Yield View of the Worst-Case Distance Method
After the Monte Carlo run for the Worst-Case Distance (WCD) method completes, the Yield view of the Results tab shows the yield estimation.
The following table describes the information displayed in different columns of the Results tab.
|
Column
|
Description
|
|
Test
|
Name of the test.
|
|
Name
|
Name of the specification.
|
|
Yield in Sigma
|
The yield value in sigma. This value is calculated using the following formula:
where, erfinv is the inverse error function.
If the yield in sigma is greater than 8.2, the yield in percentage is displayed as 100%.
|
|
Yield in Percentage
|
Displays the yield value in percentage. This value is calculated using the following formula:
where, erf is the error function.
The yield in percentage value is displayed with 10 digits by default. To change the number of digits to be displayed for this value, set the value of the digitsToShowForYieldInPercentage environment variable. You can display a maximum of 53 digits for these values.
|
|
MC Yield
|
The yield value from the Monte Carlo run.
|
|
Target
|
The target to be achieved for the given specification.
|
|
Status
|
The convergence status for each specification.
|
Convergence Criteria and Convergence Statuses
The tool uses the following two convergence criteria in the WCD method:
The following table describes the convergence statuses shown in the Status column of the Yield view in the Results tab.
|
Status
|
Description
|
|
converged after x iterations
|
The yield estimate converged after x iterations.
For example, the yield estimate for the Gain specification converged after 3 iterations.
|
|
skipped because the Monte Carlo yield is less than <value in percentage>
|
Yield estimation was skipped for the specification because the Monte Carlo yield estimate was too low.
The low yield threshold value is specified in the Monte Carlo form in the Skip specs with MC Yield < field, which is by default set to 3 sigma or 99.86%.
|
|
least error WCD, did not converge after x iterations
|
Yield estimation did not converge after the maximum number of iterations has completed. The specification failure boundary is strongly non-linear or the maximum number of iterations is too small.
The yield estimate with the least error among iterations is reported.
|
|
lower boundary, did not converge after x iterations
|
Yield estimation did not converge because the specification has an unrealistic yield estimate which is larger than 12 sigma in yield after the maximum number of iterations.
The yield estimate increased at each iteration, but never converged.
|
|
estimate based on MC data | lower boundary | least error WCD, stopped because evaluating of the WCD point sensitivity failed on iteration x
|
Yield estimation stopped before reaching the maximum number of iterations because of a simulation or measurement error in evaluating the WCD point sensitivity.
The lower boundary is reported if it is identified, if not, the yield estimate with the least error among iterations is reported.
|
|
estimate based on MC data | lower boundary | least error WCD, stopped because evaluation of the WCD point failed on iteration x
|
Yield estimation stopped before reaching the maximum number of iterations because of a simulation or measurement error in evaluating the WCD point.
The lower boundary is reported if it is identified, if not, the yield estimate with the least error among iterations is reported.
If the run was stopped on the first iteration, the estimate based on the Monte Carlo result is reported.
|
Related Topics
The Worst-Case Distance Method
Running the Worst-Case Distance Method
Return to top
