Before I changed the iso o-rings I did a test where I "hard mounted" the gimbal to the Cinestar by tywrapping the iso mounts so there was no movement. That did eliminate the soft twitch I'm seeing now but the hard mounting introduced other issues.
Looking at the whole thing as a systems engineer I think the bottom line is there are too many variables to make a perfect system that works flawlessly on each and every multirotor/gimbal combination that exists. Add to that the fact that servos currently available are not the ideal way to drive a gimbal with the speed and precision required and the best we can hope for is to get to a place where the results are the best that can be achieved with the hardware in use. There will ALWAYS be something unanticipated that affects the system as a whole and it will vary from platform to platform when dealing with systems that are built from parts that were not designed as a unified system but rather for a particular purpose that can be adapted to use with other manufacturers products. In my case the Cinestar isolation mounts are not the ideal mounting method for the weight of my camera and gimbal while using the RSGS, it's really as simple as that. The solution is much harder and usually not something that can be solved by making changes in the firmware, it requires looking at the problem from an engineering perspective and determining the root cause of the issue and implementing a change that fixes or at least minimizes the root cause.
If I was after absolute stability and precision using my NEX 5N I'd take a serious look at a DJI Zenmuse gimbal. Problem there is while it works fantastically well with a very specific camera and setup, it isn't very tolerant of things as simple as swapping to a different lens, it needs to be recalibrated to maintain its level of performance. For the $3500 price tag I would expect a bit more flexibility on what can and cannot be done by the user in the field so in that light I'm willing and able to deal with where my particular setup is at right now, it's far better than what I was dealing with before installing the RSGS and for the minimal cost of the Skyline it's a definite and concrete improvement. No, it's not perfect and to expect that adding a standalone stabilization system is going to make a less than optimal setup perfect is fantasy. A lot of professional and amateur photographers and videographers expect exactly that mainly because they don't really understand the capability of the platform they're using, they're coming into this from the perspective of it being a new tool to use for APV and not from the perspective of a hobbyist or engineering type that knows the platform is far from flawlessly stable and perfectly suited for professional quality APV. In terms everyone can understand multirotors for APV are still in the Ford Model A era, things have a long way to go before the performance levels will be up to Ferrari standards out of the box, in the meantime it's a matter of learning your particular system and how to tune it to the best level of performance it can achieve, and then work within its limitations and capabilities.
Ken
