Chat
Discord
01-01-2014, 05:36 AM
Hello tyt2y3, thanks for your concern and revealing the codes here. You see I'm no clue of such programming, and only derive the whole theory by experimenting and guessing. I see you are talking about converting rect to RGB; that's part of I did here, as we should first understand what rect means. But I also derived how to change back RGB to rect. To me this is a bit more useful, and it is actually the ultimate goal for I first went into this topic. Of course, you know the backwards conversion is just derived from rect to RGB, they are similar, but still there is a need for someone to evaluate it before everyone since not everybody could done it by themselves. Or else, much shorter posts could suffice.
Anyways somehow I was spending time to try to write out some ideas I've thought of. You're right it is indeed too long, even for me sometimes it is a bit difficult to chase one point that I had written there... One would be the binary rect as you're saying here. For the most of the article I divide RGB to 5:6:5, in respect to the color they are responsible for. But I did discussed the 5:5:5 plus 1 bit interpretation, somewhere gone through 70% of Part 6.4. About "LF2 keeps a color table internally," I guess you mean the switch in your code, I acknowledge it also and mentioned in Part 8, and it affects the rect to RGB conversion as you listed, also the available actual RGB in rect representation. What I'm aware of is that, not sure if rect: 4706, 16835, 21096 and 37770 are the only 4 values with RGB differ from the calculated one (the others are the same). Indeed one (set) of general formulas wouldn't work for the whole conversion process, they are only a part of it, together with some conditional checks.
Since 10 years this is the first time I discuss LF2 with others regarding something other than (amateur) gameplay. Thanks to you all, this place is wonderful.
Anyways somehow I was spending time to try to write out some ideas I've thought of. You're right it is indeed too long, even for me sometimes it is a bit difficult to chase one point that I had written there... One would be the binary rect as you're saying here. For the most of the article I divide RGB to 5:6:5, in respect to the color they are responsible for. But I did discussed the 5:5:5 plus 1 bit interpretation, somewhere gone through 70% of Part 6.4. About "LF2 keeps a color table internally," I guess you mean the switch in your code, I acknowledge it also and mentioned in Part 8, and it affects the rect to RGB conversion as you listed, also the available actual RGB in rect representation. What I'm aware of is that, not sure if rect: 4706, 16835, 21096 and 37770 are the only 4 values with RGB differ from the calculated one (the others are the same). Indeed one (set) of general formulas wouldn't work for the whole conversion process, they are only a part of it, together with some conditional checks.
Since 10 years this is the first time I discuss LF2 with others regarding something other than (amateur) gameplay. Thanks to you all, this place is wonderful.