DynDOLOD 3.00 Alpha 173

[mooit]

1 hour ago, sheson said:

Upload ..\skse\plugins\DynDOLOD_Data\DynDOLOD_NG_Tamriel.txt from the 171 outputs so I can compare it to the one from the 173 output you already uploaded.

 

My older setup is DynDOLOD Alpha 171, DynDOLOD Resources Alpha-49, DynDOLOD DLL NG Alpha 27.

Originally when I posted about this issue, I was not running in and out the Whiterun gate as much as I've been doing, because I started troubleshooting some other issues, it's the reason I then started to constantly go in an out of the gate, then I started to notice this. So before I didn't think there was an issue on the older setup with DynDOLOD Alpha 171 and DynDOLOD DLL NG Alpha 27, but I did notice a few times, hardly noticeable, covering very small areas, that you really have to go to extreme lengths running sometimes in and out the gate non stop for upwards of 10 minutes.

1. With DynDOLOD DLL NG Alpha 27, I spent around 10 minutes running in and out the gate. I did notice two small pops at different times, the first time two trees, then later around 100 square feet of the mountain side. Overall it was nearly flawless.

2. Next I installed DynDOLOD DLL NG Alpha 29 and DynDOLOD DLL NG Alpha 30 and the issues returned, like the video I posted.

Here are the logs...

DynDOLOD 3 Alpha 171 - DynDOLOD DLL NG Alpha 27

https://www.mediafire.com/file/1465neb9ap3fjsi/Dyn_171_Alpha_27_DynDOLOD_NG_Tamriel.txt/file

DynDOLOD 3 Alpha 171 - DynDOLOD DLL NG Alpha 29

https://www.mediafire.com/file/6kfzzkhybh37qmz/Dyn_171_Alpha_29_DynDOLOD_NG_Tamriel.txt/file

DynDOLOD 3 Alpha 171 - DynDOLOD DLL NG Alpha 30

https://www.mediafire.com/file/kx2hz0t9p2n4ml9/Dyn_171_Alpha_30_DynDOLOD_NG_Tamriel.txt/file

THANKS

Edited Thursday at 09:33 AM by mooit

[sheson]

9 hours ago, tweedledumb99 said:

I ran dyndolod a few months back with EVLaS and generated terrain underside through DynDOLOD and didn't have any problems with light shining through the ground (3 - Alpha 156).

I just reran DynDOLOD (3 - alpha 173) with underside at the same quality/height, still using EVLaS, and now there's light shining through the terrain. I've tried adding mods to give mountains a double-sided feature so light doesn't shine through, and followed the instructions on terrain underside on the DynDOLOD site on that topic. Not sure what's wrong or how to fix it, I've reran the latest DynDOLOD 3 Alpha times now with different options and they all lead to light shining through the landscape.

Is there a way I can just continue using Alpha 156 to regenerate LOD? since it seems like that version gave me working terrain underside (I still have my old output and if I swap it back in, everything's fixed). Only problem is when I try running the old one I get an error message saying I need to update to the new version, even though there isn't any feature from the new version I want to use over alpha 156.

Read the first post and/or https://dyndolod.info/Official-DynDOLOD-Support-Forum#Post-Logs which DynDOLOD log and debug log to upload when maing posts.

Properly report the problem so we can troubleshoot it in order to actually fix it.

https://dyndolod.info/Official-DynDOLOD-Support-Forum#Use-the-Latest-Versions
https://dyndolod.info/FAQ#Older-versions

There were no active changes to the underside generation or how it works since Alpha-157. Nobody else reported any similar problems with it since.

[sheson]

6 hours ago, mooit said:

 

My older setup is DynDOLOD Alpha 171, DynDOLOD Resources Alpha-49, DynDOLOD DLL NG Alpha 27.

Originally when I posted about this issue, I was not running in and out the Whiterun gate as much as I've been doing, because I started troubleshooting some other issues, it's the reason I then started to constantly go in an out of the gate, then I started to notice this. So before I didn't think there was an issue on the older setup with DynDOLOD Alpha 171 and DynDOLOD DLL NG Alpha 27, but I did notice a few times, hardly noticeable, covering very small areas, that you really have to go to extreme lengths running sometimes in and out the gate non stop for upwards of 10 minutes.

1. With DynDOLOD DLL NG Alpha 27, I spent around 10 minutes running in and out the gate. I did notice two small pops at different times, the first time two trees, then later around 100 square feet of the mountain side. Overall it was nearly flawless.

2. Next I installed DynDOLOD DLL NG Alpha 29 and DynDOLOD DLL NG Alpha 30 and the issues returned, like the video I posted.

Here are the logs...

