Red River Papers
Publishers Note: Mark has done an extensive review of both the Moab and Red River papers. In his previous review he looked at the Moab papers and in this review he covers the Red River line of papers
Red River Palo Duro Softgloss
This is a very lively, moderately textured luster-like paper with cotton rag substrate and no OBAs. It’s paper-white looks neutral-warmish, measured paper white being a bright L*98.2, b*0.2 (ever so slightly warm), whereas a* is -0.7, a little bit cool). It’s a quite hefty paper at 310gsm and 16.5 mil thickness, so you would feed this through the Canon Pro-1000 Manual feed or the Epson P800 Front Fine Art feed.
Gamut volume is remarkably large for these printers, the largest being about 867K for the OEM profile in the Canon Pro-1000 and the narrowest being about 818K for the OEM profile in the Epson P800. The custom profiles fall in-between.
Printed Maximum Black is a stellar L*1.7 for the OEM profile in the Pro-1000, and not too far behind it, L*2.2 for the custom profile in the P800. Maximum printed white is around L*98 for the four profiles – just a tad brighter in the P800 than in the Pro-1000, but nothing that would be noticeable. The neutrality of the Black and White printed maximums in both printers is very good.
I would have liked slightly better accuracy from all four profiles, the range being from total Average dE 1.02 (Custom profile, P800 printer) to dE 2.69 (OEM profile, P800 printer), the Canon results falling in between. See column “I” above.
Once the average dE of the 24 colors exceeds 1.0 or so, it usually means individual colors can show considerably larger deviations from their reference file values. For example even for the most accurate profile of the four, Epson P800 custom profile, while the average dE is just a bit above dE 1, one color mildly exceeds dE 1.75 – still OK, because this result will most likely be visually acceptable in a print of a real photograph. Figures 64 and 65 show the detailed dE graphs per color in the evaluation target. Once the dE values exceed dE 2~3, I would begin to be concerned.
Comparing the OEM with the custom profile performance please be mindful of the differences in the left-side scales in each of the graphs. The heights of the bars relate to different numbers depending on the profile, as indicated from the left-side scales. For example, in Figure 65 right side, the height of the bar for the colour Blue (toward the middle of the graphs) looks the same as that for the colour Red in the left side graph, but this similar-looking height relates to dE 2.75 in the right side graph and dE 5.5 in the left side graph.
For the Epson P800 set, the OEM profile shows only 1 color (White) near dE 1.00, while 7 others are below 2 dE. The other colors are above 2. For the custom profile, 11 values are below dE 1 and only 1 marginally exceeds dE 1.75. The rest are in-between.
For the Canon Pro-1000 set, the OEM profile shows 3 values below dE1, the remainder ranging from 1.25 to 5.5, the latter being an outlier. None of the others exceed dE 4.1. The custom profile shows 10 values below or close to dE 1, while 4 exceed dE2, the maximum being Blue at dE 2.75.
Turning to the grayscale performance, Figures 66 and 67 show the results of the 21 Neutrals test for the Epson P800 and Canon Pro-1000 printers respectively. To remind, the relevant range for Luster paper given the steps on this scale would be from L*05~95 inclusive.
For the P800 printer (Figure 66), OEM profile, the Chroma neutrality is well-behaved till level L*70, where after the dE spikes to over 1.5, then comes down again and experiences a second smaller spike to dE 1.25 at L*90. Thus, one could wish for more chroma constancy at the brighter end of the Luminance scale, but the extent of deviation from neutral is perceptually very low.
Luminance accuracy is reasonably maintained within 1 dE up to L*45, but from 0.64 at L*45 the dE spikes to 1.8 by L*50 and returns to a pattern of variability contained within 1 dE thereafter. The error at L*50 is that the patch printed less bright than it should have, but this wouldn’t necessarily be obvious to a viewer because the brightness is advancing about 4 levels instead of the 5 that it should have. By the time the scale gets to L*55 file reference value, in print, it is about 1 level lower, again not something that would jar a viewer. So once again, I must caution that while we like flat, low-lying curves, the extent of waviness we see in graphs like this does not mean that the tones or hues of the gray scale will appear choppy or biased in actual photographs.
