I'm a Director of R&D at SteelSeries and I wanted to confirm that there is indeed a fix coming out for this CPI issue mentioned frequently in the forums.
It took a very long time to reproduce this error. However - we finally figured out the cause. The current SROM from the sensor manufacturer has an issue. We've worked with them to troubleshoot and reproduce the error and they have delivered a fix to us. We are still testing it internally and with our partners before we release it, but it will be part of the SteelSeries Engine 3.2.1 release, as mentioned above. Targeting end of this week to make it public.
We _do_ read these forums all the time, so seeing your issues reported have helped us troubleshoot the issue. As stated in the forum, if you have an issue, always open a ticket. It will get to us in R&D. Thanks to the couple of you that did open tickets.
To clarify some things: yes - the issue was introduced in the firmware update released with SteelSeries Engine 3.1.5. That firmware update contained a new SROM from the sensor manufacturer. The CPI jump issue has to do with code in the SROM of the sensor responding to changing surfaces on lift off/on a surface. For instance, when using a QCK+ pad (which is very thick) if you slide the mouse off to the edge of the pad, the sensor recognizes this as a surface change and it will modify its parameters. In a very small percentage of the time, a bug would cause the small "CPI jump".
I will update the forum when we release SteelSeries Engine 3.2.1.