DynDOLOD 3.00 Alpha 182

[broncofan0211]

6 hours ago, sheson said:

6 hours ago, broncofan0211 said:

Hi, I am using DynDOLOD 3 and have the latest version, plus the latest Dyndolod Resources installed. I have a problem when running Dyndolod (Texgen works fine), where it will just close on me in the middle of the process. It closes at different times (ie, MarkarthWorld, SkuldafnWorld, FalmerValley, etc). It's inconsistent. After trying 4 or 5 times, it will eventually complete successfully. I've got my mods folder exempt from antivirus, and have even tried with antivirus disabled completely. Any ideas what is causing this? Thanks in advance!

Read the first post of the the DynDOLOD 3 alpha thread where to make post. I moved  the post.

Read the first post which log files, debug log and bugreport.txt to upload when reporting problems.

If there is no bugreport.txt it might mean the program is forcefully terminated by the OS. Check the Windows event log for entries.

Sorry about posting incorrectly. Yeah I don't get a bugreport log. But I did go to Windows Events. Details from the Application Error:

<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">

- <System>

  <Provider Name="Application Error" />

  <EventID Qualifiers="0">1000</EventID>

  <Version>0</Version>

  <Level>2</Level>

  <Task>100</Task>

  <Opcode>0</Opcode>

  <Keywords>0x80000000000000</Keywords>

  <TimeCreated SystemTime="2022-02-04T02:26:09.4438969Z" />

  <EventRecordID>103507</EventRecordID>

  <Correlation />

  <Execution ProcessID="0" ThreadID="0" />

  <Channel>Application</Channel>

  <Computer>PC</Computer>

  <Security />

  </System>

- <EventData>

  <Data>DynDOLODx64.exe</Data>

  <Data>3.0.0.0</Data>

  <Data>61e81ae6</Data>

  <Data>KERNELBASE.dll</Data>

  <Data>10.0.19041.1466</Data>

  <Data>e01c7650</Data>

  <Data>0eedfade</Data>

  <Data>0000000000034f69</Data>

  <Data>fa68</Data>

  <Data>01d8196c968499ea</Data>

  <Data>D:\TOOLS\DynDOLOD\DynDOLODx64.exe</Data>

  <Data>C:\WINDOWS\System32\KERNELBASE.dll</Data>

  <Data>1f743b0b-fd14-422a-ac99-46bd397b276d</Data>

  <Data />

  <Data />

  </EventData>

  </Event>

 

And also, leading up to the Error there are multiple warnings that say "The backing-file for the real-time session "Admin_PS_Provider" has reached its maximum size. As a result, new events will not be logged to this session until space becomes available. This error is often caused by starting a trace session in real-time mode without having any real-time consumers"

With these details:

- <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">

- <System>

  <Provider Name="Microsoft-Windows-Kernel-EventTracing" Guid="{b675ec37-bdb6-4648-bc92-f3fdc74d3ca2}" />

  <EventID>1</EventID>

  <Version>0</Version>

  <Level>3</Level>

  <Task>1</Task>

  <Opcode>10</Opcode>

  <Keywords>0x8000000000000010</Keywords>

  <TimeCreated SystemTime="2022-02-04T02:24:58.3346641Z" />

  <EventRecordID>17203</EventRecordID>

  <Correlation />

  <Execution ProcessID="4" ThreadID="13592" />

  <Channel>Microsoft-Windows-Kernel-EventTracing/Admin</Channel>

  <Computer>PC</Computer>

  <Security UserID="" />

  </System>

- <EventData>

  <Data Name="SessionName">Admin_PS_Provider</Data>

  <Data Name="ErrorCode">3221225864</Data>

  <Data Name="LoggingMode">8388864</Data>

  </EventData>

  </Event>

 

Not sure if any of this is useful or not. Thanks again.