The custom profile worked out very well over the relevant range insofar as the chroma and L* dE values for the most part remained below 0.75. This is good.
For the Canon Pro-1000 (Figure 67), OEM profile, both the chroma and luminance curves are maintained near or below dE 1.0 over the relevant range, except at L*5 where the dE error indicates a slightly warmer black than pure neutral. However, at that level of Blackness in a real photo one wouldn’t notice a b* value of about 1.39 rather than 0. For the custom profile, the whole relevant range is below dE 1.0 with most of it below 0.75, which is very good.
Turning to the tonal separation of the lower quartertones, the results of the 30 Neutrals test appear in Figures 68 to 71 below.
All four profiles indicate good shadow detail separation for this paper, the OEM profile for the P800 (Figure 68) being somewhat less accurate than the other three. With this latter profile, there is a strange downturn to the level 6 measured outcome that is not an aberration – I re-read this patch numerous times in different places on the patch and could not achieve a better reading.
Summing up for this paper – its dynamic range and color gamut are very good. Its accuracy of color rendition is on the whole satisfactory. It’s grayscale neutrality and tonal gradations are very good. It’s a nice hefty paper with a pleasing cotton rag substrate and finely textured coating with no OBA content, yet very bright and vibrant.
Red River Palo Duro Etching and M3 Profiling
This is a new and MOST interesting paper – not only the paper itself but also the manner in which it’s been profiled. Several brands provide “Etching” paper. They are heavily textured cotton rag matte papers that share much in common with other matte papers (in terms of their gamut and tonal reproduction), but they have this distinctive feature of heavy texturing.
Chromix makes Red River’s profiles (what I call the “OEM” profiles), and for this paper, the two companies opted for what we’ll call here an “M3 profile”. “M3” is a seldom-used (I believe) data measurement condition, which unlike M0, M1 and M2, uses a polarizing filter on the spectrophotometer that reads the profiling targets. Let’s spend a moment on why.
A special issue that the etching type papers (and matte papers in general) pose for profiling challenged Chromix: the surface texture scatters the light coming from spectrophotometers. When reading profiling targets, the effect of this scattering would be to under-estimate especially the dark tones, causing the profile to over-correct them by laying down more ink than those tones of the photo would call for, creating a dull or deadening effect especially in the lower quarter-tones. Red River’s commercial experience indicated customers’ concern for Black and White tonal rendition, particularly of shadow detail with matte papers, so the company was looking for ways of improving upon it and asked Chromix for assistance.
Chromix determined that using a polarizing filter on their spectrophotometer [you need a Spectroscan or a Barbieri – the latter about USD 9000- to do this) may (i) mitigate the light scatter that occurs reading the targets because it collimates reflected light from the spectrophotometer and (ii) produce a profile that creates more tonal distinction especially in the quarter-tones than is possible with profiles from M0, M1 or M2 measurement conditions. We’ll see below how this plays out.
First looking at the profile itself (Figure 72) two things looked strange for a matte paper: the very large gamut volume (749,418) while matter papers are normally in the range of 475-550,000 and the very low L* value for Black L*2, while matte papers normally fall within a range of L*13 to L*18. So I thought perhaps the profile was misnamed and applied to a luster/gloss paper rather than this one.
It’s important to keep in mind that I’m using a conventional i1Pro2/i1Profiler approach, whereas Chromix was using a very different set of tools described above, hence it is very likely that much of the high dE outcome in the foregoing results could well relate to this basic difference of instrumentation between profiling and testing. As well, the last two columns of Figure 73 indicate a huge difference between the Maximum Black value in the profile (2.0) versus that measured from the evaluation print (16.2). The former is internal to the M3 profile, and the latter from an M2 reading of the printed result from using the M3 profile.
I then created a standard M2 profile for this paper in the usual manner and opened it in the ColorThink Profile Inspector (Figure 75).
