Cool VL Viewer forum

View unanswered posts | View active topics It is currently 2025-02-12 05:24:02



Reply to topic  [ 14 posts ]  Go to page 1, 2  Next
Good setting for PBR? 
Author Message

Joined: 2016-06-19 21:33:37
Posts: 384
Location: Columbus area, OH, USA
Reply with quote
Looking for some help from the forums regarding PBR/shadow/lighting settings. I have been using version 1.32.2.12 and when I create a cube with the plywood texture and no materials but tinted <20,20,20> (almost black), I see reddish/greenish 'wood grain' pattern:
Image
The same cube in 1.32.2.11 looks like:
Image
Experiments are carried out in the Morris Sandbox https://maps.secondlife.com/secondlife/Morris/59/83/36 with midday sky. The effect seems more pronounced on anything 'dark', and occurs at every sun/moon position or lighting style and the color 'banding' shows in other colors and even affects the sky. I have tried any number of configuration setting via Preferences->Graphics, the Advanced->Render menu and even direct manipulation of some debug settings with different sky presets or exposures. Nothing has allowed me to get non-'burnt' looking darkly textured items. I don't think it is my video card/driver but maybe? I haven't configured anything esoteric other than kernel opts: nvidia-drm.modeset=1 nvidia-drm.fbdev=1 fbcon=nodefer.

Any suggestions appreciated.

Code:
Cool VL Viewer v1.32.2.12, Sep  7 2024 13:40:32
Release notes

You are at 255291.4, 256338.5, 36.2  in Morris located at
simhost-0cd3b569a6915351c.agni.secondlife.io (54.188.190.196:13007)
Alias: ec2-54-188-190-196.us-west-2.compute.amazonaws.com
Second Life RC LeTigre 2024-07-22.10048683488
Release notes

CPU: AMD Ryzen 7 3700X 8-Core Processor (4399 MHz)
Memory: 64215MB
OS version: Linux-x86_64 v6.10.7-200.fc40.x86_64
Memory manager: jemalloc v5.3.1-20240906
Graphics: NVIDIA GEFORCE GTX 1080 TI/PCIE/SSE2
OpenGL version: 4.6.0 NVIDIA 560.35.03
Detected VRAM: 11264MB
J2C decoder: OpenJPEG: 1.4.0.635f
Audio driver: OpenAL v1.1 ALSOFT 1.23.1 (OpenAL Soft: OpenAL Soft)
Networking backend: libcurl 7.47.0/OpenSSL 1.0.2u/zlib 1.3.0.zlib-ng
Browser plugin: Dullahan 1.13.2/CEF 128.4.5/Chromium 128.0.6613.18
Packets lost: 0/13648 (0.0%)

Built with: GCC v8.3.0
Compiler-generated maths: SSE2.

Compile flags used for this build:
-O3 -fno-delete-null-pointer-checks -fno-ipa-cp-clone -fno-align-labels -fno-align-loops -fsched-pressure -frename-registers -fweb -fira-hoist-pressure -DNDEBUG -std=c++17 -fno-stack-protector -U_FORTIFY_SOURCE -fno-threadsafe-statics -fPIC -pipe -g -gdwarf-4 -fno-var-tracking-assignments -fexceptions -fno-strict-aliasing -fvisibility=hidden -fsigned-char -m64 -mfpmath=sse -fno-math-errno -fno-trapping-math -pthread  -DLL_LINUX=1 -DLL_JEMALLOC=1 -DLL_FMOD=1 -DLL_OPENAL=1


Attachments:
CoolVLViewer.zip [28.65 KiB]
Downloaded 47 times
2024-09-08 17:06:42
Profile

Joined: 2009-03-17 18:42:51
Posts: 5912
Reply with quote
Disable "Fix reflection blue hue".

There is sadly no "good" and even less "one fits all" settings combination with PBR.

Thankfully, LL is apparently now willing to "fix" their mess, so the future might "shine brighter"...


2024-09-08 18:26:30
Profile WWW

Joined: 2016-06-19 21:33:37
Posts: 384
Location: Columbus area, OH, USA
Reply with quote
Henri Beauchamp wrote:
Disable "Fix reflection blue hue".

There is sadly no "good" and even less "one fits all" settings combination with PBR.

Thankfully, LL is apparently now willing to "fix" their mess, so the future might "shine brighter"...

I have been running with that setting off since noticing the 'glow' at Hippo Hollow. I have tried various combos of it, the new tone mapping function, SSAO on/off, shadows on/off, exposure settings, different 'skies'. There isn't any change other than when turning PBR off completely. I am hopeful LL will get their PBR act together but being in SL as long as I have, I know it takes time.


2024-09-08 18:50:08
Profile

Joined: 2011-09-27 11:18:31
Posts: 220
Reply with quote
I face this problem too with .12 (not with .11).

Cinema (Cool VL Viewer):
Image

Cinema (LL viewer 7.1.10.10708851543 under Wine)
Image

http://maps.secondlife.com/secondlife/R ... 1/121/3001

Code:
Cool VL Viewer v1.32.2.12, Sep  8 2024 13:54:07
RestrainedLove viewer v2.09.29.29
Release notes

CPU: AMD Ryzen 5 4500 6-Core Processor (4042 MHz)
Memory: 31882MB
OS version: Linux-x86_64 v6.6.47-gentoo-dist
Memory manager: jemalloc v5.3.1-20240906
Graphics: AMD RADEON RX 580 SERIES (RADEONSI, POLARIS10, LLVM 18.1.8, DRM 3.54, 6.6.47-GENTOO-DIST)
OpenGL version: 4.6 (Core Profile) Mesa 24.1.6
Detected VRAM: 6941MB
J2C decoder: OpenJPEG: 1.4.0.635f
Audio driver: FMOD Studio v2.03.02 (PulseAudio)
Networking backend: libcurl 7.47.0/OpenSSL 1.0.2u/zlib 1.3.0.zlib-ng
Browser plugin: Dullahan 1.13.2/CEF 128.4.5/Chromium 128.0.6613.18

Built with: GCC v13.3.1
Compiler-generated maths: SSE2.

