Questions from a small fgdata-next cleanup check
An Phúc Nguyễn Vạn <[email protected]>
| Newsgroups | gmane.games.flightgear.devel |
|---|---|
| Message-ID | <CAKX87K=NxsKtZnAHsMV0um7bv_ktb37bKPy-gr9Vvup3SZG8EQ@mail.gmail.com> |
Hi Folks, I was looking through a local checkout of fgdata-next and wanted to ask a few questions before I open any issues or MRs. Basically I had an idea in mind to just use tools to go around finding inefficient/legacy ancient stuff in fgdata that still persists. This is not a complaint or a “please delete old things” message. I am mostly trying to understand which things are intentional compatibility paths, and which ones might be useful cleanup/profiling candidates. My checkout is: next b694b4f HDR: Add debug display of hdr-forward buffer A few things stood out. First, Effects/ws30.eff still has a technique that references: Shaders/HDR/ws30_old.vert Shaders/HDR/ws30_old.frag So I assume this is not dead code. Is this an intentional fallback path for older hardware/drivers, or just an older path that has not been cleaned up yet? If it is intentionally kept, would a small comment or rename to make the fallback role clearer be useful? Second, I noticed that the runway textures are split between DDS and PNG material definitions. For example, Materials/dds/global.xml references: Runway/grass_rwy.dds Runway/dirt_rwy.dds Runway/pa_threshold.dds Runway/pc_threshold.dds while Materials/default/global.xml and some Materials/regions/*.xml files reference PNG versions of the same or similar textures. I am not assuming these are unused duplicates, since both paths are clearly referenced. Is this split still intentional, or is this an area where a clearer material/texture policy would be useful? Third, I grepped for settimer() and found a number of zero-delay or immediate-reschedule patterns, especially under Nasal/local_weather, but also in aircraft.nas, multiplayer.nas and jetways/jetways.nas. Some examples are weather interpolation, thermal/wave lift loops, weather dynamics/quadtree loops, cloud array creation, and some jetway/multiplayer update paths. Are these mostly known and harmless because they are feature-gated or short-lived, or are any of these areas already known to be worth profiling? Fourth, I found 47 tracked .xcf, .blend and .zip files. Some of the larger ones are: Aircraft/Instruments-3d/ar-6201/src/ar-6201.xcf AI/Aircraft/DH82/Paintkit/DH82-p-kit.xcf Aircraft/Instruments-3d/bxp6401/src/bxp6401.xcf AI/Aircraft/CCX/Paintkit/bomb-5000-pkit.xcf Aircraft/c172p/Paintkit/Paintkit.zip Are source art / paintkits / tool archives intentionally kept in fgdata for convenience, or would it be better to move some of those outside the runtime data tree? I am happy to split any of these into separate GitLab issues if that would be useful. I just wanted to ask first so I do not file noise or mislabel something intentional as a problem. Kind regards, Philips Nguyen _______________________________________________ Flightgear-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/flightgear-devel