These are pretty normal results for this kind of paper when using conventional measurement data parameters (M0/M1/M2). Maximum Black has a more conventional value for matte paper in the L*15~16 range as measured in both the profile and the print, and the accuracy is very good, with an average dE of only 0.81; strangely, however, Blue has a uniquely high dE (3.35) in this profile (Figure 77), so I re-measured and verified that it is what it shows. Large outliers in the context of an average dE as low as this one are unusual and I have no explanation for it.
On the face of it, I thought from the data that my custom profile is the correct one, and there must be some error with the Chromix profile. I provided all this information to Red River and Chromix, out of which we had an extensive discussion. Chromix explained that profiling with a polarizer could create large differences of patch values, comparative dE outcomes, and print appearance. From this, it emerged that the impacts deserve further observation.
My further work involved two things: (1) Black and White luminance testing for both my profile and theirs, and (2) making both color and B&W prints using both my profile and theirs, then comparing. Item (2) is the really critical one because this is what counts – what people see when they look at the photograph. Item (1) is a guide to comprehension. This is where the story gets really interesting – and no – there is no break for commercials – you get to read it just below.
Step (1) has two components: (a) the “21 Neutrals” test which measures the accuracy of the L*0~100 grayscale in increments of 5L* values, and (b) the “30 Neutrals” test which measures neutrals from L*1~30 in 1 step increments – valuable for evaluating the profile/paper/printer treatment of deep shadow detail. Both these tests are described in my “LuLa” article, “Expanded Neutrals Testing”.
Starting with “21 Neutrals”, the dE results look as portrayed in Figure 78 for the Epson P800 printer.
Recalling that the flatter the curves and the closer to zero the dE values the better, the OEM outcome looks rather unappealing over good parts of the curves. For the Custom M2 profile, the results are a whole lot more satisfying IN THE GRAPH (flatter curves beyond the customary matte non-printable range – i.e. much lower dE values with less variance); hence, on the face of it, based on indicated accuracy, once again one would have the impression that superior Black and White prints will be had from the custom M2 profile. But this isn’t the end of the story – lest we be too mesmerized by numbers and graphs, stay tuned; there’s more.
We still need to look at the question of deep shadow detail using the 30 Neutrals test. Recall, this test is looking for the extent of tonal separation in the deep quarter-tones from L*1 to L*30, which is graphically indicated by the closeness of fit between the red line for the measured values relative to the linear black line, which is a 45 degree trajectory with a slope of 1.0 meaning a 1L* rise of output [from the printer] relative to a 1L* rise of input [from the so-constructed test image file]).
As this is not a proofing exercise for measuring dE values, one may implement this test using different Rendering Intents (RI) to see what difference they make to predicted outcomes for real-world photographs. I have chosen to use two RI – Absolute Colorimetric (ABSCOL) to see how the printer reproduces unadjusted file values (as in proofing) and Relative Colorimetric (RELCOL) + Black Point Compensation (BPC) to see how the printer reproduces adjusted file values per the standard RELCOL +BPC rendition in Photoshop. Normally we print real photos with either Perceptual or RELCOL, not ABSCOL, so it’s important to see what RELCOL or Perceptual would do to the appearance of shadow detail in a printed photo. I chose to work with RELCOL + BPC, as it does less shifting of in-gamut values and many prefer this RI. I could rerun these tests with Perceptual RI, but in light of my findings below, I think perhaps not really necessary. The usual advice for printing is to try both.
Figures 79 and 80 show the graphs for ABSCOL and RELCOL respectively.
Starting with Figure 79, which shows how these profiles differ before introducing rendering intent adjustments, note the huge differences in the shape and slope of the red (measured values) curve in ABSCOL between the Custom M2 and the OEM M3 profiles, even though both start from about L*16 in the print to represent L*1 from the image file.
