World of Warships

[Wobble]

[out]

[Coffee]

MP as far as I can decipher means Misc part because they always are located in the misc folders. I could be wrong on this tho.

If that's right, then you found a tool to find where unknown HP and MP identifiers are located. MP are always locted in <country>\misc or common\misc folders. And HP are located elsewhere. Should help speed up the discovery process.

But, there are parts called:
MP_CM001_Propeller3_L
MP_JM112JSB018Stern_0_0

It may just be by coincidence that MP is always using misc folders. The abbreviations CM and JM definitely stand for "common misc" and "japan misc", so MP must stand for something else, like Medium Priority, as Andrakann suggested.

The only exception to this rule seems to be HP_flag_nation. Located in japan\misc\JM001\JM001.model. So, the flag must be important, high priority.

That could be..
I just noticed something .. and I'd like you guys to look in to this..
Im looking at the Yamato, all the guns but the secondary are pointing in the wrong direction. If this is the case with the other ships, I can filter this and rotate the NON-secondary guns.

[Wobble]

[out]

[Andrakann]

So, the flag must be important, high priority.

Tested this moment - and yes, flags loads on earlier stage if you change ships.

I can filter this and rotate the NON-secondary guns.

Not rotate, but invert by all axis, so all sides must flips and faces inverts. All this I do manually for B23 exports.
But skinning support, as Wobble mentioned - would be better
Upd.: just tested - this problem exist in B29.3 too:

- all faces inverted and small director on top of turret placed at wrong side.
Screenshot from Deep Exploration, 3dsmax automatically makes faces double sided for "realistic" renderer window.
Gun exported as OBJ, because of bug with FBX exporter.


Wobble, did you test Mike's Editor B29.3 for WoWS? I can't get it working on my PC. For WoT all's ok, but WoWS primitives loads without textures and child objects, just like in older versions
I think I miss some system libs, or maybe RU locale of windows make troubles...

[Wobble]

[out]

[Wobble]

[out]

[Coffee]

Im looking at the Yamato, all the guns but the secondary are pointing in the wrong direction. If this is the case with the other ships, I can filter this and rotate the NON-secondary guns.

How are you transforming your primitives?
Are you using the skinning information?

The main big guns like HP_JGM_3 are skinned, so you have to transform their vertices by their bones.
Other meshes are static.

If you overlay the visual skeletal system, you should see the guns pointing in the same direction as your bones:

I was looking at the .visual for the guns.
In these XML files there is a root_bendbone node matrix with the same scaling as the all other bone nodes. It does not contain rotation or translation information.

Code: Select all

         <node>
            <identifier>   Root_BlendBone </identifier>
            <transform>
               <row0>   1.000000 0.000000 0.000000 0.000000 </row0>
               <row1>   0.000000 1.000000 0.000000 0.000000 </row1>
               <row2>   0.000000 0.000000 -1.000000 0.000000   </row2>
               <row3>   0.000000 0.000000 0.000000 1.000000 </row3>
            </transform>
         </node>

This is nearly an identity matrix only the Z is mirrored (row2 z = -1.0)
Now if we look at a bend_bone:

Code: Select all

              <node>
                  <identifier>   Rotate_X_BlendBone   </identifier>
                  <transform>
                     <row0>   1.000000 0.000000 0.000000 0.000000 </row0>
                     <row1>   0.000000 1.000000 0.000000 0.000000 </row1>
                     <row2>   0.000000 0.000000 -1.000000 0.000000   </row2>
                     <row3>   -0.002679 -0.108695 -0.205988 1.000000 </row3>
                  </transform>
               </node>

You see that the scaling is the same but there is translation data. This is the pivot point for the guns (up and down).
Ok... and finally the matrix as it is stored in the compound xml.

Code: Select all

        <node>
            <identifier>   HP_JGM_2 </identifier>
            <transform>
               <row0>   1.000000 0.000000 0.000000 0.000000 </row0>
               <row1>   0.000000 1.000000 0.000000 0.000000 </row1>
               <row2>   0.000000 0.000000 1.000000 0.000000 </row2>
               <row3>   0.000000 0.682223 2.024238 1.000000 </row3>
            </transform>
         </node>

once again.. there is scale .. (Identity in this case) and translation(into the ships space).. NO rotation
So the problem remains.. However, the difference in scaling between the root_bendbone and the compound matrix is interesting.
When I have time I will look at the relationship of the secondary's and the other gun's sign of ROW2 Z (3rd position). If all guns have negitive as the main gun above and only the secondaries are positive .. than that is the indicator of the guns rotation.. negitive would mean it needs rotated 180% or maybe just mirrored as the -Z would do.
Right now, Im to tired to think about it.