Edited February 4, 2022 by broncofan0211

[sheson]

3 hours ago, broncofan0211 said:

Sorry about posting incorrectly. Yeah I don't get a bugreport log. But I did go to Windows Events. Details from the Application Error:

<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">

- <System>

  <Provider Name="Application Error" />

  <EventID Qualifiers="0">1000</EventID>

  <Version>0</Version>

  <Level>2</Level>

  <Task>100</Task>

  <Opcode>0</Opcode>

  <Keywords>0x80000000000000</Keywords>

  <TimeCreated SystemTime="2022-02-04T02:26:09.4438969Z" />

  <EventRecordID>103507</EventRecordID>

  <Correlation />

  <Execution ProcessID="0" ThreadID="0" />

  <Channel>Application</Channel>

  <Computer>PC</Computer>

  <Security />

  </System>

- <EventData>

  <Data>DynDOLODx64.exe</Data>

  <Data>3.0.0.0</Data>

  <Data>61e81ae6</Data>

  <Data>KERNELBASE.dll</Data>

  <Data>10.0.19041.1466</Data>

  <Data>e01c7650</Data>

  <Data>0eedfade</Data>

  <Data>0000000000034f69</Data>

  <Data>fa68</Data>

  <Data>01d8196c968499ea</Data>

  <Data>D:\TOOLS\DynDOLOD\DynDOLODx64.exe</Data>

  <Data>C:\WINDOWS\System32\KERNELBASE.dll</Data>

  <Data>1f743b0b-fd14-422a-ac99-46bd397b276d</Data>

  <Data />

  <Data />

  </EventData>

  </Event>

 

And also, leading up to the Error there are multiple warnings that say "The backing-file for the real-time session "Admin_PS_Provider" has reached its maximum size. As a result, new events will not be logged to this session until space becomes available. This error is often caused by starting a trace session in real-time mode without having any real-time consumers"

With these details:

- <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">

- <System>

  <Provider Name="Microsoft-Windows-Kernel-EventTracing" Guid="{b675ec37-bdb6-4648-bc92-f3fdc74d3ca2}" />

  <EventID>1</EventID>

  <Version>0</Version>

  <Level>3</Level>

  <Task>1</Task>

  <Opcode>10</Opcode>

  <Keywords>0x8000000000000010</Keywords>

  <TimeCreated SystemTime="2022-02-04T02:24:58.3346641Z" />

  <EventRecordID>17203</EventRecordID>

  <Correlation />

  <Execution ProcessID="4" ThreadID="13592" />

  <Channel>Microsoft-Windows-Kernel-EventTracing/Admin</Channel>

  <Computer>PC</Computer>

  <Security UserID="" />

  </System>

- <EventData>

  <Data Name="SessionName">Admin_PS_Provider</Data>

  <Data Name="ErrorCode">3221225864</Data>

  <Data Name="LoggingMode">8388864</Data>

  </EventData>

  </Event>

 

Not sure if any of this is useful or not. Thanks again.

Check if the event id 1000 is preceded by other events with the similar timestamp. For example like event id 1005 like shown in the screenshots of from this post https://stepmodifications.org/forum/topic/15606-dyndolod-300-alpha-65/?do=findComment&comment=256156

Make sure Antivir or other 3rd party tools are not interfering.

Check the used filesystems/disks for errors.

The second event seems unrelated.

[sheson]

6 hours ago, Mousetick said:

Thanks for the clarification. So in the case of a Deleted-flagged large reference in an ESP, shouldn't DynDOLOD issue 2 warnings for the same record: one "Deleted reference" warning on , and one "Overwritten large reference" warning?

From https://dyndolod.info/Help/Large-References (emphasis added):

Apparently that's not the case. I just tested this scenario: I created an ESP in xEdit and flagged a large reference of Skyrim.esm as Deleted in the ESP. There is no other plugin overwriting this large reference.

With Alpha-65 I only got a "Deleted reference" warning on the Deleted References page. And that's it. Nothing on the Large Reference Bugs page.

