search

niftools / nifxml

A repository for the nif.xml file, which contains the nif file format description.
http://www.niftools.org
GNU General Public License v3.0
37 stars 43 forks source link

Havok blocks update #39

Closed ttl269 closed 7 years ago

ttl269 commented 9 years ago

@neomonkeus @skyfox69 @jonwd7 @Ghostwalker71 @throttlekitty Update of some havok blocks.

  1. Most of changes are related to constraints. Changes cause that Nifskope don't render any havok constraints - structure of bhkConstraint block has been modified. If this will be merged it will require to modify Nifskope's code too. See
  2. Added three new unknown Skyrim materials. Little modification in bhkRigidBody block. See
  3. Extended the description of Unknown Int 2 in NiGeometryData. See
Ghostwalker71 commented 9 years ago

@neomonkeus @ttl269 @jonwd7 I reviewed these changes and though I haven't had a chance to run them through testing I do think they are solid and ready for inclusion. @jonwd7 what do you think would be a time frame to adding the changes to nifskope?

hexabits commented 9 years ago

@Ghostwalker71 I wouldn't make the changes until the XML is officially updated anyway (committed to master branch). That way a new XML version is already for sure finalized and I can go about doing XML version checking to decide which path to take when it comes to certain features. Thing is, I still have to write some kind of XML version checking framework.

Though in practice it might turn out to be too messy to try to support both old and new XML versions and I might end up scrapping the idea. The alternative is to enforce a minimum XML version requirement so that when a feature breaks and I adjust NifSkope to work again with that feature, I warn the user that the XML they are using no longer functions correctly with the new version of NifSkope. Realistically this would only happen if people are manually replacing their XML anyway.

Ghostwalker71 commented 9 years ago

@jonwd7 I would personally avoid trying to support multiple versions of the xml, Those who manually edit their xml can manage their end, though an error check to warn the user if they happen to be using an unsupported version is a good thing.

neomonkeus commented 9 years ago

Yeah I think it should be just rolling support and get them to update. Most people aren't averse to doing that as they would rather update and get support.

The other thing is that the xml should be tagged at points so you can allow applications to support specific versions. I know the pyffi project has been updated to the latest master, but it is better to actually select an actual version. Ideally you should be updating to the newer version, but you also need to ensure that the changes are stable too.

Also this needs rebasing, I'll see if I can do it at some point this week.

neomonkeus commented 9 years ago

@Ghostwalker71 you had a comment about morrowind meshes. Is that still an issue?

Ghostwalker71 commented 9 years ago

It is still an issue, but I discovered it is not related to this pull, but rather is already in the development branch. For some reason morrowind nifs are setting the num triangles value to an extremely high value, a mesh with 146 vertices sets num triangles to 47677, yet I can find no reason for it.