In the custom M2 profile (Figure 79, Left side), the curve is essentially flat [all printed values about the same at L*16 (vertical axis) up to about L* 14 file values (horizontal axis)]; therefore, if you printed a photo with ABSCOL having important file values in the range of L*1 to L*14, you would see no difference – they would all print as one undifferentiated deep gray blob as the first 14 levels in the photo are being squeezed into 1 level in the print. “Lift-off” (red curve upward to the right in the graphs) begins gingerly beyond L*14, such that by L*18 from the file, the printed value is only three levels above 16 in the print. So up to that point (L*18 from the file) the overall slope of this curve is about 0.17 (printed value over file value or 3/18 – see dotted blue lines), meaning that on average over the whole of the scale from 1 to 18, for every l level that the luminosity in the image file increases, the luminosity in the print would increase by only 0.17 levels – i.e. just about invisible difference, in this case also unevenly distributed over the curve. Once at Level 20 from the image file, the output (printed) tracks the input (file) in a very well behaved linear manner. But that’s 20 levels up the scale.
The curve slope in the OEM M3 profile is very different, (Figure 79, Right side). From the start point at about output L*16, there is a fairly linear slope to the curve, getting up to level 27.5 printed (vertical) at the input file value of L*18 horizontal). Hence the slope of this curve is (27-16)/18) or 0.64, meaning that on average for every 1 level increase of luminance from the image file, the print luminance will increase by 0.64 levels in a fairly linear manner – a far superior outcome that should differentiate shadow detail much better than the custom M2 profile – remember this prediction and stay tuned for the photographic evidence.
Turning to Figure 80, where I examine the same variables, but this time for RELCOL, the difference between M2 and M3 is not as dramatic because with RELCOL we can activate BPC and the RELCOL formula creates an upward shifting of printed L* values from the adjusted Black point onward, referencing the increasing brightness of the input values from the file; but still, the basic distinction between M2 and M3 remains. For the custom curve (M2) in RELCOL, the slope is 0.4, while for the OEM profile (M3) it is 0.65, again a large difference predicting superior deep shadow detail separation for the OEM (M3) profile using RELCOL Rendering Intent.
We’ve reached the point where the indications from the data need to be tested with the printing of real-world photos, so I did two, using the Epson SC-P800 printer: (i) the Atkinson printer test page which is so good at revealing many a sin, and (ii) the Romans 16 Low Key BW image which is so good at revealing issues with separation of shadow detail.
Examining the prints of the Atkinson Printer Test Image made with both profiles using Relative R.I. (Figure 81), while it may be hard to see in these reproductions, two observations are important:
(1) For the most part, the color images look the same, the exception being in the yellow-orange-red part of the spectrum, where the Custom profiled (M2) print shows a bit more color saturation than does the OEM (M3) print.
(2) Careful inspection of the B&W segments in the Atkinson test image indicates that there is less deep quartertone separation in the M2 profile compared with that of the M3 profile. So this called for a bespoke deep quartertone test print using an appropriate test image, in this case, the Romans 16 Low-Key BW image (Figure 82).
Figure 82 shows the important segment of this test photo (custom M2 profile left, OEM M3 profile right). The key observations here are: (i) OEM better separates the man’s black hair from the black background; (ii) OEM better separates shadow detail in the clothing, and (iii) OEM renders a slightly wider contrast range between the man’s forehead and the darker parts of the photo, giving an overall impression of a bit more vibrancy without blowing highlights. These outcomes are consistent with what the 30 Neutrals test predicted.
The overall conclusion I draw from this testing is that I would select to use the OEM profile over the custom profile, regardless of what the numbers say about profiling accuracy (probably heavily influenced by the disconnect between profile creation and profile results measurement technologies); this, because perceptually Chromix has provided a profiling solution that handles deep shadow detail in this challenging matte paper better than I could do with my custom profile. Most likely this results from their use of the polarizer to generate the profile data from their profiling targets and perhaps points to an important option for better profiling of matte papers. I have more to say on M3 profiling just below.
Summing up for Red River Palo Duro Etching, not to get lost in the details of profiling it, the paper itself is a very attractive medium. It has an “artsy” feel from the cotton rag substrate and the textured coating. It’s Black rendition and overall color vibrancy is very good for a matte paper, especially with the OEM M3 profile.
