Here is a result of the cross correlated analysis using the data from the last night (3 hours, starting at 6:40:00 UTC, BW = 0.1 Hz) (30780).
Also, for comparison, I overlaid the DCPD spectra from Sep. 26th (30115) where we were operating at 50 W and this was before all the TCS and ASC revert. The optical gains for the two data sets in [amps/meters], according to the Pcal height at 331.9 Hz, are almost same with the latest lock lower by 14 %. Therefore any difference in the DCPD current can be almost directly interpreted as change in the DARM displacement. I did not check the cavity pole locations.
Overall, the latest lock appeared to have lower noise floor in the jitter band (200 Hz - 1 kHz) by a factor of 2- 5. Also, the latest noise exposed a number of peaks that were not easily to see in the September data. There were two small frequency regions where the noise level became worse, namely 365 Hz and 1084 Hz. These two peaks were not present in the September data. Additionally, the laser noise contribution above 3 kHz improved. The fig file is attached too.