Compile flags used for this build:
-O3 -fno-delete-null-pointer-checks -fno-ipa-cp-clone -fno-align-labels -fno-align-loops -fsched-pressure -frename-registers -fweb -fira-hoist-pressure -DNDEBUG -std=c++17 -fno-stack-protector -U_FORTIFY_SOURCE -fno-threadsafe-statics -fPIC -pipe -g -gdwarf-4 -fno-var-tracking-assignments -fexceptions -fno-strict-aliasing -fvisibility=hidden -fsigned-char -m64 -mfpmath=sse -fno-math-errno -fno-trapping-math -pthread  -DLL_LINUX=1 -DLL_JEMALLOC=1 -DLL_FMOD=1 -DLL_OPENAL=1


I played with all the settings too (including the blue hue), but nothing helps. Some PBR aware environment make it better, but don't fix it.


2024-09-11 12:24:44
Profile

Joined: 2009-03-17 18:42:51
Posts: 5912
Reply with quote
AGAIN: the tone mapping CHANGED.

This is NORMAL that things look different.

If you want things to look the same as before, use the ACES tone mapper.


2024-09-11 13:45:01
Profile WWW

Joined: 2016-06-19 21:33:37
Posts: 384
Location: Columbus area, OH, USA
Reply with quote
Henri Beauchamp wrote:
AGAIN: the tone mapping CHANGED.

This is NORMAL that things look different.

If you want things to look the same as before, use the ACES tone mapper.
For me, I found that the different tone mapping setting doesn't seem to affect anything unless I also have 'Adjust legacy to HDR' set. When that is set, they seem to do nothing to affect the 'banding' of darker textures but just change saturation or perhaps contrast. Maybe another setting I'm using is conflicting though.


2024-09-11 14:17:44
Profile

Joined: 2011-09-27 11:18:31
Posts: 220
Reply with quote
ZaneZimer wrote:
Henri Beauchamp wrote:
AGAIN: the tone mapping CHANGED.

This is NORMAL that things look different.

If you want things to look the same as before, use the ACES tone mapper.
For me, I found that the different tone mapping setting doesn't seem to affect anything unless I also have 'Adjust legacy to HDR' set. When that is set, they seem to do nothing to affect the 'banding' of darker textures but just change saturation or perhaps contrast. Maybe another setting I'm using is conflicting though.


I can confirm what Zane says: https://i.imgur.com/EzvH1nR.png Changing the tone mapping or the SSAO or blue hue doesn't do anything at all.

What I'm pointing out is that the LL viewer shows things as they should be. So, there is a discrepancy, we are missing something.


2024-09-11 15:21:25
Profile

Joined: 2009-03-17 18:42:51
Posts: 5912
Reply with quote
g0rd0ngrfr33mailgr wrote:
What I'm pointing out is that the LL viewer shows things as they should be. So, there is a discrepancy, we are missing something.
I'm certainly missing what is "wrong" in what you are seeing, beside a different lighting (which is normal, since my viewer does not treat lighting like LL's)... :shock:


2024-09-11 17:49:02
Profile WWW

Joined: 2016-06-19 21:33:37
Posts: 384
Location: Columbus area, OH, USA
Reply with quote
Henri Beauchamp wrote:
g0rd0ngrfr33mailgr wrote:
What I'm pointing out is that the LL viewer shows things as they should be. So, there is a discrepancy, we are missing something.
I'm certainly missing what is "wrong" in what you are seeing, beside a different lighting (which is normal, since my viewer does not treat lighting like LL's)... :shock:
At least to me, especially in the images that g0rd0n posted (notably the TV texture), the gradients aren't as smooth, giving the appearance that the color depth is wrong like 16bit color vs 24bit color. It creates distinct 'banding' in the textures compared to previous versions of the viewer when using PBR.


2024-09-11 20:43:09
Profile

Joined: 2009-03-17 18:42:51
Posts: 5912
Reply with quote
ZaneZimer wrote:
At least to me, especially in the images that g0rd0n posted (notably the TV texture), the gradients aren't as smooth, giving the appearance that the color depth is wrong like 16bit color vs 24bit color. It creates distinct 'banding' in the textures compared to previous versions of the viewer when using PBR.
Ah... That ?... Fixed already for next release (I was therefore not seeing it myself any more): this is part of the latest changes that went into LL's Atlasaurus/develop viewer, and that I just finished backporting last Sunday... There is also a better SMAA shader, and more goodies.

If you compile the viewer yourself, the relevant line is in indra/newview/llpipeline.cpp, line 1302: change GL_RGBA for GL_RGBA16F to read if (!mPostMap.allocate(res_x, res_y, GL_RGBA16F)). The "banding" (that also existed in LL's viewer before Atlasaurus) was due to the fact the screen buffer and the postmap buffer did not share the same data format (floats for the first, integers for the second), while they are copied in each others several times in the renderFinalizePBR() step...


2024-09-11 21:56:00
Profile WWW
Display posts from previous:  Sort by  
Reply to topic   [ 14 posts ]  Go to page 1, 2  Next

Who is online

Users browsing this forum: No registered users and 52 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group
Designed by ST Software.