Paper Chase - The half term report

Paper Chase has now been running for 6 issues and we have not yet even left art papers behind us! Encouraged by your support we are carrying on but it instructive to take a little time out to consider what we have learned so far.

The paper suppliers are a bit like mushrooms. We no sooner pick them off than another one pops up and from afar they like quite alike! Even though we have a backlog of paper and data to give to you, at Photokina this year there were new ranges and new papers from Croppers, Innova, Permajet, Olmec, Hahnemühle, Epson, Ferrania, Kodak, Calumet and Felix Shoeller - and they are only the ones we saw! We have not even started on the gloss, lustre and semi matte ranges although we have started some analysis.

Paper Chase

the midterm report What have we learned from our chasing the paper trail? Well paper is a complex substance made even more so by the addition of an ink receptive layer followed by a ladling of inks. It is the subtle differences between the ways the various inks diffuse into the coatings which is the source of all our woes! This is what causes the tonal scales to wander about in a seemingly random fashion. Having made and tested 498 paper/ink/printer combinations so far in Paper Chase (a total of 11 million data points) we are starting to build a picture of how printers vary between each other and between models.

In a nutshell we find that the Epson 7600 and 4000 models are less variable than the Epson 2100 which is less variable than the Epson 1290.We have also tested a variety of other printers and some continuous ink flow systems. In only one instance during our 500 tests did we find that bespoke profiling a printer ink combination failed to improve the colour accuracy of the output. The worst performing printers were truly ghastly with absolutely no hope of producing a commercially acceptable print from them. Another thing we have come to note is the difference between peoples' perception of print quality. We have had photographers expressing mild dissatisfaction over awful prints and at the other end of the scale photographers tearing out their hair over prints which cannot be realistically improved upon - this is one reason why we wrote and developed our ColourAudit routine and data analysis. It is a risky business to reproduce the effects of poor print quality in a magazine but our composite image shows two of the worst offenders either side of a pure digital file. We have chosen not to reveal the source of the print as some of the heritage of the systems was not fully understood and mistakes do happen. Not all of our bad actors were inkjet printers, we have also examined a number of other technologies. There are a number of ways of expressing colour error, but for simplicity we have plotted the error in the neutral greys from some of the poor printers and compare that data with a set from a profiled inkjet.

What does the graph teach us? Well the problem is sometimes but not always the operator, there are occasional machines, which behave badly. One advantage of the Epson 9600, 7600, 4000 series is that they are the so-called "D2" range, made to higher tolerances - and it shows! Other than that, operator based errors have involved printing on the wrong side of the paper, using machines without doing a nozzle check (us included - we build one profile without using cyan at all!) and simply setting up either the profile build parameters incorrectly or the printer settings incorrectly. As we said last issue, setting up to do certain print jobs can involve up to 28 mouse clicks. Experience with RIPs has been mixed, they have produced some good and some poor results. Paper chase

The quality of profiles out there is rather mixed. The manufacturers have got to realise that the use of their product will stand and fall on the initial results the user obtains. If the manufacturer puts a poor profile on their web site and it subsequently produces a poor print, the chances of repeat orders are slight. Make no mistake there are some scabby profiles out there - you can check for yourself on how they have been made by opening an icc file in "Notepad" and looking through the raw code of the file. This tells you how many swatches have been used in the generation of the profile (more the better) and which software and spectro was used. The final arbiter is always the print though. The more progressive suppliers have got wise to the problem and in some instances it is the enthusiast and specialist user groups who have led the way. The result is that most of the major manufacturers now have good quality profiles on their web sites. This will not assist those who have machines which behave some way from the norm (like the ones in the graph) - if you are the unfortunate owner of one of these, your only option is to have a bespoke profile made.

Metamerism

Just when we thought we were getting a real handle on metamerism, the water has got a little muddier. It started when we tested a sample print of Epson Premium Semi Gloss and noticed by chance that it was abnormally metameric in low energy bulb light, in fact it had turned quite pink, visually. If you wish to gauge what we are relating, it was close to a Pantone 437c, (that is Lab co-ordinates of 50; 10; 0 (RGB 125,109, 115)).You can go to the Photoshop Color Picker and look at the colour itself. This came as a bit of a shock as the Epson Ultrachrome ink set has normally been quite low on metamerism. The index of metamerism was around 4.0 to 4.7 Delta E depending upon how it was measured. This is more than twice the normal value for an Ultrachrome ink set on a neutral 50% grey. What we are finding perplexing (and infuriating) is that the maths does not properly support our visual observations; however, the effect is high enough to eliminate Premium Semi Gloss as a contender for monochrome use particularly if you wish to match grey tones in an album. This is a topic which will bear some more investigation in the second half of the term! We had three samples to compare which were perfectly matched in a D65 viewing booth but markedly different in low energy bulb light and less different in tungsten light and a normal fluorescent (domestic) light. It is not that they were different which surprised us it was the level of difference. We gained a slight clue from the spectral output trace of the bulb, which was a riot of spikes, criss-crossing the graph. Perhaps the moral of the story is to avoid low energy bulbs in you studio but even with this precaution you may have no control over the light which your client views their print in.