[Coffee]

Im looking at the Yamato, all the guns but the secondary are pointing in the wrong direction. If this is the case with the other ships, I can filter this and rotate the NON-secondary guns.

How are you transforming your primitives?
Are you using the skinning information?

The main big guns like HP_JGM_3 are skinned, so you have to transform their vertices by their bones.
Other meshes are static.

If you overlay the visual skeletal system, you should see the guns pointing in the same direction as your bones:

I was looking at the .visual for the guns.
In these XML files there is a Root_BlendBone node matrix with the same scaling as the all other bone nodes. It does not contain rotation or translation information.

Code: Select all

         <node>
            <identifier>   Root_BlendBone </identifier>
            <transform>
               <row0>   1.000000 0.000000 0.000000 0.000000 </row0>
               <row1>   0.000000 1.000000 0.000000 0.000000 </row1>
               <row2>   0.000000 0.000000 -1.000000 0.000000   </row2>
               <row3>   0.000000 0.000000 0.000000 1.000000 </row3>
            </transform>
         </node>

This is nearly an identity matrix only the Z is mirrored (row2 z = -1.0)
Now if we look at a Rotate_X_BlendBone :

Code: Select all

              <node>
                  <identifier>   Rotate_X_BlendBone   </identifier>
                  <transform>
                     <row0>   1.000000 0.000000 0.000000 0.000000 </row0>
                     <row1>   0.000000 1.000000 0.000000 0.000000 </row1>
                     <row2>   0.000000 0.000000 -1.000000 0.000000   </row2>
                     <row3>   -0.002679 -0.108695 -0.205988 1.000000 </row3>
                  </transform>
               </node>

You see that the scaling is the same but there is translation data. This is the pivot point for the guns (up and down).
Ok... and finally the matrix as it is stored in the compound xml.

Code: Select all

        <node>
            <identifier>   HP_JGM_2 </identifier>
            <transform>
               <row0>   1.000000 0.000000 0.000000 0.000000 </row0>
               <row1>   0.000000 1.000000 0.000000 0.000000 </row1>
               <row2>   0.000000 0.000000 1.000000 0.000000 </row2>
               <row3>   0.000000 0.682223 2.024238 1.000000 </row3>
            </transform>
         </node>

once again.. there is scale .. (Identity in this case) and translation(into the ships space).. NO rotation
So the problem remains.. However, the difference in scaling between the Root_BlendBone and the compound matrix is interesting.
When I have time I will look at the relationship of the secondary's and the other gun's sign of ROW2 Z (3rd position). If all guns have negitive as the main gun above and only the secondaries are positive .. than that is the indicator of the guns rotation.. negitive would mean it needs rotated 180% or maybe just mirrored as the -Z would do.
Right now, I'm to tired to think about it.

[Coffee]

The guns local transforms have to be added to the whats stored for that gun in the compound to bring the bone locations in to the ships space.

Andrakann.
Trying downloading it again.
I installed using that same exact file and it runs fine.. other than that damn memory leak.

EDIT.. I know some one that works for WarGaming in Seattle and I sent him an email.. Maybe he can get some info. I dont think so tho because he works on WoT stuff.

[Coffee]

which matrix are you using Wobble to transform the verts?

[Wobble]

[out]

[Andrakann]

did you test Mike's Editor B29.3 for WoWS?

Nope, got a download link to it?

It's in same folder with XML Tool here. Direct link.

Andrakann.
Trying downloading it again.
I installed using that same exact file and it runs fine.. other than that damn memory leak.

I'll try it on my wife's PC for test.

Upd: test failed
After start of Editor, it's gives me error:

This file is present in Editor's root.
I press [OK], but after i set working path and try to open .primitive, i get another error:

(full log attached)
None loads after i press [Continue].

I trying reinstall with default paths and settings (checked "Just Me", not "Everyone" as always), then trying to reinstall with Administrative rights - same errors here.

So, i think my wife's PC missing a lot more libraries than my does
On my PC that UK ship loads as always - no textures or childs, but no any errors too.
Versions of Windows are the same - 7 x64

[Wobble]

[out]

[Andrakann]

I noticed there is a sub folder called "WoWs_Scripts". Does this program actually load the Yamato script? If so, how to load it?

Must loads automatically, when you open Yamato parts.
Btw, do you have WoWS models textured or white in editor?

[Wobble]

[out]