Check out https://www.pianoteq.com/ as an example. Granted, there are situations where this simulation doesn't quite come as close as a sample based solution but in other situations it is more realistic.
My point being: Even if real time solutions are technically not as detailed as a look-up-table based solution, it's benefits outweigh the negatives due to it's flexibility. For instance, a look-up-table solution is a real nightmare if you want to have a properly simulated real-time weather system that is as detailed as we have now in pC2 where puddles and dirt/mud are happening. Just simulating "every possible scenario" that a tire can go through is a monumental task in a look-up-table based solution. ISI conveniently haven't discussed the real downsides of their solution.. luckily Niels Heusinkveld discusses some of these issues on his youtube channel and basically comes to the conclusion that the rF2 tire model system is unnecessary complex considering it's limitations. Then again, he may be just making excuses as Reiza are stuck using the old system (albeit a heavily modified one).
EDIT: oh and one more benefit with the real time model is that SMS get "true" Research & Development information from this and it's going to scale very well with future hardware and software advances. In short, it's just simply smarter to do it this way when you look at it as an investment of R&D.