To explore M3 profiling further, I asked Chromix to prepare a couple of M3 profiles for me, one for Hahn Bamboo paper and the other for Moab Somerset Museum paper, but this time for the Epson 5000 printer, which has a wider gamut than either the P800 or the Pro-1000 largely on account of the addition of Green and Orange inks. The purpose of this request was to see whether the “M3 effect” discussed above is generalized to more papers, or unique to Red River Palo Duro Etching for reasons to do with the paper rather than the profiling.
Starting with Hahn Bamboo, the basic profile data is in Figure 83:
The M3 profile has a 20% greater gamut volume, which in this gamut range could have a noticeable impact on the vibrancy of real photographs printed on this paper. As well, the M3 profile indicates a Maximum Black of L*10, whereas for the M2 profile it is a more usual L*15. A difference of 5 L* levels in this range should be visible in a print.
Turning to profiling accuracy, Figure 84 shows a more accurate outcome for the M2 profile, which is what should be expected from the foregoing discussion.
The M2 profile has an average dE of 0.75, the variance of 0.18, while the M3 profile measured with an M2 parameter shows Average dE of 3.33 and variance of 2.28.
Looking at Grayscale rendition, the results of the 21 Neutrals grayscale test appear in Figure 85:
Over the relevant range for matte paper, the red chroma curve shows overall good neutrality for both profiles; the M2 L* shows lower dE versus the M3 profile, especially up to L*40. In the M3 profile printed Black values are brighter than what is printed with the M2 profile. While not very accurate based on M2 measurements, it means that shadow detail will be better revealed. We’ll look more closely at that region with the 30 Neutrals test just below (Figure 86).
With the standard M2 profile printed using Absolute RI, there would be no tonal differentiation till after input value L*14~15, where after the closeness of fit with the reference curve becomes excellent. For the M3 profile, L* brightness is consistently several levels above the reference curve, and shadow detail would begin to differentiate from about input value L*10, much as predicted from the profile analysis.
Looking at this same graph using Relative R.I., the preference for the M3 profile holds. RELCOL +BPC reduces the difference between the results of these profiles, but it is still material. While the overall slope of the Red curve from 1 to 30 looks similar in both graphs, the curve shape is not quite the same. It is steeper in the lower portion of the M3 graph as shown by the dotted blue lines. Specifically, taking L*15 as the midpoint on the input scale and a typical Maximum Black printed value for matte papers, the slope of the curve up to input L* 15 is 0.37 for the M2 profile and 0.5 for the M3 profile (*), indicating once again that one should expect superior shadow tone separation from the M3 profile [* M2: (22.27/16.77)/15 = 0.37; M3: (23.59-16.00)/15 = 0.50].
Examination of the Atkinson Printer Test Page and the Romans 16 B&W dark B&W test image largely confirm that the M3 profile provides improved deep shadow tonal distinction with hardly any visible difference between the color rendition relative to the M2 profile (Figures 88 and 89).
Examining the Figure 88 sheets under D50 illumination, the color illustrations are for all intents and purposes the same; however, the grayscale ramp shows slightly more contrast, the Black point being just slightly more Black. Taking this observation into Figure 89, it may require a bit of concentration to see this from an Internet-friendly illustration, but the shadow-tone separation is moderately better on the jacket with the M3 profile, as is the separation of the man’s black hair from the black background. In the case of this paper and profile, the distinction between the two profiles is subtler than observed for the Red River Etching paper, but it is present.
Turning to Moab Somerset Museum Rag, the other paper for which Chromix prepared an M3 profile (Epson SC-P5000 printer), the paper itself is a slightly brighter, slightly whiter product than the Hahn Bamboo. The characteristics of the M0 and M3 profiles are illustrated in Figure 90 (M0 was a better fit for the custom profile with this paper).
Key distinguishing features from the profiles data are: (i) the gamut of the M3 profile is 19% larger than that of the M0: 718K vs. 602K; (ii) The Black point of the M3 profile is L*12 vs. L*17 for the M0 profile; the White points are the same and the chroma of the Black and White points differ very little. Based on this information we expect a moderately richer color rendition from the M3 profile, as well as better deep-shadow tonal gradation from the extended range of the Black tones.