Typically if an error is logged for an item, no further processing is done for it since it is being discarded, unavailable etc.

[Mousetick]

A solution to the large reference bugs could conceivably be to remove the affected references individually from the RNAM list in the ESM, right? It seems it would be a more efficient and simpler way to address the issue, rather than splitting mods into ESM for large references + ESP for the remainder, or than burying all large references in a cell + re-creating new identical but non-large ones.

I tried this approach by removing one entry in the RNAM of Tamriel in a small ESM (RWT2.esm) with xEdit and that got rid of the corresponding DynDOLOD warning.

Unfortunately xEdit is hardcoded to skip the RNAMs in Skyrim.esm, where most of the large references overwritten by mods are defined, so they're not viewable/editable. I'm going to try a workaround but I fear that xEdit's RNAM editor might explode due the very large RNAM of Tamriel. We'll see...

 

---

Unrelated: what to make of the following warnings? "DynDOLOD.esm missing an overriden record" is not listed in https://dyndolod.info/Messages.

Spoiler

[00:00] Gathering LOD assets
[00:15] Processing patches
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24) is missing an overriden record for SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24) is missing an overriden record for SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24) is missing an overriden record for SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23) is missing an overriden record for [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23) is missing an overriden record for [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23) is missing an overriden record for [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23) is missing an overriden record for [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23)
[00:24] Processing 24125 base records for 2906 object LOD and 231 dynamic LOD models

 

Thanks.

[sheson]

2 hours ago, Mousetick said:

A solution to the large reference bugs could conceivably be to remove the affected references individually from the RNAM list in the ESM, right? It seems it would be a more efficient and simpler way to address the issue, rather than splitting mods into ESM for large references + ESP for the remainder, or than burying all large references in a cell + re-creating new identical but non-large ones.

I tried this approach by removing one entry in the RNAM of Tamriel in a small ESM (RWT2.esm) with xEdit and that got rid of the corresponding DynDOLOD warning.

Unfortunately xEdit is hardcoded to skip the RNAMs in Skyrim.esm, where most of the large references overwritten by mods are defined, so they're not viewable/editable. I'm going to try a workaround but I fear that xEdit's RNAM editor might explode due the very large RNAM of Tamriel. We'll see...

 

---

Unrelated: what to make of the following warnings? "DynDOLOD.esm missing an overriden record" is not listed in https://dyndolod.info/Messages.

  Reveal hidden contents

[00:00] Gathering LOD assets
[00:15] Processing patches
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24) is missing an overriden record for SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24) is missing an overriden record for SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24) is missing an overriden record for SolitudeArch01 [CELL:0006E76E] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,24)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23) is missing an overriden record for [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23) is missing an overriden record for [CELL:00037EEB] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -17,23)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23) is missing an overriden record for [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23)
[00:17] <Warning: File [4D] DynDOLOD.esm with Group GRUP Cell Children of [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23) is missing an overriden record for [CELL:00037EEC] (in SolitudeWorld "Solitude" [WRLD:00037EDF] at -16,23)
[00:24] Processing 24125 base records for 2906 object LOD and 231 dynamic LOD models

 

Thanks.

Yes, updating the RNAM data in the vanilla plugins to make reference not large will work. Burying all large references in a cell  (at least the once that have visual flicker) and add new non large reference works, too. That is how DynDOLOD creates the workaround for one vanilla issue for the College of Winterhold bridge.

xLODGen terrain LOD beta (used in edit mode) shows RNAM data on Skyrim.esm. Or rename Skyrim.esm. Its a lot of data.

I purposefully refrained from considering patching RNAM data in plugins to address the bugs. A better solution in this direction would be a DLL plugin that skips loading the data. Aers and I looked at it for a bit a couple years back (which let to reference cap explanations) but we couldn't easily make it work at the time. Also I do not do any serious reverse engineering while there are still updates to the game.