Two Glossies

Just to get the ball rolling on the glossy and lustre papers we have pulled out two of the papers to give you a taster.T hey are the Kodak professional Gloss and the Calumet Brilliant Gloss.

Kodak Professional Gloss

Launched at Photokina, this is a medium gloss paper weighing in at 255gsm and 23 microns thick, about the same as an en print material (maybe a little thinner). The back of the paper is embellished with the logo "Kodak Professional Paper". The base white of the media is a cool blue and it does contain optical brightening agents; there is a lift of about 3% at the blue end of the spectrum due to the fluorescence.

Bearing in mind our remarks at the head of this edition of Paper Chase we were anxious to test the profile provided on the Kodak web site.The one we used was named KproIPaper_EpSP2100_v1.icm.This is quite a large profile built from 514 swatches, presumably using Kodak's own colour management profiling system. The gamut volume of the profile was 816,000 putting it in the upper bracket and a little below the HiFi ink system of the R800 reviewed elsewhere in this issue. We tested on our Epson 2100 using Premium Semi Gloss photo paper as our media setting, 2880dpi as our printer resolution, High Speed was set to OFF BPC was On and we used a perceptual rendering intent.

The Colour Performance

Whole Gamut

The print was quite pleasing with an overall error in colour of 3.9 (DE2000) points compared to an average for bespoke profiled 7600 machines of around 3.3 on the same scale. Measured on the more common Lab error this averaged at 5.5 delta E which should be assessed against a contract proof standard of between 4 and 7. So, for an off-the shelf profile this was a good result, one of the best we have tested. There was an overall drift towards green of about 5 Lab points on the a and b scales. This was most noticeable in the greys, the most sensitive part of the perception of the human observer. The overall error in the lightness channel was to make the print between 2 and 3% too dark.

Paper Chase Flesh Tones

The all important flesh tones had an overall error of 2.84. The errors in the Lightness, Hue and Saturation channels broke out to 2.1; 2.9; and 0.36 - hence the error in the hue was low. The graph shows the plotted skin tones and the resulting print values. The saturation is maintained for most of the sweep of flesh tones and the hue just starts to push into the more yellow part of the quadrant for the paler skin tones.

Photokina the paper round up

Hahnemühle had new papers on display, some of which we have already had a look at. Significantly they have extended their range to include 13" and 17" stock for Epson 2100 and 4000 owners as well as introducing a complete new range following their collaboration with Lumijet, the famed American media supplier. The Lumijet Preservation (4 papers) and Portfolio (5 papers) ranges are now to be marketed in the UK. Hahnemühle have always made at least some of the Lumijet range. The Preservation series has been re-engineered for use with pigment ink sets. Also newly announced from Hahnemühle are board finishes at 1.2 mm and in sizes 36x44; 36x22 and 24x30 inches. We have tried these out and they look really nice.

The press release also mentions Canvas Satin which we assume is the 340gsm material we tested for the issue before last. This is one of the best canvases we have tested to date although it is not fully water resistant. Other new papers were a White Velvet 268gsm (similar to German Etching but with a brightener) and glossy 270GSM a premium high white paper for the photo retail industry.

We called at the Ilford stand, a little surprised to see them at all (as were others apparently). In terms of supply security for papers everything seems OK. They have decided to stop selling machines but will keep on selling a RIP for their media. The Studio Range of papers has been subsumed into the Galerie Range, where it came from despite claims to the contrary! This, the most profitable part of the Ilford empire, is likely to survive the knife of the receiver as the Moberly site is sold and business is moved to Switzerland.

Epson have rationalised their range a little and added a PremierArt Water Resistant Canvas, a really lovely 350gsm media which delivered sumptuous images and tremendous colour accuracy over the gamut. They also had Ultra smooth Fine Art Paper, a 250gsm highly calendared finish. They now have both Velvet and Somerset velvet Fine Art Paper and many of their materials are available in roll sizes from 13" to 44".