Comparative profiling accuracy is very similar to this paper compared with the previous two. Figures 91 and 92 show the dE graphs for the 24 test colors for the M0 and M3 profiles respectively (be mindful of the scale difference between these graphs, left side). The Average dE for the M0 profile is 0.71 with Variance of 0.12, while the Average dE of the M3 profile is 3.49 with Variance of 2.97. Again, to remind, the M3 profile was created with a completely different instrument and light filtering (“M” Condition) from that used for measuring the printed evaluation target, most likely accounting for a good part of the higher dE results here and in the grayscale tests.
Turning to full grayscale tonal rendition, the Average L* dE for the M3 profile over the L*15~95 range is 1.41, dE Chroma 0.57 and the Total Average dE over the same range 1.99. The heightened dE Luminance values of the M3 profile for roughly the darker half of the L* curve (Figure 93, Black curve) are similar to those encountered for the Hahn Bamboo (Figure 85), but less than those of Palo Duro Etching (Figure 78). The cause of the high L* dE values is that the measured deep gray values are considerably brighter than those in the test image reference file till about L*45~50, where after the dE values are below 1.0.
For the M0 profile, Total Average dE over the L*15~95 range is 0.68, of which L* 0.22 and Chroma 0.46 – excellent outcomes from a proofing accuracy perspective, but the shadow detail from the M3 profile will be superior nonetheless.
As for Hahn Bamboo, I conducted the 30 Neutrals test for both Absolute and Relative R.I. for Moab Somerset Museum (Figure 94 ABSCOL comparison between M0 and M3, and Figure 95 RELCOL comparison).
Recall that ABSCOL R.I shows the results of a rendering unaided by Relative Colorimetric and BPC adjustments. Unaided, the M0 profile shows no tonal differentiation up to input value L*17, beyond which there is a close linear fit with the file reference values of 1:1 input: output. However, with the M3 profile, tonal distinction starts at a blacker level of L*12~13, where after the measured values track the linear scale quite well, but at about 4 levels brighter. This means that the M3 profile has an inherent capability of differentiating shadow detail earlier in the tone scale and brighter; while not so accurate, it is also not erratic – the tracking is quite linear. This outcome is very similar to that seen above for Hahn Bamboo.
With this paper, if one print using Relative Intent with BPC (Figure 95) there isn’t a lot of difference in the rendering of shadow detail. The M3 profile starts one level darker and finishes one level brighter than the M0 profile – so slightly higher incremental tonal separation from the slope of the M3 curve, but it’s subtle.
Now, leaving the world of curves and data, and turning to the test images for Moab Somerset Museum, I was going to include comparative scans of these photos, but I decided not to because the differences between them printed in Relative Intent are too subtle to be observed in a published article. The blackest tones look a trifle blacker in the Romans-16 photo, but the difference again is subtle.
For the time being, my overall conclusion of this foray into M3 profiling is that it does have the inherent capability of bringing out deeper blacks and more shadow detail from matte papers, but printing with Relative Intent, and depending on the paper, these differences vary from being quite noticeable to very subtle. The production of the M3 profiles is somewhat of a big deal because the equipment needed to do it costs about USD 9000+ and I am told the adjustment of the polarizer to optimize the results is quite finicky. A matter for further exploration in this area will be the use of dE2000 rather than dE76 measurements. When the visually perceived results differ by less than the dE(76) proofing data would suggest they should, the perceptual adjustments in the dE 2000 formula may help narrow the gap between perception and data. But this will be for another day. Finally, while I could not do it myself, I have been provided good reason to believe that if one measured the results of an M3 profile with the same polarizing-equipped spectrophotometer, the resulting data would show much greater profile rendition accuracy (lower dE) than reported here.
