Difference between revisions of "Mesh/Upload Model UI reference"

From Second Life Wiki
Jump to navigation Jump to search
 
(16 intermediate revisions by 3 users not shown)
Line 1: Line 1:
{{Multi-lang}}<noinclude>{{Navbox/Mesh|advanced}}</noinclude>
Use the '''Upload Model''' window to upload COLLADA (.dae) files to Second Life.  This window includes three tabs and preview section that is always visible.
== Preview ==
== Preview ==
The Preview area displays a preview of the selected model at various levels of detail.  Use the {{Vh|Orbit Zoom Pan|standard camera controls}} to pan around and zoom your view of the model.
{{UIref|Name|Name of the model being uploaded.}}
{{UIref|Level of Detail|Select a level of detail at which to preview the model.}}
{{UIref| {{Icon_Gear|Gear icon}} |
* '''Show edges:''' Shows the edges of your model, between vertices.
* '''Show physics:''' Shows the physics representation of your model as either a yellow wireframe mesh or a series of colored hulls.
* '''Show textures:''' Shows any textures that are applied to the model.
* '''Show skin weight:''' If the model contains skin weight information, this shows a preview of what the model looks like when worn by an avatar in the default standing animation.
* '''Show joint positions:''' Shows the position of custom joint locations, if any are included in the model.}}
=== Weights and fees ===
Displays  the model's calculated download, physics and server costs, its [[Mesh/Prim equivalence|prim equivalent]] cost, and the resulting upload fee.


== Level of Detail ==
== Level of Detail ==
 
{{:Viewerhelp:Upload_Model_-_Level_of_Detail}}
=== Select Level of Detail ===
 
=== Mesh ===
 
==== Load from file ====
You may manually upload your own models for each level of detail.  This allows you to maintain tight control over how your model looks at each LOD, preserving key visual features that the automated algorithms may otherwise distort.
 
==== Auto generate ====
Generates a new model at the currently-previewed level of detail, based upon the model you uploaded.
 
===== Method =====
Choose a complexity level for your object, based upon a maximum number of triangles or a maximum number of... {{red|errors???}}:
 
* '''Triangle limit:''' Enter the maximum number of triangles for the model at the selected level of detail.  This number must not exceed the number of triangles in the LOD above it.
* '''Error threshold:''' {{red|How does this work?}}
 
===== Build Operator =====
Selects the method used for incrementally reducing the geometry while building the multiresolution hierarchy. {{red|Need a higher-level explanation with less jargon.  I ''think'' this means, "Selects the method by which the number of vertices in your model is reduced."}}
 
* '''Edge Collapse:''' Merges two vertices currently connected by an edge. Both vertices are replace by a new, unified vertex, which is chosen to minimize error.
* '''Half Edge Collapse:''' Merges two vertices currently connected by an edge. One of the two vertices is merged into the other, which remains (essentially) unchanged.
 
{{red|Under what circumstances is one of these better than the other?}}
 
===== Queue Mode =====
Changes the order in which vertices are chosen to be collapsed. {{red|(Does "collapsed" mean "merged", as above?)}}.
 
* '''Greedy:''' Recalculates the priority values for each neighboring edge after each simplification operation.
* '''Lazy:''' Marks the priority values of each neighboring edge as "dirty" after each simplification operation. Recomputation of costs is only done on required edges. {{red|What does "dirty" mean?}}
* '''Independent:''' This method performs non-overlapping sets of operations on the input geometry, resulting in hierarchies of logarithmic height. {{red|Need a higher-level explanation with less jargon.}}
 
{{red|Why would I choose one of these over the others?}}
 
===== Border Mode =====
Changes the handling of vertices on a geometric border; these are vertices without a full ring of triangles around them, like the edges of a flat plane {{red|Is this example correct?}}.
 
