Hard Light Productions Forums
Modding, Mission Design, and Coding => FS2 Open Coding - The Source Code Project (SCP) => Topic started by: General Battuta on January 31, 2014, 12:17:18 pm
-
Now, the next challenge is to update the mods since everything except FSO is out of date. But unfortunately once Blue Planet and possibly some other mods gets around to update theirs, the official FSO build won't run said mods. Is there any possibility of adding support for FSO executable dependencies? And let me be clear here, I think it is stupid for official mod releases to require unofficial FSO builds. BP should never have gone this route.
BP, the MediaVPs, JAD...sometimes the official builds just aren't viable.
-
BP, the MediaVPs, JAD...sometimes the official builds just aren't viable.
Perhaps so but that is something to handle with the SCP in cooperation. Going with wide variety of unofficial builds is just making it much harder to end-users to follow which build should be selected with whichever mod.
Edit: Edited my earlier post somewhat to add some more points.
-
I agree, it's a pretty ****ty position to end up in.
-
The BP builds theoretically won't remain separate from trunk; at some point, those lighting improvements will wind up in an official release. The reason BP went with a separate build at all is because the code freeze was in effect at the time; the reason those changes are still in a separate build is because some computers cannot run those builds at all.
-
The lighting 'improvements' are devastating performance-wise, though. I'm not sure what their status will be in the long run.
-
The lighting 'improvements' are devastating performance-wise, though.
Well, they perform a lot better than trying to get that many light sources without deferred rendering. :P
-
No, they don't, that's the problem.
-
No, they don't, that's the problem.
Bzzzt, wrong.
There are issues with the deferred renderer (and swifty is working on them), but saying that the old renderer would have rendered the same number of lights faster is flat-out wrong. We tried that approach once. It was not at all pretty.
-
AFAIK we've added the ability to render more lights, though. I'm not concerned with the ability to render an equivalent number of lights more rapidly, I'm concerned with the fact that net performance on an identical scene can in some cases drop.
e: because the engine is attempting to do more, even if it's doing so more efficiently
-
This is off-topic...