neomonkeus

Core Team
  • Content count

    1638
  • Joined

  • Last visited

Community Reputation

0 Neutral

1 Follower

About neomonkeus

  • Rank
    Advanced Member
  1. Each unique material effective results in having to split out what were commonly shared vertices into a new mesh. The nif format doesn't support sharing vertices across materials like it does in blender.
  2. Yes, that version is not supported atm. I think a few people have been inquiring about it recently. one of the main issues with that version of the format is that it uses the NiMesh block instead of NiTriShapes, for holding topology data. The main difference is that it is a streaming block, so has to be decoded in a different manner, something which the nif.xml is not currently suited to support.
  3. Care to share?
  4. I ended up creating a project for all property related tickets - https://github.com/niftools/blender_nif ... projects/1
  5. Thanks for the really thorough report. That whole area needs a refactor. I have had a go a few times at this stage. I am designing a system to handle all custom data that we need to store - https://github.com/niftools/blender_nif ... processing Design is ok, just that there is currently limitations in what object types i.e bones, mesh, lights can store certain property types. Talking with the Blender developers, there are hoping to apply the system to all types essentially.
  6. The probably are using another tool eg. Nifskope or 3ds to output into a format that blender can read.
  7. If you have a github account can you create a new issue, filling in the details of the template. If not, please let me know and I will copy it here and we can complete it together.
  8. Fallout 4 is not support by the plugin, thus you are getting the "ValueError: string too long (0x64060000 at 0x0000003A)" error. The 2.49b release is nolonger supported. The 2.6 series is in development, but is in pre-alpha. An there are several updates required to the underlying nif reading library to support Fo4, like this one. - https://github.com/neomonkeus/pyffi/tree/feature/hfloat This will be part of 2.2.4, but I am actively working on several priority bugs for the 2.2.3 release - https://github.com/niftools/pyffi/pulse/monthly The above updates will need to be complete before we can even think about the blender side, which we need to stabilize the old functionality still.
  9. Verified the downloads also.
  10. Bone names are stored in file as part of the blend file, so if they imported incorrectly, then I would suspect it would remain as such in the existing .blend file, even post fix Updated the issue on github with the additional information.
  11. Created - https://github.com/niftools/blender_nif ... issues/262 If you can provide I can further fill out the ticket.
  12. Created issue - https://github.com/niftools/blender_nif ... issues/261 If you can provide any more info that would be great.
  13. Is there any progress on the fix? There are a few of us in my modding circle that would really like to use the advanced features of Blender 2.7x, especially now with Skyrim-SE. Thank you for the nif tools; and my apologies if I am being impatient. :wink: ETA1: I copied the 4 scripts that have been listed in this thread and saved them as .py files in notepad++. I made a new zip file with the changes and attempted to load it into Blender; which failed. Blender wouldn't read it. So, I loaded the original plugin and then went into Blender's plugin folder and copied the changes into it. When I exported my model I got only one error: The nif file looks really good. I haven't yet gone over it in detail, nor have I tried it in game; but it's a step in the right direction. ETA2: Exported both bodies (0,1) and tested in game. The body exploded. So, I tried both of the batch spells: Update all Tangent Spaces and Make all Skin Partitions. I used only the Bones per Vertex and Bones per Partition, which found a few verts with greater than 4 influences (probably from the HDT Weights). I reloaded it in game and all looks good. While I was posting this I tried exporting with an increase of Weights per vertex to 5. That eliminated the warning. I will be tweaking the Weight Maps anyway, so that will no longer be a problem. Everything else seems to be working as intended. I cannot express how very nice it is to be able to import and export meshes without all the adjustments in nifskope, as well as using all the advances of Blender 2.78. I am a very happy camper. :lol: The person updated the zip file with the fix manually.
  14. Can you post the output failure you were getting when you try with the above. Also note that I think the specific toaster spell would need to be updated if the block types have changed in f04. Also, just to point out that the version of nif.xml reference in the pyffi commit is probably several versions behind the NifSkope version.
  15. You can try the last thing I was working on here before taking some time off - https://github.com/neomonkeus/pyffi/tree/hfloat I can't remember if there was any additional work required after that point but if you see how it goes and provide any errors then can go from there.