* '''Unlock:''' The default behavior, does not treat geometric border vertices specially.
* '''Lock:''' Prevents any modification to or removal of border vertices.  This is often useful when preserving the edges of models that are meant to align with each other, such as blocks of terrain.
 
===== Share Tolerance =====
This floating point parameter is intended to compensate for small floating point errors present in your data set. It controls the amount of distance between two vertices before they are considered coincident. Try increasing this number if cracks appear in your object during simplification.
 
==== None ====
You may choose not to include a model for any level of detail below High.  The model will not appear in Second Life when rendered at a level of detail set to None.  This setting may be useful for improving performance and decreasing resource cost for small, visually complex models that do not need to be viewed from far away.
 
=== Crease Angle ===
The Crease Angle adjusts the smoothness of the model.  Increase this value to soften the hard, "faceted" look of models with relatively few vertices.


== Physics ==
== Physics ==
The physics decomposition step analyzes the model and generates a series of "hulls" that represent the physical shape of the model as it will be interpreted by Second Life's physics engine. Additional steps allow you to simplify the physics shape, thereby reducing the amount of load for the physics engine.  Models meant to be used as vehicles must not have a physics cost higher than 32.
{{:Viewerhelp:Upload_Model_-_Physics}}
 
=== Step 1: Analysis ===
 
==== Method ====
Each of the three available decomposition methods is best suited to a particular type of shape, and will return the most accurate and least costly results when paired with a model of that type.
* '''Surface:''' Works best on curved, organic shapes, such as an animal body.
* '''Solid:''' Works best on objects with sharp angles and a clear inside and outside, like buildings.
* '''Wrap:''' Works best on very complex shapes, such as trees and branches.
 
==== Quality ====
These values determine how much combining and overlap reduction is performed on the hulls at the decomposition stage.  '''High''' returns the best simplification results, but requires the most time to process.  If you experience performance problems during this step, try using the '''Preview''' setting first in order to confirm your smoothing and '''Close Holes''' results, then switch to '''High''' to get the best simplification results.
 
==== Smooth ====
The Smooth value sets the tolerance when the utility tries to merge coplanar triangles. It refers to the angle between adjacent triangles for which merging is permitted. A value of zero allows some angle between triangles, resulting in gently curving surfaces being flattened.
 
A lower '''Smooth''' setting results in a lower physics cost for objects with curved surfaces.
 
==== Close Holes (slow) ====
Attempts to fill any holes in the physical representation of the model in order to fix minor errors in the mesh.  However, checking this setting also closes intentional openings in objects and should not be used for models that are meant to be passed through, such as door frames.
 
=== Step 2: Simplification ===
 
==== Method ====
* '''Retain %:''' Select a percentage of the hulls to retain from the original analysis.
* '''Detail:''' Reduces the physics model to a single hull that resembles the basic shape of the model. {{red|Is this correct?}}
* '''Better Detail:''' Similar to Detail, but does a better job preserving openings in your model.
 
==== Passes ====
Increasing the number of passes creates results with fewer overlapping hulls and more accurate overall representation of your model's physical shape.
 
==== Detail Scale ====
Reducing this number preserves more fine detail in your model during the simplification process. Larger values result in a decreased physics cost, but lower accuracy in the physics model.
 
=== Preview Spread ===
Move the Preview Spread slider to explode your view of the hulls that comprise your physical model.  On objects with many hulls, this can help you to find and examine hulls that would not otherwise be entirely visible.


== Modifiers ==
== Modifiers ==
 
{{:Viewerhelp:Upload_Model_-_Modifiers}}
=== Scale ===
Increases or decreases the base size of your model.  Increasing the scale of a model can result in an increase in resource and upload costs.
 
=== Include ===
* '''Textures:''' Includes the model's textures in the upload, and automatically applies them to the model.  Each texture increases the upload fee by L$10.
* '''Skin weight:''' Includes the model's skin weight information for the purposes of avatar rigging.
* '''Joint positions:''' Includes the model's joint positioning information for the purposes of avatar rigging.
 