Red River San Gabriel Baryta Semigloss 2.0
To conclude this round of research and reporting on interesting inkjet papers, I discuss Red River’s San Gabriel Baryta Semigloss 2, used in an Epson SC-P5000. This is a wide gamut PK-ink paper so it deserves the full treatment in the widest gamut 17-inch printer on the market.
Now, to begin with, what’s in a name? Paper naming has always intrigued me. I have this vision of serious paper company executives sitting around a boardroom table testing ideas of what to name their next new baby. There are nine cities, two rivers, one valley and one mountain range with the name San Gabriel, and I haven’t asked the Red River people which inspired the name of this paper. The Baryta and the Semigloss components of the name are more self-informative.
Somehow, Red River got wind of the fact that I like Ilford Gold Fibre Silk (IGFS), so they suggested I try this product, as it is “similar but different”. “Similar but different” is a recurrent theme in the paper business these days. So sure, why not have a look. Those of you who know these other papers will have a good idea of the look and feel of this one.
The basic profiling data (Figure 96) does indeed situate it in the family of papers such as IGFS, Canson Baryta Photographique and similar but different.
Gamut volume is right up there in the big league at about 972K for my custom profile and 947K for the Red River profile, not a significant difference, and both very large. It’s a bright paper, with Maximum printed White measuring at L*97.4~97.5; it will reveal ample shadow detail with printed Maximum Black at L*2.4 with the OEM profile and L*2.2 with the custom profile. The paper white is perceptually neutral with a* and b* being only -0.30 and –0.27 off neutral respectively (0 is neutral). Printed neutrality of the Black and White points is also very good, as inferred from the data above.
The paper has a very mild OBA/FBA content, evident from the Magenta curves in Figure 97 (from X-Rite i1Profiler, measured paper white).
The paper performs well for profiling accuracy, the Average dE for the 24-colour test being 2.03 for the OEM profile and a stellar 0.56 for the custom profile.
By the way, I took advantage of this opportunity to test Ethan Hansen’s finding published in the LaLa Forum recently indicating some advantage to the XRite 1877 patch set versus the 2033 set. The average dE was 0.56 for the former and 0.58 for the latter – nothing significant on the face of it, but perhaps more important is the difference of the variance around the average – 0.027 for the 1877 target and 0.059 for the 2033 target. On balance, Ethan is right – use the 1877 set – fewer patches but more accurate with less variance (Figures 98 and 99, note the scale differences).
The OEM profile (Figure 100) shows an outlying dE result for Yellow, which re-measuring did not alter. I always think of such outcomes as strange and for which I have no explanation.
Turning to the 21 Neutral grayscale (Figure 101), despite the jagged-looking results of the OEM profile, the overall outcome is not bad. Over the relevant range for this PK paper (L*5~95), neither the Luminance nor the Chroma dE values exceed about dE 1.25, and the range of variance from one five-level interval to the next seldom exceeds 0.75, so the perceptual impact would be minimal if any.
The custom profile shows very satisfactory grayscale properties with this printer and paper. Over the relevant range, only one Luminance dE reaches 0.64, the remainder being at or below 0.5; the Chroma dE values are below 0.5.
The 30 Neutrals test shows good tonal separation in the deep shadow tones, whether for the OEM or Custom profiles, the Custom profile providing a remarkably close fit to the linear curve (Figure 102) – it doesn’t get better than this.
Unlike for IGFS, there is no reported longevity testing data for this paper yet; however, the company is of the view that longevity should show as satisfactory once it does get tested because of the standards it adheres to, per their website: <<Archival grade materials; Baryta whitener layer (pure barium sulphate); Base stock is acid free, lignin free, and buffered with calcium carbonate; FOGRA Certified ISO 9706 (paper aging standards)>>. I have no view on the implications of these properties for years of stability due to the paper alone (recall the ink plays a major role in this too).
My testing with the above tools and prints of photographs indicate this paper would please anyone who likes the other papers in its class. Colour vibrancy, resolution, dynamic range, shadow detail, printing accuracy and the gray scale all look fine. The surface texture is relatively unobtrusive, which helps for non-distracted viewing of the photo itself. In sum, a nice paper I would use.