and answers to some of my technical questions. I received no other financial or material support beyond what’s mentioned above.