HP Photosmart 9180

HP Photosmart 9180
and HP Papers

This is an inkjet printer with a 13” throat and a capacity for A3+ prints. With a footprint of around 27” wide by 17” deep, it requires an additional 14” behind when using the flat-fed art media. The front tray extends by an additional 7” or so. Physically it is the same size as the Epson 3800, although the Epson runs up to A2 size. The 9180 uses the Vivera eight cartridge ink set consisting of pigment inks in the following colours: Photo Black, Matte Black, Light Grey, Cyan, Magenta, Yellow, Light Cyan and Light Magenta.

There are a number of papers available as shown by this selection image in natural light and

ABOVE in Ultra Violet light. Paler blues show increased OBA activity.

Unpacking and Installation

There is nothing unusual to report here, a similar amount of sticky tape needed to be removed, comparable to its rivals. We noticed right away that trays and sliders were sturdy and well engineered , as indeed, is the whole machine.

Speed and Connectivity

The printer is provided with both USB and network connectors. We set up a network-connected printer and installed the drivers and profiles on a Windows XP box, using Photoshop CS2 to output test prints from. An A3+ print took 6m 16s to print, in line with the 3m10s reported elsewhere for an full coverage A4 print. The feed tray can hold stacks of thinner media but not the fine art and canvas types.

The printing interface for the HP is straightforward and simple to follow. We did not use the installed Photosmart Print Plug-in much, although this was even easier to use – a real joy in fact! It is accessed by opening the print in Photoshop and then going to the Automate menu and clicking HP Photosmart Pro Print. Initially only the HP media is shown but it is possible to set-up new papers and assign profiles.

The printing process did not exit gracefully when a print was cancelled. It stopped instantly when the ‘cancel’ cross on the front of the printer was pushed, however, we could not delete the spooled file from either the printer dialogue box or from the spool folder. When we rebooted the computer the spooler kicked into life and printed the file perfectly. This behaviour is more an interaction between Photoshop and Windows, as far as we can tell, at least the HP allowed us the stop printing in a timely fashion.

Paper Options

HP provides a large number of papers under their own name, along with a number of co-branded papers with Hahnemühle. The options increase yet further if you include the roll media available for the HP Z2100/Z3100 series of printers. A useful selection is included as a swatch set inside the printer's packing box. These are tabled along with those other media that come with A3+ sizes from the ‘large format’ media catalogue. The 9180 has now been around for a sufficient time that many third-party paper providers have profiles available on their web sites.

Measurement of the gamut volumes for the various profiles installed by the HP driver suggests that the actual profile data are doubled up, with only the name of the profile being changed. This is somewhat sloppy technique in our view, especially from the manufacturer of the printer. It assumes for example that the white points are all the same which is patently not true.

Testing

HP Advanced Photo Paper Glossy 250gsm

This is likely to be a bread and butter option for the professional photographer who needs a gloss surface. The paper has a slightly cool base white with just a hint if a lift at 440nm but no really obvious OBA activity in the UV-booth. We colour audited the printer performance using the HP profiles provided with the driver and also built our own.

The HP profile produced a reasonably pleasant print. The skin tones were desaturated by about 11% and rotated a couple of degrees towards magenta. The overall error was evenly split across hue, saturation and brightness channels. The lightness values were the most accurate and averaged about 2½% too dark.

Bespoke profiling tightened things up considerably, bringing the average error down from 5.49ΔE 2000 to 3.92ΔE 2000. The bulk of the errors shifted to the lightness channel leaving the colours about 5% too dark across the gamut. The residual errors in saturation were concentrated in the upper quadrants of the Lab chart, that is the greens, green-yellows, yellows, oranges, and reds. The highest error was the Moderate Red at 5.5 ΔE 2000. Overall the combination of this printer, paper and profile could not reach contract proof standard but the resulting prints were quite acceptable commercial print quality for professional output.

To see just what a finely tuned printed could do we include data from a client trial we conducted recently. This was using Calumet Brilliant Gloss paper – not one we would choose for proofing normally. However, after some effort we did tune it to contract proof standard and the data are tabled for comparison.

The gamut volume was measured at 788,658 for our profile, higher for the HP profile at 869,582. The Dmax was a deep 2.24. The metamerism was high and this topic is covered separately. The Granger chart from all profiles was smooth.

Greyscale testing

Making neutral, well-graded monochrome prints is the ultimate test for an inkjet system. The HP 9180 can make mono prints using composite inks (ie all the inks) or using grey inks only. We tried both options using the same image onto the same piece of paper. We turned the colour management off in Photoshop for one set and then allowed ‘Photoshop manages colors’ for the other giving four prints in all. The resulting prints were different because of using composite inks or grey inks, but the colour management settings had no influence on the outcome. Use of all the greyscale settings created prints that were too dark, by 13% using all inks and by 22% using the grey inks only. These data are plotted alongside a greyscale image made using full colour inks and our own profile to act as a reference. The HP versions were not up to standard and some experiment would be needed before this route could produce exhibition-quality prints. There were only single point differences in the base colour of the 50% density greys. The grey ink only versions were a little cooler. The colour ink versions shifted towards cyan in north daylight, the grey ink only version remained neutral (see metamerism section below).

Metamerism

We noticed the metamerism quite early on in our testing, before we actually made any measurements. For the full colour (ie all colourants used) greys the metamerism peaked at 4.5 Lab ΔE when shifting from D65 light measurements to Tungsten Illuminant A. Things were better with the grey only inks when the metamerism was measured at 0.9ΔE Lab. The change of metamerism down the tone scale followed the hump shape found with Epson K3 ink, although the peak in the HP was very much higher (ie worse metamerism).