"missing an overriden record" is a message from xEdit that should not happen with a DynDOLOD plugin in any normal operation.
Read the first post which entire log files to upload when making posts.
If this happens while trying to update existing DynDOLOD plugins, generate from scratch instead.

[Mousetick]

2 hours ago, sheson said:

A better solution in this direction would be a DLL plugin that skips loading the data. Aers and I looked at it for a bit a couple years back (which let to reference cap explanations) but we couldn't easily make it work at the time.

Yes ideally DynDOLOD would generate a "blacklist" of overwritten large references. Then a DLL like SSE Engine Fixes or similar engine-level runtime patcher could read that blacklist and skip the RNAM entries while loading the ESM WRLD records. That'd be quite neat... Are you familiar with CommonLibSSE? I'm not, but it looks like it reverse-engineers a lot of data and APIs. Impressive stuff. For example, the in-memory data-structure of a Worldspace, including the Large References: https://github.com/Ryan-rsm-McKenzie/CommonLibSSE/blob/master/include/RE/T/TESWorldSpace.h.

2 hours ago, sheson said:

"missing an overriden record" is a message from xEdit that should not happen with a DynDOLOD plugin in any normal operation.
Read the first post which entire log files to upload when making posts.
If this happens while trying to update existing DynDOLOD plugins, generate from scratch instead.

I thought this was related to updating the DynDOLOD plugins too, but it's actually not, as I've re-run it from scratch several times and the warnings are still issued. Using Alpha-65.

Here you go for the logs: https://drive.google.com/drive/folders/1fQvv6PVQ5uC3w_lQuFlT0K29O6BkgAeO

Please change your logger so that it overwrites logs instead of appending, or it appends a session timestamp to log file names. It's really annoying to have to scroll past past sessions to find the last one in the logs - both for your and your users' sake.

There is only one session contained in the logs linked above.

Thanks.

[sheson]

47 minutes ago, Mousetick said:

Yes ideally DynDOLOD would generate a "blacklist" of overwritten large references. Then a DLL like SSE Engine Fixes or similar engine-level runtime patcher could read that blacklist and skip the RNAM entries while loading the ESM WRLD records. That'd be quite neat... Are you familiar with CommonLibSSE? I'm not, but it looks like it reverse-engineers a lot of data and APIs. Impressive stuff. For example, the in-memory data-structure of a Worldspace, including the Large References: https://github.com/Ryan-rsm-McKenzie/CommonLibSSE/blob/master/include/RE/T/TESWorldSpace.h.

I thought this was related to updating the DynDOLOD plugins too, but it's actually not, as I've re-run it from scratch several times and the warnings are still issued. Using Alpha-65.

Here you go for the logs: https://drive.google.com/drive/folders/1fQvv6PVQ5uC3w_lQuFlT0K29O6BkgAeO

Please change your logger so that it overwrites logs instead of appending, or it appends a session timestamp to log file names. It's really annoying to have to scroll past past sessions to find the last one in the logs - both for your and your users' sake.

There is only one session contained in the logs linked above.

Thanks.

Put in a feature request with xEdit for an command line option to replace logfiles. Otherwise simply delete the log folder before doing such a run or truncate the log file in notepad to last meaningful generation as explained on the first post. Simply search for "starting session" to jump to a start of a session. Typically it is easier for people to simply upload the logfile without having to worry about the session with the messages of interest being already replaced.

If you load the generated DynDOLOD.esm into xEdit afterwards, does xEdit still show the message or is it corrected once it has been saved?

[Mousetick]

6 hours ago, sheson said:

Put in a feature request with xEdit for an command line option to replace logfiles.

Oh, it's xEdit's logger. I hadn't thought about that. But yes of course, makes sense. Well, never mind then 

7 hours ago, sheson said:

If you load the generated DynDOLOD.esm into xEdit afterwards, does xEdit still show the message or is it corrected once it has been saved?

No, once DynDOLOD.esm is saved, everything is fine. Both in xEdit and in game. No "missing an overriden record" in xEdit when loading DynDOLOD.esm.