=== Pelvis Z Offset ===
Changes the vertical offset of a rigged mesh when it is worn on your avatar.  This parameter can be used to help properly align meshes that are not meant to be centered on an avatar's root location (pelvis).

Latest revision as of 11:10, 15 November 2011

Use the Upload Model window to upload COLLADA (.dae) files to Second Life. This window includes three tabs and preview section that is always visible.

Preview

The Preview area displays a preview of the selected model at various levels of detail. Use the standard camera controls to pan around and zoom your view of the model.

Name   Name of the model being uploaded.
Level of Detail   Select a level of detail at which to preview the model.
Gear icon Gear icon  
  • Show edges: Shows the edges of your model, between vertices.
  • Show physics: Shows the physics representation of your model as either a yellow wireframe mesh or a series of colored hulls.
  • Show textures: Shows any textures that are applied to the model.
  • Show skin weight: If the model contains skin weight information, this shows a preview of what the model looks like when worn by an avatar in the default standing animation.
  • Show joint positions: Shows the position of custom joint locations, if any are included in the model.

Weights and fees

Displays the model's calculated download, physics and server costs, its prim equivalent cost, and the resulting upload fee.

Level of Detail

Level of Detail > Physics > Modifiers > Calculate weights & fees > Upload

This tab is the first step in the model upload process, in which you specify levels of detail for your model. Levels of detail (LOD) determine how the model looks at various distances. As you get farther away from a model, it renders in less detail to boost performance.

Preview your model's automatically generated levels of detail on the Level of Detail tab by clicking High, Medium, Low, or Lowest. Notice how the object's complexity is reduced at each level.

After you make selections in this tab, click the Physics tab.

Select Level of Detail

Click High, Medium, Low, or Lowest to preview and modify your model's various levels of detail (LOD). The status indicator color will show the whether your model is suitable for upload:

  • Green - No problems detected. Ready to upload.
  • Red - Too many vertices at this level of detail. Models at each level must have fewer vertices than the level above them.

Source

Select from the Source dropdown how you will specify your model's LOD:

  • Load from file
  • Auto-generate
  • Use LoD above


Load from file: Manually upload .dae files for models at any level of detail. This enables you to control how your model looks at each level of detail, preserving key visual features that the automated algorithms may otherwise distort.
Auto generate: Generate a new model at the currently-previewed level of detail, based upon the model you uploaded.
Method   Choose a complexity level for your object, based upon a maximum number of triangles or a maximum amount of overall change in your model's shape:
  • Triangle limit: Enter the maximum number of triangles for the model at the selected level of detail. This number must not exceed the number of triangles in the LOD above it.
  • Error threshold: Increase this value to decrease the amount of detail retained during simplification. For example, setting this value to 0.1 results in a model that is simplified to 90% of the original model's accuracy.


Use LoD above: Copy the next higher level of detail. If you set Medium, Low, and Lowest to this setting, only the model for High will be used.


Generate Normals: Provides options for generating surface normals on your model, affecting its shading and appearance of "hardness" along edges between vertices.

Crease Angle  

Adjusts the smoothness of the model. Increase this value to soften the hard, "faceted" look of models with relatively few vertices.


Physics

KBcaution.png Important: This article has been translated. The translation was imported on Aug 19 2011. Any changes will require re-export for incremental translation.

Level of Detail > Physics > Modifiers > Calculate weights & fees > Upload

This tab is the second step in the model upload process.

A model's physics shape defines how it collides with other objects, and is typically simpler than its visual appearance to reduce computational cost. For example, a complex vehicle model might have a square physical box shape. NOTE: Models meant to be used as vehicles must not have a physics weight higher than 32.

Step 2: Analyze and Step 3: Simplify are not mandatory for creating a physics shape for your model, but can significantly decrease your model's physics weight.

After you make selections in this tab, click the Modifiers tab.

