New shadow leaks in 2.12 dev builds

[datiswous]

I tested. This is a shadow leak:

(it's faily minor and hard to notice though)

Edited January 25, 2024 by datiswous

[datiswous]

--

 

Edited January 26, 2024 by datiswous
wrong place

[nbohr1more]

New report from the 2.12 beta thread:

Added to the wiki

[Daft Mugi]

Thief's Den 3: Heart of Lone Salvation

1736.48 3691.68 -137.75 -5.4 101.1 0.0

Light leak:

No light leak:

[nbohr1more]

Added to the wiki

[stgatilov]

This case in Poets and Peasants is something new:

In this case we have on same line:

• light
• brush wall
• fully closed room
• closed door with visportal in a brush wall
• player

Since only backfaces cast shadows, only the door and the first brush wall can cast shadows.
The brush wall does not cast shadows because the entirety of its area is unreachable by light.
So the door should.

It turns out that all doors that are connected to visportal should automatically get this "forceShadowBehindOpaque" flag. When the door is closed, it closes the visportal and thus becomes part of the wall, just like a brush would be in its place.

 

 

[thebigh]

Discovered another in The Gatehouse. Torchlight is leaking through the stairs

 

[nbohr1more]

Thank you!

Added to the wiki.

[stgatilov]

On 2/18/2024 at 11:26 AM, thebigh said:

Discovered another in The Gatehouse. Torchlight is leaking through the stairs

This one has nothing to do with the change.
It happens in 2.11 too.

I have no idea what is wrong here, and why one patch casts shadow while the other similar one does not.
There is even auxilliary surface with "shadow" material inside the stairs, which should also cast shadow but it does not.

[thebigh]

That is weird. Something funky with the torch itself, maybe?

[stgatilov]

11 hours ago, thebigh said:

That is weird. Something funky with the torch itself, maybe?

It casts shadows, and it is a proper point light.
Not sure what else can go wrong with a light.
 

[HMart]

Could it be anything to do with the shape of the geometry? Is a curved staircase looking at the images. 

Or the thickness? Thou by that image those stairs are anything but thin.  Is just that I have seen thin walls leak light, while ticker ones do not and I have seen this, in more than one engine even in lightmapped ones.

Also is that brush geometry or a imported triangle mesh? You will correct me if wrong but afaik the engine treats triangle mesh's, a tad differently from brush geometry, like it doesn't automatically inline the triangle mesh geometry, into the overall baked brush geometry (unless your force it), so it leaves a invisible "gap" there where the mesh connects with the brush wall? 

Could be floating point accuracy problems, like you recently talked about?

[thebigh]

Another one, in The Outpost, started happening between dev16809-10394 and dev16814-10408

 

[nbohr1more]

Added to the wiki. Thank you!

[stgatilov]

Another leak in Written in Stone near start (230 -1820 65):

This case is more interesting because it is not just light leaking vs not leaking through a wall.
Here we see view-dependent scissoring rectangle which limits the light effect.

The light source is in the farthest area behind a wall (on the middle of the two far visportals).
There is no way to cast a ray from the light source through visportals that would reach the viewer area: if you hit one of the far portals, then you surely don't hit the near portals. So the frontend decides that the light is not active in the viewer's area. Then the light scissor is bounding rectangle around all the visportals.

Normally, the wall in between should occlude the light, but apparently it was set to noshadows.
This wall seals the areas, so this is the same kind of bad input as usual. But the resulting behavior is indeed interesting.

So I wonder how should we fix it @Amadeus @Dragofer

[Amadeus]

I do remember this being a problematic area for a few reasons. I'll take a look and try and fix it; hopefully some time this week

[Amadeus]

So, I just retextured the caulk brushes to shadowcaulk and this seems to be working fine. Suitable fix???

However, I did find another light leak right next to this one:

[stgatilov]

2 hours ago, Amadeus said:

So, I just retextured the caulk brushes to shadowcaulk and this seems to be working fine. Suitable fix???

I think yes.

The root of the issue is always an opaque wall (in terms of areas/visportals) which does not cast shadows (in terms of lighting). Shadowcaulk is both opaque and casts shadows, so there is no contradiction.

[Amadeus]

okay, I'm gonna go through and fix these lightleaks with shadowcaulk then. There are a handful of other lightleaks in WIS, do you want me to log each one and report it on the wiki page, or can I just go ahead and fix them as I see them?

[nbohr1more]

56 minutes ago, Amadeus said:

okay, I'm gonna go through and fix these lightleaks with shadowcaulk then. There are a handful of other lightleaks in WIS, do you want me to log each one and report it on the wiki page, or can I just go ahead and fix them as I see them?

Go ahead and fix them if you want. The purpose of the wiki was to attempt to resolve as many leaks as possible prior to 2.12. If a map contained too many leaks, there is a map-wide workaround that reverts to 2.11 mode at the cost of reverting to 2.11 performance. It's up to you if you want to clean-up the leaks and retain the optimizations or just use the easy-button and set the mission to 2.11 mode.