Synopsys
From CVL Wiki
Line 7: | Line 7: | ||
==Installing Synopsys on Windows (2000/XP/Vista)== | ==Installing Synopsys on Windows (2000/XP/Vista)== | ||
− | ''Step 1'' Double click on My Computer and type in the address space \\filebox.ece.vt.edu\saber. To gain access to this folder you must type in your ECE username and password. (Sometimes, the \\filebox.ece.vt.edu\saber will not work, if it doesn't work, try \\128.173.88.43\SPB i.e. try the IP address of the machine.) | + | <br>'''Step 1''' Double click on My Computer and type in the address space \\filebox.ece.vt.edu\saber. To gain access to this folder you must type in your ECE username and password. (Sometimes, the \\filebox.ece.vt.edu\saber will not work, if it doesn't work, try \\128.173.88.43\SPB i.e. try the IP address of the machine.) |
− | + | <br>'''Step 2''' | |
== VERA == | == VERA == | ||
Revision as of 16:16, 14 February 2008
This is the begining of Synopsys documentation -- so far very incomplete.
The following packages are installed on the CVL & CESCA machines.
Contents |
Installing Synopsys on Windows (2000/XP/Vista)
Step 1 Double click on My Computer and type in the address space \\filebox.ece.vt.edu\saber. To gain access to this folder you must type in your ECE username and password. (Sometimes, the \\filebox.ece.vt.edu\saber will not work, if it doesn't work, try \\128.173.88.43\SPB i.e. try the IP address of the machine.)
Step 2
VERA
What does this package do?
How to run it
To run VERA, in a command window, type:
VERA
(Note, thats all capital letters.) Your command prompt will change to signify that you are ready to run VERA. Now type
vera <file>.vr
(Lower case this time.) For some help, try:
vera -h
There is some documentation in this directory (type the following):
cd $VERA_HOME/doc ls
Medici
What does this package do?
How to run it
To run Medici, in a command window, type:
Medici
(Note the capital letter 'M'.) Your command prompt will change to signify that you are ready to run Medici. Now type
medici
(Lower case 'm' this time.)
TCAD Sentaurus
What does this package do?
Sentaurus is a sophisticated software package that can simulate semiconductor device processing steps as well as semiconductor device parameters with a given set of biasing conditions. Sentaurus can simulate electrical, optical, and thermal properties of a device quite accurately.
How to run it
To run TCAD Sentaurus, in a command window, type
Sentaurus
Now you can run any number of TCAD Sentaurus programs. (This documentor does not know the names of them...)
There is some documentation in this directory (type the following):
cd $STROOT/tcad/current ls
Look under any of Example_Library / Manual_Examples / manuals
Sentaurus Training
The training documentation provided is quite good although very time intensive. It is recommended that exercises contained in the training documentation be completed. Furthermore, it would be very beneficial to look at the core files of a sample simulation and try to replicate it after completing the training documents. Lastly, although the training documentation is good, the order in which the training modules are presented is not optimal. Completing the training modules in the following sequence will make understanding the tool easier:
-Tool Overview -Ligament -Sentaurus Process -Sentaurus Structure Editor -Sentaurus Devcice -Inspect -Tecplot SV -Sentaurus Workbench -TCL -Noffset3d
Note: Although the total estimated training time is roughly 24 hours, this time can be reduced by skimming sections of little importance. In addition, some tools are quite intuitive and therefore require very little training. Do not be intimidated.
Using Sentaurus
When using Sentaurus, always check material parameters; do not assume the default parameters are correct. Finally, finding the correct mesh size for a given simulation takes both experience as well as trial and error. If the mesh size is too fine, then the simulation may not only take incredibly long, but it may never converge and give false results. On the other hand, if the mesh size is too sparse, the simulator may again produce inaccurate results. In cases where multiple device characteristics are being tested, it can be easier to use two different mesh sizes depending on the biasing conditions.