Step 1: Level of Detail

Choose one...   Select a level of detail to automatically generate a physics model based on an existing visual level of detail:
  • Lowest
  • Low
  • Medium
  • High
  • From file (see below)

Lowest or Low are adequate for most purposes and incur a relatively small physics cost.

From file   Select this option to upload your own physical model from a .dae file. This gives you the greatest control over the model's physics shape. Click Browse to select a file.

Step 2: Analyze

Select options in this section, then click Analyze.

Method   Select a decomposition method for the model. Each method is best suited to a particular type of shape, because it will return the most accurate and least costly results for models of that type:
  • Surface - best for curved, organic shapes, such as an animal body.
  • Solid - best for objects with sharp angles and a clear inside and outside, like buildings.
  • Wrap - best for very complex shapes, such as trees and branches.
Quality   Select how much combining and overlap reduction is performed on the hulls at the decomposition stage. High returns the best simplification results, but requires the most time to process. If you experience performance problems, try Preview setting first to confirm your smoothing and select Close Holes, then switch to High to get the best simplification results.
Smooth   Set the tolerance to which coplanar triangles are merged. The value is the angle between adjacent triangles for which merging is permitted. A value of zero allows some angle between triangles, resulting in gently-curving surfaces being flattened.

A lower Smooth setting results in a lower physics cost for objects with curved surfaces.

Close Holes (slow)   Select to fill any holes in the physical representation of the model (if possible) to fix minor errors in the model. Selecting this option will also close intentional openings in objects: do not use for models that have intentional openings, such as door frames.

Step 3: Simplify

Select options in this section, then click Simplify.

Method   Select a simplification method:
  • Retain %: Select a percentage of the hulls to retain from the original analysis.
  • Detail: Reduces the complexity of the physics model based on the Detail Scale slider.
  • Better Detail: Similar to Detail, but does a better job preserving openings in your model.
Passes   Increase this value to create results with fewer overlapping hulls and more accurate overall representation of your model's physical shape.
Detail Scale   Reduce this number to preserve more fine detail in your model during the simplification process. Larger values result in a decreased physics cost, but lower accuracy in the physics model. For many models, most of the simplification takes place at very low values, between 0.0 and 0.3, after which the physics model is reduced to a single hull.
Preview Spread   Move the Preview Spread slider (below the Preview pane) to explode your view of the hulls that comprise your physical model. Make sure the Physics box below the Preview pane is also checked. On objects with many hulls, this can help you to find and examine hulls that would not otherwise be entirely visible.

Modifiers

KBcaution.png Important: This article has been translated. The translation was imported on . Any changes will require re-export for incremental translation.

Level of Detail > Physics > Modifiers > Calculate weights & fees > Upload

This tab is the third step in the model upload process, in which you specify additional features on your model, such as textures.

When you have specified all the information you wish to on this tab:

  1. Click Calculate weights & fee, then review the resource weights, prim equivalent cost, and Linden dollar cost of your model.
  2. Click Upload to upload your model to Second Life.


Scale   Increases or decreases the base size of your model. Increasing the scale of a model can result in an increase in resource and upload costs.
Dimensions   Shows the model's dimensions with the specified scale.
Include textures   Select to include the model's textures in the upload, and automatically apply them to the model. Each texture increases the upload fee by L$10. Uploaded textures also appear in your inventory inside a new folder in your Textures folder. The new folder has the same name as the uploaded model.


For avatar models only:

Skin weight   Select to include the model's skin weight information for the purposes of avatar rigging.
Joint positions   Select to include the model's joint positioning information for the purposes of avatar rigging.
Pelvis Z Offset   Set to change the vertical offset of a rigged mesh when it is worn on an avatar. Use this parameter to help properly align meshes that are not meant to be centered on an avatar's root location (pelvis).
Retrieved from "https://wiki.secondlife.com/w/index.php?title=Mesh/Upload_Model_UI_reference&oldid=1157894"