DynDOLOD 3 Alpha 171 - DynDOLOD DLL NG Alpha 27

https://www.mediafire.com/file/1465neb9ap3fjsi/Dyn_171_Alpha_27_DynDOLOD_NG_Tamriel.txt/file

DynDOLOD 3 Alpha 171 - DynDOLOD DLL NG Alpha 29

https://www.mediafire.com/file/6kfzzkhybh37qmz/Dyn_171_Alpha_29_DynDOLOD_NG_Tamriel.txt/file

DynDOLOD 3 Alpha 171 - DynDOLOD DLL NG Alpha 30

https://www.mediafire.com/file/kx2hz0t9p2n4ml9/Dyn_171_Alpha_30_DynDOLOD_NG_Tamriel.txt/file

THANKS

All the uploads are the same file, because the data file ..\skse\plugins\DynDOLOD_Data\DynDOLOD_NG_Tamriel.txt that is created by DynDOLOD when generating the LOD patch does not change, even when running the game.

In any case I can now compare file between in the 2 outputs.

Now lets do this:
set debug=true and add debuglevel=5 in each ..\skse\plugins\DynDOLOD_Data\DynDOLOD_NG_Worlds.txt of the old and new output.
Install and use latest DynDOLOD DLL Alpha-30, with both old and new output.
Start the game, coc WhiterunDrunkenHuntsman and go through the gate for both old and new outpout.
Between switching old and new output rename/upload C:\Users\ROG\Documents\My Games\Skyrim Special Edition\SKSE\DynDOLOD.log accordingly for the 2 tests.

[tweedledumb99]

1 hour ago, sheson said:

Read the first post and/or https://dyndolod.info/Official-DynDOLOD-Support-Forum#Post-Logs which DynDOLOD log and debug log to upload when maing posts.

Properly report the problem so we can troubleshoot it in order to actually fix it.

https://dyndolod.info/Official-DynDOLOD-Support-Forum#Use-the-Latest-Versions
https://dyndolod.info/FAQ#Older-versions

There were no active changes to the underside generation or how it works since Alpha-157. Nobody else reported any similar problems with it since.

thanks, when I try rerunning it and generate logs, will post them along with the other requirements.

[sheson]

1 hour ago, tweedledumb99 said:

thanks, when I try rerunning it and generate logs, will post them along with the other requirements.

Also make two screenshots to show the issue. One normal and the other after typing tll in console to disable all LOD.

[mooit]

8 hours ago, sheson said:

All the uploads are the same file, because the data file ..\skse\plugins\DynDOLOD_Data\DynDOLOD_NG_Tamriel.txt that is created by DynDOLOD when generating the LOD patch does not change, even when running the game.

In any case I can now compare file between in the 2 outputs.

Now lets do this:
set debug=true and add debuglevel=5 in each ..\skse\plugins\DynDOLOD_Data\DynDOLOD_NG_Worlds.txt of the old and new output.
Install and use latest DynDOLOD DLL Alpha-30, with both old and new output.
Start the game, coc WhiterunDrunkenHuntsman and go through the gate for both old and new outpout.
Between switching old and new output rename/upload C:\Users\ROG\Documents\My Games\Skyrim Special Edition\SKSE\DynDOLOD.log accordingly for the 2 tests.

 

Hello sheson,

DynDOLOD_NG_Worlds.txt - this is correct?

[Config]
bunchofnumbers=0N3O4S3E7H5S88O5T8T1I5M3B2U4S2
esmplugin=dyndolod.esm
espplugin=dyndolod.esp
glowmultiple=1.75
neargrid=11
fargrid=21
debug=true
debuglevel=5
[Worlds]
tamriel
dlc2solstheimworld
dlc01soulcairn
dlc2apocryphaworld
skuldafnworld
sovngarde
deepwoodredoubtworld
dlc1hunterhqworld
dlc01falmervalley
japhetsfollyworld
blackreach

Old Output
https://www.mediafire.com/file/gijl7kcy1j0yzb8/Dyn_171_DLL_NG_30_DynDOLOD.log/file
New Output
https://www.mediafire.com/file/zx0gxd0hbs2uzo5/Dyn_173_DLL_NG_30_DynDOLOD.log/file

Something else I noticed, at first I thought I was imaging this. On the Old Config it only takes 5 seconds to go through the gate. On the New Config it is twice as long, 10-13 seconds, it will vary, and I've been noticing this for several hours, it's always twice as long. It's as though this is also introducing a lag.

THANKS

Edited Thursday at 11:50 PM by mooit

[sheson]

9 hours ago, mooit said:

Hello sheson,

DynDOLOD_NG_Worlds.txt - this is correct?

