DynDOLOD 3.00 Alpha 180

[z929669]

Yes, that should do it. Change and restart PC. If you still get the error, then sheson may have some instruction.

[Viv]

Same result unfortunately. 

TexGen_SSE_Debug_log.txt

TexGen_SSE_log.txt

[sheson]

6 hours ago, Viv said:

Same result unfortunately. 

TexGen_SSE_Debug_log.txt

TexGen_SSE_log.txt

How much VRAM does the graphics card have? Ccheck its memory consumption while running the TexGen.

[Viv]

3 hours ago, sheson said:

How much VRAM does the graphics card have? Ccheck its memory consumption while running the TexGen.

Sorry for the hasty screenshot but I get the "out of memory" error seconds after the RAM peaks at 100%. There was one point here when the integrated GPU briefly bit 100% but Texgen kept going no problem. I'll assign my dedicated GPU (6GB VRAM) to it and try again today but based on this, not sure if it will make any difference.

In the end I ran the Beta version yesterday as usual and didn't give any issues, but I won't give up on 3.0 just yet.

[sheson]

49 minutes ago, Viv said:

Sorry for the hasty screenshot but I get the "out of memory" error seconds after the RAM peaks at 100%. There was one point here when the integrated GPU briefly bit 100% but Texgen kept going no problem. I'll assign my dedicated GPU (6GB VRAM) to it and try again today but based on this, not sure if it will make any difference.

In the end I ran the Beta version yesterday as usual and didn't give any issues, but I won't give up on 3.0 just yet.

The out of memory message refers to video memory. What physical memory the video memory consists of depends on the hardware/OS/drivers. TexGen should be using the "first" GPU.

[Viv]

3 hours ago, sheson said:

The out of memory message refers to video memory. What physical memory the video memory consists of depends on the hardware/OS/drivers. TexGen should be using the "first" GPU.

Hah now I see my misunderstanding! I assigned my dedicated GPU to Texgen this time and everything ran smoothly. Thanks so much for your help.

 

[wbell03]

When using Dyndolod 3.0, some of the trees LOD are like twice the size of the normal trees in game, making everything look extremely weird as the trees shrink in size dramatically when I get close.

Did I do something wrong? I basically used all the setting as default with the ultra tree box checked. Im using Jedi Trees.

[sheson]

31 minutes ago, wbell03 said:

When using Dyndolod 3.0, some of the trees LOD are like twice the size of the normal trees in game, making everything look extremely weird as the trees shrink in size dramatically when I get close.

Did I do something wrong? I basically used all the setting as default with the ultra tree box checked. Im using

 

Yes you did. Read the first post. Especially the part about which log files to post.

Old and new billboard texture and text file generation is explained in great detail in the included documents.

[z929669]

1 hour ago, wbell03 said:

When using Dyndolod 3.0, some of the trees LOD are like twice the size of the normal trees in game, making everything look extremely weird as the trees shrink in size dramatically when I get close.

Did I do something wrong? I basically used all the setting as default with the ultra tree box checked. Im using Jedi Trees.

If you are using EVT - Lush/Large with the Alternative Trees optional file, then this is a known issue. I have submitted an updated version to the mod authors, which should be available soon-ish.

If you are not using EVT with those options, please disregard this post.

[z929669]

DynDOLOD quit unexpectedly after about 40 min with no helpful details but for the following from end of LOSGen_SSE_Tamriel_log.txt:

...
plants\srg_shrub01half_n.dds=7: 2162 -> 2245 | 130, 4
Log ended at 2:29:57 PM (35:41)
Code: 0

What is "Code: 0"? First issue I have had for alpha 33 (I have run with many varieties of settings around 30x on full modded setup). The new *Threads* INI params have completely resolved memory management issues for me (32 Gb system RAM, 8 Gb VRAM here), which occasionally 'broke' running object LOD in 2.xx beta when upwards of 30 Gb system RAM utilized.

I will run again and post back with more detail if issue repeats.

[sheson]

6 minutes ago, z929669 said:

DynDOLOD quit unexpectedly after about 40 min with no helpful details but for the following from end of LOSGen_SSE_Tamriel_log.txt:

...
plants\srg_shrub01half_n.dds=7: 2162 -> 2245 | 130, 4
Log ended at 2:29:57 PM (35:41)
Code: 0

What is "Code: 0"? First issue I have had for alpha 33 (I have run with many varieties of settings around 30x on full modded setup). The new *Threads* INI params have completely resolved memory management issues for me (32 Gb system RAM, 8 Gb VRAM here), which occasionally 'broke' running object LOD in 2.xx beta when upwards of 30 Gb system RAM utilized.

I will run again and post back with more detail if issue repeats.

Are you saying it quit without saving any log files or a bugreport?

Code: 0 has always been the last line of the LODgen logs when everything went well.

[z929669]

It's strange and looks like it's going to happen again. Symptom is that the DynDOLOD xEdit window closes, while LODGenx64.exe process continues running. Once that is finished, it just closes, so my issue is the DynDOLOD process. I will post relevant log(s) once completed.

[wbell03]

2 hours ago, sheson said:

Yes you did. Read the first post. Especially the part about which log files to post.

Old and new billboard texture and text file generation is explained in great detail in the included documents.

Fixed the issue. I went to the "trees addon" page on nexus and put the settings they suggested for dyndolod on their main page and now everything is working perfectly. Thank you for the quick response. 

[z929669]

So no bug report or anything useful. All logs for this run attached, including TexGen logs in case they have any clues. I am running again and will watch the DynDOLOD xEdit process this time ... meh. Logs.7z

EDIT: My DynDOLOD settings for these failed runs. This is the only piece for which I have made any changes. INI settings haven't changed for past 20-ish successful runs. Maybe the Billboard2 changes for the wrTemple* trees? I assume all settings here have fallbacks.

EDIT: Well, I guess the app wanted my attention is all. Absolutely no changes aside from me watching it for 20 minutes straight to confirm the LODGen64 process started without DynDOLOD closing. Completed successfully this time with those settings.

[sheson]

8 hours ago, z929669 said:

So no bug report or anything useful. All logs for this run attached, including TexGen logs in case they have any clues. I am running again and will watch the DynDOLOD xEdit process this time ... meh. Logs.7z

EDIT: My DynDOLOD settings for these failed runs. This is the only piece for which I have made any changes. INI settings haven't changed for past 20-ish successful runs. Maybe the Billboard2 changes for the wrTemple* trees? I assume all settings here have fallbacks.

EDIT: Well, I guess the app wanted my attention is all. Absolutely no changes aside from me watching it for 20 minutes straight to confirm the LODGen64 process started without DynDOLOD closing. Completed successfully this time with those settings.

I a program is being terminated like without an exception it is typically because of something outside  of a program code error, like memory, overclocking,  error etc. There might be something in the Windows event log.