Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[western0221]

I tested Max's auto reducing "MultiRes" down to 85%.

Then Faces are reduced, but SSC tower modeling is broken... 
It seems to be difficult to CPU auto reducing.
If more Faces 90% or 95%, there is no sense evenif modeling keep.

Finally I will (have to) divide objects more, like ashehouse said. Thanks.

[asheshouse]

Multi-Res will not create a good LOD0.
All LOD0 work needs to be done by hand to look best.
However I do use Multi-Res for making LOD1, LOD2 etc
Before using it manually remove minor details that you know will not have visual impact in the LOD
Then with the slimmed down mesh run MultiRes. Its ok for that stage.

Buggybuggy Exporter is not good at handling smoothing groups --- unless it has been improved.
I use the Maraz Exporter for getting my models into the game.
It was a little harder to learn but now is much easier and better quality export than buggybuggy

I cannot find it on SAS but its still available on AAS
http://allaircraftsimulations.com/forum/viewtopic.php?f=31&t=11214

Ashe

[SAS~GJE52]

Ashe,

I still use the Buggy exporter on Max 2009, not sure if it will support the Maraz Exporter - what version of Max do you run ?

G

[asheshouse]

I'm using it on Max 9.
It works as a Plug-In  accessed through File/Export on the menu bar.
Its currently in general use by the HSFX team so I would guess it works with later versions of Max but I don't really know. I have both this and the Buggybuggy Exporter installed but its a long time since I've used the latter, which I guess says a lot.

Maybe worth contacting Maraz to check on the current status of it?

[SAS~GJE52]

Thanks Ashe,  will look into this 

G

[western0221]

A new question...

Has IL-2 1946 hard-coded about .mat(material) files' file name?

Nagato MOD doesn't use default material "Matt1D0o.mat", "Matt2D0o.mat", "Beam.mat", etc.
But IL-2 1946 wants to read ship's default material "Matt1D0o.mat", "Matt2D0o.mat", "Beam.mat",
and errors (in log file, Can't open Beam.mat, or Can't open Matt1D0o.mat).

I try text search application to find what .msh file where line "Beam.mat" is needed...
But application show me "nowhere Beam.mat calling exists".
Why?

[asheshouse]

As far as I am aware it is not hard coded.

If you get the error messages that you refer to then the material name must be in the [Materials] section of one of your msh files.

To get a "clean" export of materials from Max I apply a single Multi/Sub-Object material to the whole model. Each separate tga texture has a different Material ID (CaioDulio has 10 Material ID's)



For your testing look in all the msh files and make sure you have a tga and a mat file for every material listed in the [Materials] section.    Check that the file reference in the mat file refers to the correct tga file.

While we have the image of Caio Duilio's bridge structure, how does the Nagato bridge structure mesh compare?  Caio Duilio's has 1242 faces so Nagato, with a more complex bridge, should maybe have about 4000 faces.

[western0221]

Now Nagato main guns fire!


Bullets arrives at half distance to an enemy....
I think this is caused by Startshell Hook vector...
I will ckeck this again.

Other AAA/guns exist funny places , this is Hier.him positions are incorrect.

So I did today...

• I divided parts that have many meshes to small. Now all parts have under 3000 meshes.
• Default .mat files (I think not-used DUMMY) place in IJNNagatoBB42 folder. Then Can't open error disapper. It's a mystery to me.

I think at .mat problem...
Evenif I miss material export from Max to .mat,
any .msh has the .mat's filename in text in own .msh [Materials] lines.
I try text auto find/search application around all .msh and all .mat , but application shows "No line in any .msh".

[asheshouse]

For testing main guns place target at 20km so that the gun can be seen to elevate.
If gun elevates correctly and gun flash appears to be in the correct direction then adjust the rotation of startshell in *.msh.

If the gun flash appears correct then roty and rotz are probably correct.
Try rotating rotx 90deg 180deg and 270deg to see the effect.

With target at 20km you are not likely to achieve a hit but the shell splash should be close to the target.
Its easier to see the effect of adjustments with a distant target.

Before starting these tests check the pivot point alignment in Max.

Pivot points must be located on rotation axis.

For turrets z-axis is vertical pointing upwards, x axis points in the direction that the turret faces.

For guns x axis points in the direction the gun faces, z axis is horizontal, y axis is vertical pointing downwards.

These must be correct before you export the mesh.

 

[western0221]

Today I try myself Gun's .msh rotate 90 deg.
I looked at default well-running Head and Gun vector in Hier.him, copyed it to MOD's Hier.him.

Now Nagato Guns seem to reach well to enemy.
But I don't confident that is correct.

In 3D model here.


Hier.him is here.

Code: [Select]

[Head1]
Mesh Head1
Parent _ROOT_
Attaching 1 0 0 0 1 0 0 0 1 62.52 0 8.78
[Gun1]
Mesh Gun1
Parent Head1
Attaching 1 0 0 0 0 -1 0 1 0 -0.2 0 0

When I load Daidalos default 3D model of Gun, it's not rotated.
Even Hier.him Attaching 1 0 0 0 0 -1 0 1 0
Why? But it runs well. I don't understand.

[asheshouse]

You are using Buggybuggy Exporter I think, but I am using Maraz Exporter.
They work differently.

Maraz Exporter works out the msh coordinates relative to the pivot point axes.

Buggybuggy, I think, works out the msh coordinates relative to the real world axes.

So with Buggybuggy, before you export, you must move the object so that the pivot axes and world axes are aligned. This means gun barrels have to be rotated 90deg to line up the axes.

In the hier.him they are rotated back again.

Using MarazExporter you do not have to move the parts before export as long as the pivot axes are correctly positioned. This is one feature that makes MarazExporter better. Another is that it correctly exports smoothing groups from Max.

I hope you can understand my poor description.

In all my models I have the bow of the warship pointing in the +ve x axis direction.
So in hier.him  the rotation for the hull parts is       1 0 0   0 1 0   0 0 1
If your model is not aligned with the x axis then your hier.him rotations will be different.

In your example:

[Head1]
Mesh Head1
Parent _ROOT_
Attaching 1 0 0 0 1 0 0 0 1 62.52 0 8.78
[Gun1]
Mesh Gun1
Parent Head1
Attaching 1 0 0 0 0 -1 0 1 0 -0.2 0 0

Head1 is a turret facing forward.   --- so no rotation needed.
Gun1 is rotated 90deg rotx   ----  to make the pivot z axis horizontal

Ashe

[Sakai07]

Learning a ton watching your dialogue guys, keep it up!!!   The only thing I am interested in learning how to build is ships, so maybe after the completion of this project I can make an attempt on my own. The game is critically short of cruisers from all nations, and the Americans and Japanese come to mind first. A Tone class Heavy Cruiser would be a nice addition, as would the Atlanta Class Anti Aircraft Cruiser (wow, talk about EPIC flak in game, lol) as well and the total absence of ANY light cruisers at all leave a gap needing to be filled. I hope I learn enough here to give me the confidence I need.

@ Western - Thank you so much for your work. A Japanese light carrier was a wonderful treat, and now a real live Japanese Battleship!!!! Thank you!!!

@ Ashe - Your reputation for ship building is renown, and the results are always of the highest quality. You have done more then any one person to make my FMB sessions the most fun I have with the game. Thank you for all your hard work!!!