by mankrip » Wed Sep 16, 2015 7:00 am
I've pushed my health to near death until the end of August, and got a shoulder injury that hasn't fully healed yet. This somewhat affected my sense of humor. Anyway, enough of that.
On the upside, every time a new good idea pops up in my head, I write it down on paper, so someday I can implement it. And there have been quite a few good ideas.
As an example, I've given quite a lot of thought about color rendering, coming up with very optimized algorithms for color blending and for colored lighting, but both features seems to be somewhat mutually exclusive; an ultra-optimized high quality color blending algorithm would impossibilitate an ultra-optimized high quality colored lighting algorithm, and vice versa. I'll only figure out the best algorithm when I get enough time to dive into the code again, as I suspect it should require some major refactoring of all palette-related code.
In retrospect, trying to implement FitzQuake's "big map" support was a mistake. Retroquad wasn't ready for that yet, not in a clean way. Also, the FitzQuake protocol has its own limitations, some of which Retroquad didn't have, and I don't want to restrict my engine's features. A proper implementation will take much more work than I could do at the moment, but I don't want to undo all the work I've already done on it. I can't do anything else in Retroquad until this is done, but there's no better way; it's too late to turn back now.
Ph'nglui mglw'nafh mankrip Hell's end wgah'nagl fhtagn.
==-=-=-=-=-=-=-=-=-=-==
/ /