Tuesday, 23 February 2010

Print Pricing

The usual practice today for fine art prints seems to be selling limited editions, artificially restricting the number of prints to a fixed number, say 20 or 100 for example. As long as the photographer keeps their end of the bargain and does not print more than the promised number, the buyer can be guaranteed to have a limited edition product with a degree of exclusivity. The smaller the edition, the greater the photographer could price their prints and the more exclusive the photo.

I have two issues with this. One is that the photographer is handcuffing themselves at the start. They have to guess the tradeoff between price and exclusivity. The amount of money a print edition can generate depends on the product of price per print times the number of prints sold, the latter capped at the beginning by the edition quantity, and the former set as high as possible but not too high so that there are unsold prints in the edition. The problem for the photographer is that if a particular print is unexpectedlyy popular, then the price may have been set too low at the beginning and the photographer misses out on the volume of prints they may have been able to sell had the edition quantity been higher. If the photographer places the price too high, then there are unsold prints in the edition which again is a waste, considering the greater exclusivity which could have been guaranteed if the edition quantity had been set lower and hence justified a higher selling price. Finding that balance between price and exclusivity is tricky, especially for someone new to the field.

The other issue I have is that limiting the number of prints to a fairly arbitrary fixed number seems to be contrary to the spirit of photography, where there are no natural constraints to the reproduction of an image, in contrast to painting where one can usually point to an ur-painting. Limited editions try to mimic the exclusivity which buying a Monet, or van Gough gives a collector, which is something which I am uncomfortable with.

As a compromise, I am developing a system of print pricing which allows in principle an unlimited number of prints, however in practice should be self-limiting. If an image is unexpectedly popular, the extra demand can be catered for. This is achieved by using a sliding scale, low numbered prints are less expensive than higher numbered prints. Early buyers of a print are rewarded by a smaller entry cost. If a print becomes popular, late buyers pay extra. This should encourage speculation yet not denying collectors who are able to buy any print at any stage, albeit at prices reflecting the popularity if the image.

For simplicity, I have adopted an exponential pricing structure. Print N=0 is the photographer's proof version and is nominally set a price P. Each additional print is numbered and priced at a fixed proportion r of the price of the proceeding print, i.e. print number N is priced at P*r^N, where N=1 is the first print to be sold. There are two parameters to be set, P and r. The nominal initial price P set an overall scale (perhaps depending on the size of the print, e.g. P=£10 for a small print, £100 for a large print), whereas the rate r sets a scale for the exclusivity. As an example, if I set P=£100 and r=1.023293, then the first sold print would be £102.30 whereas the 100th print would £1000. If I managed to sell the 400th print, it would be £1 million. Naturally this would limit the edition quantity, depending on just how popular the photo was.

For small prints, it may not make sense to have limited editions, a flat pricing may cater better to non-collectors. But for large prints, this pricing structure may have benefits for both the photographer and buyer as it naturally self-limits to a level dictated by the popularity of the image.

Thursday, 18 February 2010

Epson 2100 Print Head Alignment Woes

I recently switch over my Epson Stylus Photo 2100 from matte black to photo black inks so I could do glossy or semi-gloss/pearl prints. However, I noticed that my prints were displaying a really unfortunately problem. In certain dark tones, the texture was fuzzy, blurry, or basically reminiscent of an impressionistic painting rather than display the detail that should have been there. My suspicion landed on my use of the print head alignment utility, especially the last part of the test which involves choosing the least grainy patch for different print densities. I always had trouble discerning which of the patches in the #3 column was supposed to be least grainy, they all look equally bad.

I probably compounded things by running the print head alignment utility using photo paper instead of plain paper. Also, it is recommended that high speed printing be turned off, something which I didn't make sure of. Hence, I probably managed to put in totally wrong settings for this part of the alignment test.

Trying to get back to a reasonable baseline has been problematic. From a support chat with Epson, it seems as if the alignment settings are actually stored in the printer, but I couldn't find out whether there were a series of printer button presses which would reset the printer to default. In the end, it seems that the only "solution" is to uninstall the printer driver, reinstall, set up the default printing options to disable high speed printing (turned off finest detail, and smoothing for good measure), and run the utility again and try to stick to the default numbers, (8,8) for the vertical line test, (4) for the horizontal banding test, and (4,4,4) for the graininess test.

Update: The actual problem was the fact that the paper could not absorb the ink fast enough leading to coalescence. After a lot of experimenting, I found that by using a different media setting (Watercolour paper at 1440dpi) and reducing ink density by 15%, I could limit the coalescence. The print quality is slightly reduced and the colours aren't accurate any more, e.g. red prints as orange. This requires a profile for the ink+paper combination which I managed to achieve using the ColorMunki. The final results are more hue accurate but the density is a bit higher than I am used to. The detail still seems to be there in the shadows but requires quite a lot of light to see. A bit more experimenting with media setting might be required in order to get better results.