I'll let chronus chime in on the diff between the commercial and pro versions, but that doesn't change the fact the closer the data is to real life values, the more a simulation must represent reality, which is all we can do, so to some degree your question is kinda pointless, we may as well have no distinction between any racing game if you're going to self define it that way.
Indeed.
Something the other poster said:
"
f a pc is restricted by hardware to produce a real to life simulator with forces, motion, exact data for tyres etc etc.
Arent we then all just playing dumbed down simulators anyway?
So arent we just playing simcade versions of the ones f1 teams/multimillion dollar simulators etc use?"
A motion simulator is not a requirement for the "physical realism" or "accuracy" of a sim. Nor should it be.
If that poster and others think it is, then consider this:
1-The plant at Sindelfingen (used by Mercedes GP), is the most advanced motion platform with 360º motion and screen and a 12 meter long rail for crosswise, horizontal motion...and it cannot deliver sustained high-g's for several seconds. [There were rumours about building a 5 to 7 meter arm based centrifuge for F1 training, but the associated costs were exceedingly large.]
Professional racing motion platforms fall short of the real experience and cannot simulate the actual level of accelerations experienced in real life for duration and magnitude - regardless of the impressive payload figures. In our "cheap" simracing corner, it's even worse.
2- Traditional platforms used by air forces to train pilots can exert substantial "forces" (several years back, max g-force was between 1.0 and 1.5g in any given direction - and not sustained), but again, not delivering (limited) sustained forces higher than 5g's.
Current DFS technology, on the other hand, which uses the good old centrifuges (and somato-sensory perception devices - inflatable suits and seats, plus mechanically actuated harnesses that tighten up), can deliver high sustained forces in 4 axis of motion - if memory serves me well, 14g's are attainable. This is only possible via use of cumbersome and powerful centrifuges, obviously.
3- Motion platforms used by racing teams can (within low limits) simulate impact forces, but they cannot simulate sustained lateral or longitudinal forces that last for seconds. Case in point, with LMP1's (and even more so with F1's) there are circuits with corners which allow/demand the driver-car system to experience "g-forces" from 2.3g, up to a peak of 3.4g and down again to about 2g, and this can last over 2, 3 or even 4 seconds - this cannot be simulated in any non-centrifuge based motion platform.
That's, in a nutshell, why we should leave motion platforms aside when discussing the physical fidelity of our sims. Better to focus on how we model the physical interactions of the contact patch, the suspension elements, the chassis and aerodynamics devices - for the foreseeable future we will not be able to fully or accurately represent the physical sensations of actually driving a real car (definitely, there's nothing like the real thing). Maybe some day, with some "holodeck" type of tech we will, but not now.
As for a pc being "restricted by hardware to produce a real to life simulator [...] exact data for tyres etc etc."...
Not sure what he intended to say in regards to "exact data for tyres", but if he meant to say we cannot use "exact data" then that is not correct.
I have used "exact data" and seen others use "exact data" (i.e., data provided by tire companies testers/engineers, or data from data acq. systems) in different platforms, including old "dumbed down" rF1. The work around "exact data" is indeed overtly complex (collecting it, cleaning it, interpreting it, verifying it and applying it to a set of models), but it is routinely done by performance centres or even inside the teams themselves by a host of "technicians".
The problem, as always, is not whether we can use "exact data" but if at all it is useful within the confines of a simulation platform. It's not uncommon to filter out data considered to be "out of range", it's not uncommon either to discard recent data sets in favour of "old" batches of data if inconsistencies are found. It's not uncommon to employ a variety of noise reduction methods (to deal with low-level errors, often resulting from flaws in the data collection processes) and pick that which provides the best approximation (or the one that hinders data analysis the least).
Errors, uncertainties, flaws, exist in all levels of "simulation", professional or non-professional. There is no "PERFECT" system, no "100% accurate" system, but we can use that which works the best (regardless of price, complexity, supported hardware).
My 2 cents, mate.
msportda said:
another thing if rf2 is so close to real numbers
Define:
1) what "real numbers" are those?
2) what do you mean by "rf2" being close to real numbers? The physics engine (i.e., the physical/mathematical modelling) or the physics calibration used? If you mean the latter (physics calibrations), then by all means, anyone can try their luck with it, change it and see for yourself. Then, and only then, you can comment on both aspects.
Because you really dont know anything about pcars and its tyre model.
