<sigh>We all know it's possible to build "custom profiles", what it isn't possible to do is know which actual printer Blurb will use for any individual book. Hence why any profile they supply will only give a general idea of what you might get, not a specific output profile.
<sigh> clearly this is a topic that confuses some of the people posting here.
This has nothing to do with custom or non custom profiles!
ICC profiles define device behavior or they do not. Blurb's not only doesn't, the bigger issue is, their process control sucks! I don’t know if some here purposely trying not to understand this, or if they are really struggling with it.
I sent the SAME RGB values from 24 patches to Blurb over the course of a year. They are a MILE off in terms of matching. No profile defines both or either I suspect but it doesn't matter. Even the SAME book is a mile off between the cover and the inside yet some here continue to believe the ONE profile Blurb supplies is of any use. It isn't. Nor do they conduct adequate quality control; patches should not be 19dE off from each other, ever!
I've worked in shops that have far more digital processes, spread across the world than Blurb does, and yes, they can be managed such they never produce a single dE error over 5! In EVERY press in operation. People like Rhossydd, Tim and now presumably Blurb have
no such experience setting up color management systems with profiles that notify a client within minutes if their press run exceed a set of deltaE metrics. It's called process control. it is time consuming and expensive and hobbyists labs like Blurb, the labs
some here use for soft proofing and utterly inconsistent print output use them too. Now you know why I do not, expect for actual colorimetric testing so few here can conduct and a few do not understand.
NO! Their one profile is
not a general idea of what you '
may' get, and if you understood the ramifications of a dE 19 (or even dE9) means, you'd retract that non colorimetric comment really quickly.
Blurb can't maintain process control. Blurb's profile is not based on their output unless and perhaps you're lucky enough to print on a day when the moon and star's align perhaps. The next day, the next month, all bets are off. They were off 19dE in one example; that's horrible. Again, without data, you're just a person with an opinion. My data suggests it is based on misunderstanding of maintaining the ideal behavior of ICC profiles for large numbers of digital presses. I'm sorry again if actual data has ruined your day.