- New gainCorr is the one obtained in iteration at Aug. 14
- gainCorr values in DB vs. new calibration (last iteration)
- w/wo channels NOT converged
- Any gainCorr larger than 4.0 was set to 4.0 to draw them within range

- Convergence criteria used for this iteration:
   a. Use gainCorr obtained in last 4 iterations (ex. use i=0 to i=3, at the end of iteration i=3)
   b. Check if derivative between adjacent iteration (=subtraction) is smaller than '0.01', and then count # of satisfied
   c. If # of satisfied is 3 (ex. derivative btw i=0 and i=1, 1-2, and 2-3 are satisfied at the end of iteration i=3),
       check if derivative between first and last iteration (ex. i=0 and i=3) is smaller than '0.005'
   d. if b and c are satisfied judge the channel is converged and fix the gainCorr

   * Report for criteria: from p8,
      https://drupal.star.bnl.gov/STAR/system/files/CKim_FMScalib_180723_1.pdf