The effect of the metamerism was that greyscale images that were neutral in D50 or D65 light were green in natural north daylight and magenta in tungsten light.

Real prints

It is important to make real prints when reviewing printers; test targets sometimes give a false impression. We made a number of excellent prints. The very slight banding that we observed in solid blocks of 50% grey were not obvious in real prints with their mixtures of tones and details. It is also worth noting, before we leave the gloss material, that the higher metamerism in this paper has been replicated with some Epson semi gloss paper although not to the same degree. As we show later the metamerism was very much lower on other surfaces.

HP Hahnemühle Smooth Fine Art Paper 265gsm

With sufficient weight to pass Fine Art Trade Guild standards, this is a 100% cotton rag, acid free offering that is classed as ‘bright white’. There is no evidence of any OBA activity and so the brightness is assumed to be due to premium quality materials in the base.

With fine art media selected in the printer driver the 9180 chooses the matt black ink option seamlessly. The only care you need to take is to ensure that there is sufficient space behind the printer – it pushes the paper further back as it takes up an aligns the sheet.

We tested the paper using the profile provided with the driver CD and automatically installed with the printer. Unusually for a fine art matt media, the lightness error was the smallest. This was backed up with a good Dmax of 1.66, good tone separation down to 15RGB points and good highlight separation right up to 252 RGB points. The metamerism was measures at 1.0 Lab ΔE (D65 to Tungsten on 50% grey). The look of the print rather belied the rather poor statistics. The bulk of the error lay in the saturation component. Most viewers would deem the print to be commercially acceptable for portrait work.

In terms of gamut volume the Vivera ink set is slightly lower than that of the Epson 3800 onto Hahnemühle Photo Rag, even though the HP Dmax is greater

HP Hahnemühle Watercolor Paper 210gsm

This paper has the characteristic rough texture and is similar (identical?) to Hahnemühle Albrecht Dürer. It is too light in weight for Fine Art Trade Guild limited edition print standards, but is a good-looking paper even so.

Assuming that we are correct in aligning the paper to Albrecht Dürer we are able to make a direct comparison between the Epson 3800 and the HP 9180, using publicly available profiles. As with the other profiles we tested, the HP variant is lacking in saturation and it is this component that contributes to the highest error. The skin tones were 10% desaturated but on the nail in terms of hue. Conversely the Epson 3800 using Hahnemühle profiles was on the nail on saturation but rotated about 5° too yellow. As with Smooth Fine Art, this was a good-looking print despite our reservations on the statistics. The Dmax was 1.66, the metameric index was 1.1; the shadows were differentiated down to 15RGB points and the highlights to 252 points.

The Hahnemühle (web site) profile for the 9180 is significantly better than the HP one provided with the driver Honours between the Epson 3800 and the HP 9180 (both on the Hahnemühle profiles) are more evenly shared out – some are up some are down.

HP Artist Matte Canvas 380gsm

This is a matt finish canvas substrate made up of a cotton polyester blend. The intrinsic ability to print to canvas is one area where the HP has an edge over its rival Epson 3800. The canvas is feed through the speciality media slot and performed flawlessly for the prints that we made. Like all canvas media it is likely that a clear giclée varnish will be applied to the finished print. We tested the HP canvas with Clearshield UV varnish applied as a single coat with a soft brush. There was no ill effect from the aqueous-based varnish.

Before Coating

The as-printed surface has an average error of 11.5ΔE Lab/5.2ΔE 2000. The errors were split evenly between the hue, saturation and brightness channels. Skin tones were about 11% desaturated and rotated a couple of degrees towards yellow. The greys were neutral, mapped to the base white. Dmax was good for a canvas at 1.60, the greyscale linearity was held down to 20RGB points. The whites had detail up to 252 RGB points. The metamerism was spectacularly low, at just 0.3 Lab ΔE points. As with other results from the printer/profile combination the saturation was low overall. To put some perspective on this the hat of the model in our audit test sheet should be cream, it was desaturated to a white.

After Coating

The effect of the coating was to deepen the tones across the gamut by about 5%. The Dmax was deepened from 1.6 to 2.10 and the saturation was slightly increased. The error in the lightness channel was doubled due to the darkening. Despite this, we fancy that most observers would go for the extra punch of the coated sample. The correct procedure would be to make a swatch set for profiling, varnish that, and then use it to make the profile. Then the print would come out light, before varnishing, but correct afterwards.

Summary

In spite of days of profiling, measuring and testing we are left with quite a few unanswered questions about this printer. Most of our baseline data are from Epson K3 ink sets and we have only occasionally tested Vivera ink. We are not even sure if the HP 9180 should be rightly tested against the Epson 3800 or the Epson 2400, then you have to throw the Canon 9500 and the HP Design Jet 90 into the ring. There are too many variables to test everything but our overall feeling is that the HP 9180 is capable of doing a thoroughly professional job but that it is a little behind the Epson 3800 in absolute quality. This concerns the gamut volume, quality of profile, banding and audit precision. However, there is a significant cost difference. Against that the 3800 will do slightly larger prints and has much larger cartridges. Which ever way you decide the 9180 will not let you down.

Read related digital articles Colour Correction

What our members say
Why I like the Societies: Informal and informative - Tim S
Find out more about the Societies here

Convention testimonials Neil Redfern: Had an amazing time at the Convention! So nice to catch up with friends & make so many new ones. HUGE thank you to @TheSocieties
Find out more about the Convention here

Photo Quote: Nothing is ever the same as they said it was. Diane Arbus

There are 227 days to get ready for the SWPP Convention and Trade Show at The Hilton London Metropole Hotel ...
which starts on Tuesday 8th January 2013