On 2019-03-07 10:07 p.m., Graeme Gill wrote: > Michel Dänzer wrote: > >> Yep. The alternative is that the different mechanisms clobber the >> hardware LUT from each other, which sucks from a user POV. > > Which user though ? > > It certainly does the opposite of suck if you are a user > who wants reliable color management, and so want a simple > and direct mechanism to set the post frame buffer manipulation > to a known state. > > What does suck if you are a color critical user is that > what used to be a reliable system (i.e. "run the profile > loader") just became unreliable due to a system update.
It was never reliable for that. Other clients using any of those mechanisms could always interfere, at least for the RandR compatibility output. > From this perspective I'm puzzled as to why such a change > was implemented. To make all these mechanisms work reliably and consistently at all times. >> Welcome to >> the wonderful world of "colour management" in X, please pick your >> poison. I guess you can see why Wayland has a different design. :) > > X11 has only got that way with such changes in behavior. It's > been pretty reliable up to now, and at least is possible > thanks to some foresight on the designers and implementer s part. If you have specific suggestions, please post them to the xorg-devel mailing lists or create a merge request at https://gitlab.freedesktop.org/xorg/xserver/merge_requests . -- Earthling Michel Dänzer | https://www.amd.com Libre software enthusiast | Mesa and X developer _______________________________________________ wayland-devel mailing list [email protected] https://lists.freedesktop.org/mailman/listinfo/wayland-devel
