Lightened Skyrim (by nicola89b and Firegtx)

[Mousetick]

3 hours ago, z929669 said:

I am not in favor of dropping mods just because they have LR warnings. The vast majority of LR warnings to not manifest --in any reasonably-detectable way-- into LR bugs in game.

That's my thinking as well. And even if some LR bugs do manifest in game, the worst that can happen is some z-fighting between LOD(s) and full model(s) in affected cells. It's not like they cause the game to crash or anything game-breaking. It may be in a cell that users may visit once or twice, or perhaps never, in a playthrough and therefore are unlikely to notice or be annoyed.

I'd suggest using a finer set of criteria than simply "trigger DynDOLOD LR warnings or not" to decide whether to keep a mod:

1. How many cells are affected by the LR warnings
2. How many LR are there in the affected cells* (i.e. how likely are LR bugs to noticeably manifest in those cells)
3. Are there visible defects with the LRs in the affected cells^
4. How frequently is an affected cell visited by an average player (e.g. is this near a town or in a remote mountainous area)
5. How "important" is the mod in the whole build

(*) Admittedly this is impossible to determine without dumping the RNAM records of the parent worldspace in the master plugins and cross-referencing them with all references defined in a specific cell. Can't be done by a human being so some kind of xEdit script would be needed to automate the process. Perhaps sheson can provide some advice on that.

(^) The LR bug affects all LRs in a given cell. The DynDOLOD warning doesn't necessarily indicate that the LOD will be noticeably bugged in game for that specific overwritten reference. Other LRs in the same cell could be however.

I reckon the proposed list of criteria is too complex and time-consuming to be of practical use. So perhaps going with #1, #4 and #5 in a first pass may be good enough, and then go with #3 (and ideally with #2 first) in a second pass to make a finer assessment if needed.

[TechAngel85]

LR bugs will cause produce z-fighting or pop-in. The root of the issue is the altered LR loaded reference no longer matches the "supermesh" for the cell. As such, the LOD will show whatever was or wasn't there in the vanilla supermesh. It's the data that is used to build the supermesh that needs to be updated. XEdit doesn't do this on its own, but sheson has a script that will do it.

[DoubleYou]

You guys are right. Not sure what happened. There are no Large Reference bugs with this mod. I checked all versions even, and none of them showed it, so I don't know what I was smoking, but it must have been pretty good stuff. Unless perhaps it was something to do with uncleaned DLCs? I don't care anymore as long as it's fixed.

[z929669]

2 hours ago, TechAngel85 said:

LR bugs will cause produce z-fighting or pop-in. The root of the issue is the altered LR loaded reference no longer matches the "supermesh" for the cell. As such, the LOD will show whatever was or wasn't there in the vanilla supermesh. It's the data that is used to build the supermesh that needs to be updated. XEdit doesn't do this on its own, but sheson has a script that will do it.

Yeah, but it is only a problem when it is apparent (which it rarely is). It's best practice to fix the plugin, but it probably doesn't matter for this mod, and I don't think this mod has any LR ref bugs or warnings ... at least none that I have seen.

But see Mouse's earlier posts. He has looked at this more closely.,

10 minutes ago, DoubleYou said:

You guys are right. Not sure what happened. There are no Large Reference bugs with this mod. I checked all versions even, and none of them showed it, so I don't know what I was smoking, but it must have been pretty good stuff. Unless perhaps it was something to do with uncleaned DLCs? I don't care anymore as long as it's fixed.

Yeah, that's possible, but I looked at a log I had from before I cleaned and didn't find any.

No worries! I've made worse mistakes

[z929669]

Voting to accept this one

[TechAngel85]

Same, vote yes.

[z929669]

Marked Accepted

[TechAngel85]

After evaluating this mod in xEdit, we need to decide if it's still desired.

• It's not just disabling hidden objects. It disables references that the author simply deemed unnecessary (leaves in Riften, clutter, wood piles, tickets, etc.)
• It has a LOT of overlap with USSEP, despite USSEP being a master. Many things that are fixed by USSEP, Lightened Skyrim opts to disable or "refix" anyway.
• It has several overlapping edits with it's other master, SLaWF. Some are "additional fixes" and others seem to remove things already fixed (just like with USSEP).
• Many edits seem to be made on a whim. Some references of the same object are moved or corrected while others are disabled...can't find the logic other than "author's choice".

 

With this new information, I think we accept it at face value or drop it, rather than patching.

[z929669]

Moved from testing. OP updated