Felix Schoeller had a huge stand and they launched their new range of digital media Ultra 9. Available in three weights from 240 to 300gsm the prints have what they are calling a Glacier finish a kind of light lustre.The print provided certainly looks nice quality. It is described as having an enormous colour gamut, which is a puzzle; it looks quite white to me! It appears to have a certain amount of optical brightener in its PE-micro porous coating.

All the other big paper names had a presence at Photokina, the sharks are obviously circling the lucrative digital market expecting a feeding frenzy to break out over the next few years.

The Grey Tones

The greyscale linearity was good ranging down to a Dmax of 2.15 with even gradation all the way along. The blacks blocked up at around 20RGB points, better than many profiles we have tested. The grey tone balance was biased towards green (see graph). The cast was most obvious in natural daylight and D65 illumination, more neutral in low energy bulb light and quite close to neutral in tungsten light. The metamerism index was higher than we normally find with an Ultrachrome ink set on an art paper, mimicking the effect described above in this feature. As the greys move away from pure neutral the metameric index rises (there is more colour for the effect to leverage upon) and so we plotted the changes over the set of near neutrals placed in our audit target. The graph shows the grey of the print centred upon the Lab colour point a= -4 b= +3.The surrounding data points are those of the patches surrounding the image of the Spanish Dancer. The big triangle in the lower left quadrant of the graph shows where a perfectly matched profile would have mapped the greys, the base colour of the media. The metameric index of the most neutral swatch was 5.1, about twice the value we expect from an art paper and a well-matched profile. Overall though the grey errors were good for a canned profile and would satisfy all but the critical monochrome specialist. The metameric shift towards green is the most unpleasant of the possibilities and we would recommend adjusting the image in Photoshop ahead of printing if using this profile with a 2100 that behaved in the way ours did.

Calumet Brilliant

This is a new paper stocked by Calumet. At £14 ex VAT for a box of 50 A4 it is very competitively priced .The full range includes a gloss, a lustre and a matte. We have only looked at the gloss and used an Epson 2100 to test bed it. We bespoke profiled the paper so we could see what it really does.T he paper is 270gsm, 26 microns thick with a very smooth gloss finish.

The Colour Performance

Whole Gamut

We had two profiles with which to measure gamut volume. Profiles are available from the brilliant paper web site at http://www.brilliantpaper.com/brilliantdigital.html. We also had our own. We measured slightly higher than the canned profile at 835,000 typically high for a quality paper and Ultrachrome ink set. Straight away then we could see that there was no compromise to colour quality from this inexpensive product. When profiled the error overall came right on our average for bespoke profiling at 3.40 DE2000.The print also looked very good. The base tone is slightly blue and the paper was a couple of points behind some of the other papers around in base reflectance. There was a slight lift in the spectral response at the blue end of the scale but any optical brighteners (if present at all) were quite unobtrusive. As one would expect our greys were accurately mapped in to the base tone. Overall the print was exactly the correct tone density. With a slightly lower base tone the paper would make a good proofing media for many prepress applications.

Flesh Tones

The flesh tones were a little biased towards magenta and desaturated. However the error was small. The average error in Lab units was 5.7 and it was the rotation towards magenta that was the significant part of the error.

The Grey Tones

paper chase The greyscale linearity was excellent, a straight line all the way down to 7% brightness, a Dmax of 2.05.There was no detectable tone cross over in the bias of the greys they were all mapped extremely close to the base tone of the paper. There was a slight warming of the tone in low energy bulb light and the metameric index was a middling 3.8 Lab Delta E.The shadows held detail all the way down to 20RGB points. Overall this was an excellent greyscale image.

Conclusion

This paper range seems well worth investigating, it performs well and is very competitively priced. It is available in A4, A3, A3+ and 24" x 30m rolls. At £89 ex Vat the roll is a good price.

Paper chase

The SWPP 2008 Convention was an outstanding success,
we have 178 days to get ready for the 2009 convention - which starts on January 14, 2009

Photo Quote: It is light that reveals, light that obscures, light that communicates. It is light I "listen" to. The light late in the day has a distinct quality, as it fades toward the darkness of evening. After sunset there is a gentle leaving of the light, the air begins to still, and a quiet descends. I see magic in the quiet light of dusk. I feel quiet, yet intense energy in the natural elements of our habitat. A sense of magic prevails. A sense of mystery. It is a time for contemplation, for listening - a time for making photographs. - John Sexton