On old Brit iron for a smooth motor the first place you started was the cam - the factories couldn't seem to consistently get any better than +- 5 degrees at best, most aftermarket cams lucky if they were twice that, with one or two exceptions, Piper and Crane were reputed to sometimes get it right.
There was usually 60 thou of case hardening on the cam so a little work with the proper hones and you could get the a bit closer, and it wasnt hard to find offset keys, etc, to get the timing closer than the nearest tooth - at the ragged edge on or two degrees makes a decent sort of difference. Timing figures were often available as lift at TDC, which removed a few other variables, and was sometimes different for individual cylinders to allow for camshaft twist. Made a huge difference to the Commando ,in particular.
Never seen any figures for BMW factory cams, but when they changed their supplier Richie was the only guy who noticed that they were way out, which suggests that he pays close attention to his cams, and of course he, and most other top tuners sell premium cams which I think we can assume are as near to spot on as manufacturing tolerances allow, and if my experience with Brit bikes is anything to go buy can make a big difference.
It also used to be standard practice to measure and balance the combustion chamber volumes, and where the equipment existed, to balance the port flow, and to " blue print " every other part of the motor, including selective assembly of all bearings.
This seems to have gone out of fashion these days in favor of buying the most exotic and expensive components and processes you can find / post pictures of, and as you cant post pictures of hard graft it doesn't seem to happen, outside of the select few who produce winning machines.
The original tale of the 800 cc bikes being the optimum capacity was related to the thermodynamics and was based on the ratio of the total surface area of the cylinder and chamber to the volume, the 800's having the most favorable ratio.
How much the rocking couple contributes to the perceived vibration on I have no way of assessing, but I read somewhere it was very little, and in any event was much less than a slight variation in main bearing alignment / tolerances or "fly wheel " run out.
The guys who have spent a lot of time and money working out what works best justifiably keep their cards close to their chests, but in my experience simply taking the time and effort to make certain that everything you can reach from the outside of the motor is performing exactly as intended will go a long way towards a smooth running motor, whatever the capacity.
