What makes dye ink so attractive is the ultra smooth gloss of the finish. Can you tell the difference once face mounted? I would think at that point the print surface isn't a real factor any more.
Wayne, I think a lot of it has to do with how receptive the surface is to ink.
I've been playing with some BC Silverada Canvas and it just doesn't suck in the dye ink well. Some may even evaporate. They also mention pulling back the density in the printer driver (on the Canon 9000 II dye ink printer) to keep ink from puddling on the surface - and I found that true. I've made an error and sent the same print through the printer twice so the dye ink built up and the dMax was higher as the areas that were off register on the second pass that showed the density buildup of the dye ink. Probably 4 passes would hit max dMax, or one pass of pigment, but the canvas just isn't that receptive to "dye" on one pass - or many. One pass is a muddy warm black, and a different canvas is definitely blacker.
Sort of dismayed by the BC material for dye use since so many other metallic papers work well with my printers and the various dyes (OCP & Cone) soaks in but lets the metal shine through. If it were called "Opalescent" or "Pearl" and "Not recommended for dye ink due to layer construction" I'd be $130 richer.
Sidebar: I'm also seeing issues where the OCP dye is much blacker than the Cone dye too (e.g. OCP dye showing dMax 2.4, and Cone dye is 1.9) on other type of paper, but that's another issue too. Canon for the "dye" win so far over my Epson. Hopefully, I can get that addressed, but it might take another encapsulated black dye from someone else to pull off.
SG