How does PROC ESM optimize the damping weight when model=damptrend

[cau83]

I am forecasting using PROC UCM and want to do a post-forecast adjustment to dampen the trend in some situations. I am familiar with using damptrend through PROC ESM and I understand that the damping weight is selected by minimizing SSE/RSS. The same thing is done by R's forecast package when dampening is selected. 

I could pre-select a few damping weights (say, 0.98, 0.5, 0.25) and then create adjust slope values for each, calculate SSE, find the minimum and adjust the forecast using the difference b/w the new slope values and what PROC UCM used. I would rather not do this for a couple reasons:

1) I have to pre-select the weights and I am optimizing b/w these options but not considering anything else (and I assume that SAS/R are somehow considering the entire range of values between zero and one)

2) It will be time intensive to run this test - may run 1000s of times depending on number of forecasts and pre-selected weights to test.

How does SAS (and/or R) quickly and broadly do this optimization in the background, and is it something I can approximate? 

[cau83]

http://support.sas.com/documentation/cdl/en/etsug/63348/HTML/default/viewer.htm#etsug_tffordet_sect0...

According to this, The optimization is initialized by choosing from a predetermined grid the initial smoothing weights that result in the smallest sum of squared, one-step-ahead prediction errors. The optimization process is highly dependent on this initialization. 

I guess this would provide more nuance to my question-- I may be better off knowing about this predetermined grid (not necessarily the exact grid used by SAS but how one would come up with these starting values) and then how it optimizes around that.