sl-service-account
commented
7 years ago Whirly Fizzle commented at 2016-10-01T23:55:18Z, updated at 2016-10-01T23:56:16Z
BUG-9439 has never been fixed & I can confirm it still reproduces on the current default LL release viewer: Second Life 4.0.8.319463 (Second Life Release)
The reason that Linden probably couldn't reproduce it is because they had their RenderVolumeLODFactor setting down low. This bug will reproduce more easily at higher RenderVolumeLODFactor settings. As far as I'm aware, it's only the RenderVolumeLODFactor setting that affects the behaviour of this bug - it isn't OS or card dependent or dependent on any other graphics settings.
I get the feeling it may be some kind of "griefer" object protection - too many verts in small space cause the viewer to just stop rendering objects at the highest LOD in that space. If that is the case then I feel the threshold needs to be raised - this problem gets reported pretty frequently to Firestorm support. Though also to note, Firestorm has slightly higher RenderVolumeLODFactor default settings then the LL viewer so it's expected to be noticed by Firestorm users more frequently.
Steps to Reproduce
I had uploaded a number of mesh objects that were complex with a high number of vertices. Once a certain number of these mesh objects were rezzed parts of those meshes and other objects in my field of view disappeared. This also includes trying to rezz a simple prim - it rezzed invisible.
This happened previously for me at one particular sim and led me to wonder whether there was something wrong with that sim server. At that time I lodged a support ticket and a nice Linden came to help me troubleshoot - at that time the Linden could not replicate the issue and the issue did not occur on my alternative laptop. We deduced at the time that the issue was an unknown local machine graphics issue.
Fast forward to now. The issue still occurs at the problem sim. The issue suddenly starts occurring at my other building location on the mainland - on multiple computers and on different viewers (tried SL viewer and Firestorm). Common denominator: there are a number of complex mesh objects in the scene.
Actual Behavior
When there are a number of high complex meshes in the same area, parts of those meshes and other objects disappear (including simple prims) on variety of LOD factors including LOD factor 4.
Screen record of issue: https://youtu.be/hpG5zJ3C3QI
Expected Behavior
Things to not disappear on high LOD.
Other information
I tried reducing the vertices in the mesh objects and re-uploaded them. Now all of the same mesh objects in the same area do not cause the issue. Therefore I am assuming the issue is related to the complexity of the meshes in the area are too high for something to cope with. I have a suspicion that this is related to local machine performance as well - purely based upon the fact that in the past the issue was not reproducible by the Linden in the same area. However, as this bug was logged in 2015 here: https://jira.secondlife.com/browse/BUG-9439? I am also assuming it is replicatable by others.
Links
Related
Original Jira Fields
| Field | Value | | ------------- | ------------- | | Issue | BUG-40618 | | Summary | Recurrence of Bug-9439 - Objects disappear when highly complex meshes are present | | Type | Bug | | Priority | Unset | | Status | Closed | | Resolution | Duplicate | | Reporter | Julala Demina (julala.demina) | | Created at | 2016-10-01T09:17:20Z | | Updated at | 2016-10-04T23:56:58Z | ``` { 'Business Unit': ['Platform'], 'Date of First Response': '2016-10-01T18:55:18.269-0500', "Is there anything you'd like to add?": 'I tried reducing the vertices in the mesh objects and re-uploaded them. Now all of the same mesh objects in the same area do not cause the issue. Therefore I am assuming the issue is related to the complexity of the meshes in the area are too high for something to cope with. I have a suspicion that this is related to local machine performance as well - purely based upon the fact that in the past the issue was not reproducible by the Linden in the same area. However, as this bug was logged in 2015 here: https://jira.secondlife.com/browse/BUG-9439? I am also assuming it is replicatable by others.', 'ReOpened Count': 0.0, 'Severity': 'Unset', 'System': 'SL Viewer', 'Target Viewer Version': 'viewer-development', 'What just happened?': 'When there are a number of high complex meshes in the same area, parts of those meshes and other objects disappear (including simple prims) on variety of LOD factors including LOD factor 4.\r\n\r\nScreen record of issue: https://youtu.be/hpG5zJ3C3QI', 'What were you doing when it happened?': 'I had uploaded a number of mesh objects that were complex with a high number of vertices. Once a certain number of these mesh objects were rezzed parts of those meshes and other objects in my field of view disappeared. This also includes trying to rezz a simple prim - it rezzed invisible. \r\n\r\nThis happened previously for me at one particular sim and led me to wonder whether there was something wrong with that sim server. At that time I lodged a support ticket and a nice Linden came to help me troubleshoot - at that time the Linden could not replicate the issue and the issue did not occur on my alternative laptop. We deduced at the time that the issue was an unknown local machine graphics issue. \r\n\r\nFast forward to now. The issue still occurs at the problem sim. The issue suddenly starts occurring at my other building location on the mainland - on multiple computers and on different viewers (tried SL viewer and Firestorm). Common denominator: there are a number of complex mesh objects in the scene.\r\n\r\n\r\n\r\n\r\n\r\n', 'What were you expecting to happen instead?': 'Things to not disappear on high LOD.', 'Where': 'You are at 30.8, 40.4, 3,001.4 in Albulata located at sim10202.agni.lindenlab.com (216.82.49.68:13020)\r\nSLURL: http://maps.secondlife.com/secondlife/Albulata/31/40/3001\r\n(global coordinates 254,751.0, 262,184.0, 3,001.4)\r\nSecond Life Server 16.08.31.319299\r\nRetrieving...\r\n\r\nCPU: Intel(R) Core(TM) i3-3220 CPU @ 3.30GHz (3292.62 MHz)\r\nMemory: 8156 MB\r\nOS Version: Microsoft Windows 7 SP1 64-bit (Build 7601)\r\nGraphics Card Vendor: NVIDIA Corporation\r\nGraphics Card: GeForce GTX 650 Ti BOOST/PCIe/SSE2\r\n\r\nWindows Graphics Driver Version: 21.21.0013.7290\r\nOpenGL Version: 4.5.0 NVIDIA 372.90\r\n\r\nRestrainedLove API: (disabled)\r\nlibcurl Version: libcurl/7.47.0 OpenSSL/1.0.1i zlib/1.2.8\r\nJ2C Decoder Version: KDU v7.8\r\nAudio Driver Version: FMOD Ex 4.44.61\r\nLLCEFLib/CEF Version: 1.5.3.FS6-(CEF-WIN-3.2526.1366.g8617e7c-32) (Chrome 47.0.2526.80)\r\nVoice Server Version: Vivox 4.6.0017.22050\r\n\r\nSettings mode: Firestorm\r\nViewer Skin: Firestorm (Grey)\r\nFont Used: Deja Vu (96 dpi)\r\nFont Size Adjustment: 0 pt\r\nUI Scaling: 1\r\nDraw distance: 216 m\r\nBandwidth: 1500 kbit/s\r\nLOD factor: 4\r\nRender quality: High (5/7)\r\nAdvanced Lighting Model: No\r\nTexture memory: 1024 MB (1)\r\nVFS (cache) creation time (UTC): 2016-9-30T20:11:16 \r\nBuilt with MSVC version 1800\r\nPackets Lost: 0/31,811 (0.0%)', } ```