[Config]
bunchofnumbers=0N3O4S3E7H5S88O5T8T1I5M3B2U4S2
esmplugin=dyndolod.esm
espplugin=dyndolod.esp
glowmultiple=1.75
neargrid=11
fargrid=21
debug=true
debuglevel=5
[Worlds]
tamriel
dlc2solstheimworld
dlc01soulcairn
dlc2apocryphaworld
skuldafnworld
sovngarde
deepwoodredoubtworld
dlc1hunterhqworld
dlc01falmervalley
japhetsfollyworld
blackreach

Old Output
https://www.mediafire.com/file/gijl7kcy1j0yzb8/Dyn_171_DLL_NG_30_DynDOLOD.log/file
New Output
https://www.mediafire.com/file/zx0gxd0hbs2uzo5/Dyn_173_DLL_NG_30_DynDOLOD.log/file

Something else I noticed, at first I thought I was imaging this. On the Old Config it only takes 5 seconds to go through the gate. On the New Config it is twice as long, 10-13 seconds, it will vary, and I've been noticing this for several hours, it's always twice as long. It's as though this is also introducing a lag.

THANKS

The changed load order resulted in a different DynDOLOD output. Whatever changed in the load order between generation may cause directly or indirectly more work/longer loading times. For example, the number of cells requiring large reference bugs workarounds seemed to have more than doubled from 299 to 678. The work in DynDOLOD DLL NG is done on thread independent from the mainthread loading things, which is one of the reasons you actually see what you are seeing.

The video you uploaded earlier shows a loading time fade to/from black of under 5 seconds. Restore whatever config changes you did for testing, including turning off debug.

Replace DynDOLOD.DLL with this test version and check if you no longer see the LOD loading in https://mega.nz/file/ABIy2aAY#29GWeeITWeJlf_S_3kFVJVctEKsZBh2lwnH6GvB0dhU

 

[mooit]

1 hour ago, sheson said:

The changed load order resulted in a different DynDOLOD output. Whatever changed in the load order between generation may cause directly or indirectly more work/longer loading times. For example, the number of cells requiring large reference bugs workarounds seemed to have more than doubled from 299 to 678. The work in DynDOLOD DLL NG is done on thread independent from the mainthread loading things, which is one of the reasons you actually see what you are seeing.

The video you uploaded earlier shows a loading time fade to/from black of under 5 seconds. Restore whatever config changes you did for testing, including turning off debug.

Replace DynDOLOD.DLL with this test version and check if you no longer see the LOD loading in https://mega.nz/file/ABIy2aAY#29GWeeITWeJlf_S_3kFVJVctEKsZBh2lwnH6GvB0dhU

 

 

Hi sheson,

Ok...

Thanks

Edited Friday at 09:36 AM by mooit

[sheson]

10 minutes ago, mooit said:

Hi sheson,

I accidentally deleted my new load output from xLODGen, TexGen, DynDOLOD, so I needed to rerun them all over.

I'm assuming I'm placing this DynDOLOD.DLL in \Skyrim Special Edition\Data\SKSE\plugins?

So when I tried to do this with DynDOLOD.DLL and DynDOLOD.pdb I get a popup from DynDOLOD; Wrong papyrus scripts.

 

Use the Copy message to clipboard and paste the text instead of posting screenshots of text. See https://dyndolod.info/Official-DynDOLOD-Support-Forum#Copy-and-Paste-Text

Obviously keep the existing scripts from DynDOLOD DLL NG and Script Alpha-30 and only replace the DynDOLOD.DLL and .PDB in the ..\SKSE\Plugins\ folder with the test version to check if it makes a difference.

[mooit]

2 hours ago, sheson said:

Use the Copy message to clipboard and paste the text instead of posting screenshots of text. See https://dyndolod.info/Official-DynDOLOD-Support-Forum#Copy-and-Paste-Text

Obviously keep the existing scripts from DynDOLOD DLL NG and Script Alpha-30 and only replace the DynDOLOD.DLL and .PDB in the ..\SKSE\Plugins\ folder with the test version to check if it makes a difference.

Unfortunately I accidentally deleted my new load output, so I'm remaking them over.

Sorry about the screen shot...

I am not seeing any issues at the moment with the Test DLL, it's looking better, but I haven't had the time to test it long enough. I will test more later and get back to you. But for the time I tested it, I posted a DynDOLOD log.

In regards to the video I posted showing 5 seconds through the gate, occasionally the time is ok, most of the time it's not, and using this Test DLL, if that's the issue, 10 minutes of testing running in and out the gate, every time was 10-13 seconds. Something has changed, because it was always around 5 seconds before, now it's twice as long.

Here's the new output 173 with the test DLL, with the debug off.

https://www.mediafire.com/file/g7lhvpo3ckz9m3t/Dyn_173_Test_DLL_DynDOLOD.log/file

THANKS

Edited Friday at 12:09 PM by mooit

[sheson]

7 hours ago, mooit said:

Unfortunately I accidentally deleted my new load output, so I'm remaking them over.

