The quality of the mathematical ground model in combination with the motion and visual ques of a simulator is usually not high enough to allow sufficient confidence for use of training crosswind roll outs under slippery conditions. These limitations are often not fully recognised by the instructors and even the suppliers of these simulators.
I disagree. These limitations are fully recognised by both simulator manufacturers and operators.
Whilst the airframe manufacturers (probably) have adequate flight test data for X winds at the edge of the certified limits, they typically only provide data within their data packs at fairly mundane levels.
The archaic regulations that govern simulation don't help either. There is currently nothing driving this industry into the 21st century. Despite massive leaps forward in technology, FFSs are still being accredited to regulations written when banking was thought to be an honorable profession and aircrews weren't prisoners within their own workspace!
Today's visual and motion systems are more than capable of providing the necessary cueing but there is no objective data provided to quantify or measure it against.
This is no different to unusual attitude training, accurate and representative severe engine vibration malfunctions and so forth which repeatedly get raised as FFS requirements. The technology exists, the regulatory requirement and
supplied data does not.
Without both of these, nothing will move forward.
Simulation will never replace the 'real thing' but they are capable of providing so much more training value.