Page 1 of 1
Shadowvolume problem [Solved]
Posted: Tue Feb 23, 2016 6:31 am
by giftheck
I'm getting a munge crash on my model when I go to munge. This is after creating a new shadowvolume.
Here is the mungelog:
Code: Select all
WARNING[PC_modelmunge msh\imp_inf_sandtrooper.msh]:SplitSkinnedSegments: WARNING: out of space in splitting catalog!
WARNING[PC_modelmunge msh\imp_inf_sandtrooper.msh]:_CreateShadowMesh(bone:DummyRoot,lod:0): ignoring duplicate tri 201,203,200
WARNING[PC_modelmunge msh\imp_inf_sandtrooper.msh]:imp_inf_sandtrooper has 7094 vertices and NO COLLISION GEOMETRY (WILL BE SLOW)!
WARNING[PC_modelmunge msh\imp_inf_stormtrooper.msh]:_CreateShadowMesh(bone:DummyRoot,lod:0): ignoring duplicate tri 1,3,0
ERROR[PC_modelmunge msh\imp_inf_stormtrooper.msh]:_CreateShadowMesh(bone:DummyRoot,lod:0): edges to close (54) not predicted correctly
ERROR[PC_modelmunge msh\imp_inf_stormtrooper.msh]:_ReduceShadowMesh: edge->next also belongs to another face!
ERROR[levelpack req\imp_inf_rifleman.req]:Expecting bracket, but none was found.
File : munged\pc\imp_inf_stormtrooper.model.req(1)...
ucft <--
ERROR[levelpack req\imp_inf_rifleman.req]:Expecting bracket, but none was found.
File : munged\pc\imp_inf_stormtrooper.model.req(1)...
ucft <--
2 Errors 0 Warnings
ERROR[levelpack impshell.req]:Expecting bracket, but none was found.
File : munged\pc\imp_inf_stormtrooper.model.req(1)...
ucft <--
ERROR[levelpack impshell.req]:Expecting bracket, but none was found.
File : munged\pc\imp_inf_stormtrooper.model.req(1)...
ucft <--
2 Errors 0 Warnings
The .req files have not been changed in any way. This has only happened after creating a new shadowvolume from a duplicated mesh. I also did one for my Sandtrooper but I recieved no such errors for it during the munge.
Re: Shadowvolume problem
Posted: Tue Feb 23, 2016 7:01 am
by Anakin
can you post a screenshot of your scene with the explorer expanded??
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 4:23 am
by giftheck
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 5:13 am
by thelegend
I am not sure about tgat but maybe rename
sv_stormtrooperr to
sv_stormtrooper_mesh_new OR just
shadowvolume. I am not sure if the last one works for Battlefront II too.
Edit: By the way. How do you export your unit? Which steps do you take?
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 6:18 am
by giftheck
I'll try your suggestion out.
What I do to export is middle click the dummyroot then go to "Export .MSH" in Mod Tool.
I just find it curious it won't do it this way. I did a different method for my Sand Trooper as it's two meshes, I just made a merged copy without any UV data and that worked fine.
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 6:30 am
by Anakin
sv_stormtrooper should work just fine.
can you send me the corrupted msh file??
btw you need a normal klick on the dummyroot for ZETools. No middle klick.
Try to delete the cluster Sample. Maybe that works
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 6:59 am
by giftheck
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 7:41 am
by thelegend
I tested it and I got the same errors as you had. Then I duplicated the original mesh, renamed it to sv_stormtrooper_mesh_new and exported again. I didn't receive any errors.
But when I reduced down the poly count of the shadowvolume mesh I received about 20 Errors. I have just imported the stock stormtrooper model, used it's shadowvolume and I had no errors nor crashed while munging. I didn't test it ingame yet but it should work. My advice: Use the stock shadowvolume. You could still modify it adding some simple details onto it but it might be the best to use a stock sv instead of a too high polyed shadow volume which could break the game. Especially munging the model took a while. You can keep the default poly count of your model but I would recommend using the stock shadowvolumes (With minor or major modifications) or a custom one with less than 500-600 triangles.
Maybe Anakin or someone else will get your sv_ working. I tried my best. By the way I had the same problem and used -hiresshadow to generate a 1:1 shadowvolume. It works like a charm, but It eats some memory and munging could take longer than usual. Otherwise it's a good alternate way.
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 7:49 am
by giftheck
Ah, I see.
It's interesting because this only happens with a duplicated version of the original mesh. I have a Sandtrooper too, and the way I created the shadowvolume for that is a little different as instead of duplicating the mesh I merged it with another mesh (my sandtrooper has the main body mesh and the backpack mesh separate), reduced polys then exported and it works fine.
For the moment I will take your advice and borrow the stock Stormtrooper's SV.
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 9:07 am
by Anakin
i tested, too. And i got the same result.
I really don't know why, since it worked for me. I made the same steps for the SWBF3 clones (the ep3 style. The others are merged, to get their addons)
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 9:28 am
by giftheck
Could it be because it's an already-merged model?
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 9:45 am
by Anakin
i don't think so. But i'm waiting for Ace to explain it to us
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 3:08 pm
by AceMastermind
You guys already have the best solution figured out here by using a stock Stormtrooper sv mesh, but I had a look at your character anyway.
Shadowvolume geometry should be solid, or closed on all sides. Several armor pieces were 2d geometry, I removed these from the sv mesh and exported and munged with only 2 warnings.
Code: Select all
WARNING[PC_modelmunge msh\imp_inf_stormtrooper.msh]:_CreateShadowMesh(bone:DummyRoot,lod:0): ignoring duplicate tri 23,25,24
WARNING[PC_modelmunge msh\imp_inf_stormtrooper.msh]:imp_inf_stormtrooper has 6324 vertices and NO COLLISION GEOMETRY (WILL BE SLOW)!
0 Errors 2 Warnings
Warnings like the first one might be a result of the merging you were doing, I usually merge objects with a .001 tolerance.
Re: Shadowvolume problem
Posted: Wed Feb 24, 2016 3:20 pm
by giftheck
Ah, thanks. I'll give my copy another looking over. I think this can be filed as 'solved now'.
