Lots of misleading info on this thread. First of all, lets talk about the good ole standard, Dye Transfer. Besides storage and display lighting variables, it just all depends on the specific dye combination involved, and there were lots of them. There were not only at least four different manufacturers of dye transfer supplies, and all kinds of variations to the process, but it was amenable to all kinds of custom tinkering, including dye options. Some dye transfer public installations held up under twenty years of strong light due to special dyes. So blanket statements are inherently misleading. Then Cibachrome came along, which is a somewhat more cut and dried subject. Their dyes seem to be very stable in dark storage but don't like direct sunlight or artificial UV sources. Inkjet is analogous to dye transfer because all kinds of inks are potentially involved, and some of these ink particles are basically lakes (dyed inert particles) containing some of same dyes once used in dye transfer printing. There are also various pigments involved; so again, making any kind of blanket prediction about permanence would be irresponsible and misleading. When it comes to true pigment processes like color carbon, carbro, and Fresson there's a question about the integrity of the bond between sandwiched layers. Carbro prints are especially prone to cracking and blisters; but this has been detected to some extent on all of them. So it's not only about the hypothetical permanence of individually tested pigments. The entire print has to have mechanical integrity too. Gum and related Quickprint largely fall in another category visually, while several attempts to modernize carbon like the Polaroid Permanent process, Evercolor, and Ultrastable relied on halftone techniques that visually defeated their considerable expense trying to simulate more conventional forms of prints. One of these processes was ridiculed as having " all the beauty of a plastic place mat", and did indeed resemble a laminated halftone poster. It quickly went bankrupt, but I still have token samples. Chromogenic prints have conspicuously been improved - but by how much? Now that Cibachrome is gone, I happen to work with Fuji Crystal Archive products, including their superb polyester-based product, Fuji Supergloss. I probably won't be around as long as the prints themselves, but do suspect they'll show some conspicuous yellowing in a few decades, perhaps even prior to noticeable fading per se. So there is no real silver bullet out there. Even paintings involve choices where their respective pigments are rarely created equal. And some of the treasured works of both old and modern masters are still around only because staggering amounts of time and money have been spent restoring and conserving these paintings. In other cases,
the pigments involved simply cannot be supplied on industrial scale. The grade of lapis blue which was used on the ceiling
of the Sistine Chapel, for example, is now more expensive per ounce than gold. No wonder the Popes like Leo bankrupted entire nations with their decor projects. And even he went deeply into debt.