【资料】【灌水】i-deas...............

Universal Files (*.unv)
Can choose whatever items in the model file--parts, assemblies, draft setups, etc--that the user wishes to include.
Two types:
Simulation Universal file -- can be created from the export form.
Design Universal file -- must use the hidden command (/xto f ea {for export})
Does not maintain associativity between Draft Setup and the drawing.
Used to transfer data between model files that cannot be shared using libraries (FEM, etc.)
Used to transfer data between non-compatible computer systems (UNIX and NT).
Universal files are upwardly compatible from one version to the next version of I-DEAS.
May be imported into an existing model file with other data (parts, assemblies, etc.).

Archive Files  (*.arc)
Writes out the entire model file.
Must be imported into a empty model file.
Used extensively for archiving data (FEM, etc.).
Maintains associativity between Draft Setup and the drawing (as long as both are in the model file!).
Used to bring data from very old versions of the software into the latest versions.
  
ONE FINAL WORD:
.arc files can be renamed to .unv files and imported via the universal file pick.  The import usually goes faster and circumvents the need for the archive reader/writer

Library Data (Package Files) *.pkg
Used extensively to share data across installations.
Maintains associativity between parts, assemblies, draft setups and drawings.
All data (parts, assemblies, etc) to be exported out MUST be checked into a Library.  Data cannot be retrieved from a model file.
Export can be for check out or for reference only.  (Copy is not an option.)
Once a export from one installation has been completed, all items listed in the package file will have a ‘Import/Export’ lock on them.
Always export to a directory specifically for Package File exports (otherwise, the file will go to your current working directory).
Exporting Library data creates a *.pkg file that is ONLY metadata about what is being exported.  Therefore, if all the contents of a Library have been included in the export, along with the .pkg, all the .prt, .asm, etc. files must be sent.  On UNIX a .tar file should be created to gather all these items and the .tar file is then sent to the new installation.

.......
Individual items cannot be imported from a package file.  The package will only import as a whole.
When importing a .pkg file, there are options available for where the data will be saved:
Use Original - I-DEAS will attempt to import the items into projects and libraries that match the ones they came from
Supply Missing -- Any items that have a missing project or library will be place in the project and library that the user supplies.
Override Original -- Creates local copies of items.  ALL associativity will be lost.  All items will be placed in the project and library that the user specifies regardless of whether the original exists.
Create Missing -- Creates any Projects and Libraries that are needed to make the installations ‘match’.
By default, drafting setups and related detail drawings are NOT included when the export of the related part or assembly is done.  Either specifically select the draft setup or look on the Options form.
To check if you have import/export privileges, at the Command Prompt type in:  /ma idm
The information will be displayed in the List window.
Different file types can be used for exporting and importing .pkg files (from UNIX to NT or vice versa).  When the import takes place the difference in the file type will be noticed and a dialog box displayed for data conversion.  This conversion must complete before the .pkg can be imported.

                 Class-1 Modeling Using Surfacer  notes
To introduce the concept of a Class-1 surface model  
To explain the various levels of Class-1 modeling
To provide familiarity with Imageware’s Class-1 tool-set
To ensure that the student is aware of the usual methods used to create surfaces under most modeling scenarios
To provide a firm base for students to build Class-1 expertise
Introduce feature detection from Scan data
Contrast methods for sparse & dense data
Gain familiarity with point cross-section tools & methods for class-1 modeling
Creation of reference cloud geometry

Class-1  Definition
  
Class-1 surface models capture styling intent as well as being geometrically close to the desired shape
The main objective in creating a class-1 model is usually to capture a particular ‘high-light’ or flow of light over the part
This usually involves the creation of surfaces which are least “Curvature continuous” -- i.e. the patch boundaries flow flawlessly into each other

Aims
To introduce the concept of a Class-1 surface model  
To explain the various levels of Class-1 modeling
To provide familiarity with Imageware’s Class-1 tool-set
Introduce feature detection from Scan data
Contrast methods for sparse & dense data
Gain familiarity with point cross-section tools & methods for class-1 modeling
Creation of reference cloud geometry

Curve Based
Create ‘clean’ curves network
Sweep curves, loft or blend to create patches
edit to fine tune quality
Advantages
Forces patch structure to be defined early in process
Good curve network = good surfaces
Disadvantages
Very Time consuming
Sometime hard to define the ‘right’ curve
Surface Based
Create simple patches & edit to data
Stitch surface edges & manage flow of highlights between patches
Edit to fine-tune quality
Advantages
Quick & straightforward approach
Quality of surface is held through editing process
Disadvantages
Requires careful management of patch breakdown before surfacing starts to avoid rework later

Recommended Strategy
Create major slab surfaces using surface editing approach
Transition surfaces created using the most effective method depending upon the specific geometry

Feature Extraction
Curvature Based
Uses point cloud  curvature diagnosis
Neighborhood size variable determines size of features found (and speed of operation)
Calculates a curvature value of each location in the cloud (dense or scan line)
‘Extract’ facility creates a new cloud with only points matching a desired criteria
Sensitive to noisy data
Use the combination of settings in the extraction base to filter out the actual feature lines   (combination of 4 settings)
Use just the first two sliders to isolate larger areas / slabs of data
gives good indication of possible patch structure
  
Notice how each method extracts the data in a slightly different way
In many cases, the level of detail vs. noise in the data will dictate the method which gives the best result