I'm not too worried about this warning since I can't see any negative effect in-game, but I mentioned it in case you'd want to know about it. Thanks for looking into it anyway. Let me know if you need more info or you'd like me to make specific tests.

And thanks for all your answers, much appreciated.

[sheson]

8 minutes ago, Mousetick said:

Oh, it's xEdit's logger. I hadn't thought about that. But yes of course, makes sense. Well, never mind then 

No, once DynDOLOD.esm is saved, everything is fine. Both in xEdit and in game. No "missing an overriden record" in xEdit when loading DynDOLOD.esm.

I'm not too worried about this warning since I can't see any negative effect in-game, but I mentioned it in case you'd want to know about it. Thanks for looking into it anyway. Let me know if you need more info or you'd like me to make specific tests.

And thanks for all your answers, much appreciated.

Good to know.

There should be no more missing overriden record and no more warning with the next alpha version.

[zhuabaobao]

Ask, why I am far away, there are seams in the water, and the distance is close to normal

 

[Sayoregg]

All the guides that are available online aren't useful for the new version as the options differ, and I couldn't find anything about tree LOD in the manual. I just want to know what settings I have to set to have all the trees in the worldspace visible at once and have them appear on the map. As far as I understand it means generating tree LOD as static LOD, but it's not clear what i actually have to change in the settings to achieve that.

[sheson]

1 hour ago, zhuabaobao said:

Ask, why I am far away, there are seams in the water, and the distance is close to normal

 

As explained again and again and at https://dyndolod.info/How-LOD-Works and https://dyndolod.info/Help/Terrain-LOD, DynDOLOD generates tree and object LOD and does not do generate or change terrain or water LOD.

Water LOD as part of terrain LOD generation / BTR meshes only defines where water LOD is.

How water LOD looks depends entirely on the water LOD record defined on the worldspace record. LOD generation does not change this record and vice versa the record does not affect LOD generation.

Because of this single record for the entire worldspace, all water LOD looks the same, unlike water in the active cells, which can each define their specific water record.

1 hour ago, Sayoregg said:

All the guides that are available online aren't useful for the new version as the options differ, and I couldn't find anything about tree LOD in the manual. I just want to know what settings I have to set to have all the trees in the worldspace visible at once and have them appear on the map. As far as I understand it means generating tree LOD as static LOD, but it's not clear what i actually have to change in the settings to achieve that.

Read the first post of the DynDOLOD 3 Alpha thread where to make posts. I moved your post to the thread.

https://dyndolod.info/Help/Tree-LOD explains tree LOD. It says this:

While this is commonly referred to as tree LOD, it might also be referred to as standard tree LOD or traditional tree LOD when trying to distinguish between similar tree LOD with tree LOD billboards done in object LOD, e.g. ultra tree LOD.

Read the documentation https://dyndolod.info/Help/Ultra-Tree-LOD. It happens to explain how to generate ultra tree LOD, how to have ultra tree LOD in all object LOD levels and on the map, which happens automatically because the map uses object LOD level 16.

[matashina]

Hi, it looks like the .NET Script Framework, which is a requirement for DynDOLOD grass LOD via No Grass in Objects, will not be receiving an update for AE. Is the right thing to do now as an end user who isn't interested in downgrading to continue to use DynDOLOD without grass?

Thank you.

[wasilee]

hey can i get some help please on the DynDOLOD advanced setting the grass lod select box is grayed out and its saying its found 0 grass cache files when i hover over the box.

[DoubleYou]

1 hour ago, wasilee said:

hey can i get some help please on the DynDOLOD advanced setting the grass lod select box is grayed out and its saying its found 0 grass cache files when i hover over the box.

You need to generate a grass cache with NGIO first. See the second post. You can also look at this Grass LOD Guide. Be aware that if you are on AE, it will be very difficult to generate the necessary grass cache.