Sorry about the screen shot...

I am not seeing any issues at the moment with the Test DLL, it's looking better, but I haven't had the time to test it long enough. I will test more later and get back to you. But for the time I tested it, I posted a DynDOLOD log.

In regards to the video I posted showing 5 seconds through the gate, occasionally the time is ok, most of the time it's not, and using this Test DLL, if that's the issue, 10 minutes of testing running in and out the gate, every time was 10-13 seconds. Something has changed, because it was always around 5 seconds before, now it's twice as long.

Here's the new output 173 with the test DLL, with the debug off.

https://www.mediafire.com/file/g7lhvpo3ckz9m3t/Dyn_173_Test_DLL_DynDOLOD.log/file

THANKS

Since you generated new output, did you verify the LOD issue still happens with Alpha-30 or older?
The log looks like you loaded a not cleaned save again.

If debug=true/false does not affect times (it shouldn't really anyhow), it seems to be unrelated to DynDOLOD.

[drift123]

I read the page about the Child/Parent Worldspace copies but it's still unclear to me. I know the Parent/Child settings affect what you see when inside Whiterun looking over the city walls. Does it affect what you see when you're outside of Whiterun looking over the city walls?

It's unclear to me what the difference between Parent > Child and Child > Parent settings are in terms of in-game visuals.

Edited Friday at 08:14 PM by drift123

[sheson]

35 minutes ago, drift123 said:

I read the page about the Child/Parent Worldspace copies but it's still unclear to me. I know the Parent/Child settings affect what you see when inside Whiterun looking over the city walls. Does it affect what you see when you're outside of Whiterun looking over the city walls?

It's unclear to me what the difference between Parent > Child and Child > Parent settings are in terms of in-game visuals.

https://dyndolod.info/Help/Child-Parent-Worldspace-Copies
Walled cities like Solitude, Riften, Whiterun, Windhelm etc. are child worldspaces that are embedded into the Tamriel parent worldspace that they use for LOD.
The Child > parent option copies references that could have LOD from the child worldspaces to the parent worldspace.
The Parent > child option copies references that have LOD from the parent worldspace to the child worldspaces.

The Child > parent option makes sure that things only added to the inside of city can also be seen from the outside.
The Parent > child option makes sure that things can also be seen when inside a city and peeking past the walls.

[mooit]

4 hours ago, sheson said:

Since you generated new output, did you verify the LOD issue still happens with Alpha-30 or older?
The log looks like you loaded a not cleaned save again.

If debug=true/false does not affect times (it shouldn't really anyhow), it seems to be unrelated to DynDOLOD.

 

Hi sheson,

I used the test dll with Alpha 30 as you mentioned...

The game save was the new game I created.

On Alpha 173 with DLL Alpha 30 using the Test DLL, I do not see this issue, it looks like it has corrected the problem. I do have other issues to work on, and I will just be sticking around Whiterun doing these other modding chores, and if I happen to see it again, I will reply back.

All I know is, the only thing I changed when I started to notice the increase in wait times through the gate was upgrading DynDOLOD to 173. After adding in the test dll to Alpha 30, yesterday the wait times through the gate were still twice the time. Now today, it only happened a few times, and the majority of times it went back to 5 seconds. So I really don't get why it changes like this at times, because in the past I never saw this, and since the beginning of the year, I have been modding daily at Whiterun. 95% of my modding for the past year has only been at Whiterun, I've run in and out of that gate hundreds of times, possibly a thousand, and when I saw the change, I noticed it.

Oh well, I guess the game can have a mind of it's own sometimes... hmm

THANKS

Edited Friday at 10:55 PM by mooit

[sheson]

5 hours ago, mooit said:

Hi sheson,

I used the test dll with Alpha 30 as you mentioned...

The game save was the new game I created.

On Alpha 173 with DLL Alpha 30 using the Test DLL, I do not see this issue, it looks like it has corrected the problem. I do have other issues to work on, and I will just be sticking around Whiterun doing these other modding chores, and if I happen to see it again, I will reply back.

All I know is, the only thing I changed when I started to notice the increase in wait times through the gate was upgrading DynDOLOD to 173. After adding in the test dll to Alpha 30, yesterday the wait times through the gate were still twice the time. Now today, it only happened a few times, and the majority of times it went back to 5 seconds. So I really don't get why it changes like this at times, because in the past I never saw this, and since the beginning of the year, I have been modding daily at Whiterun. 95% of my modding for the past year has only been at Whiterun, I've run in and out of that gate hundreds of times, possibly a thousand, and when I saw the change, I noticed it.

Oh well, I guess the game can have a mind of it's own sometimes... hmm

THANKS

It looks like there have been plenty of other changes in the load between old and new output.
It seems you made further changes in the load order that rectified whatever caused seeing the LOD loading in late, too.