Regarding RT Sebring Bumps (physical track mesh)
I just wanted to post this as I thought it might be a interesting discussion point.
I have been having a interesting discussion today about RT Sebring being 20% too bumpy..it's is claimed by a well known scratch track maker that the original source material (the s397 laser scan data untouched in the conversion) when compared to the lidar (presumably) data is approx 20% higher in elevation.
I'm super interested in all your opinions on this...why would s397 tamper with what must have been a fairly expensive and expansive in accuracy laser scan data set?.. would this not undermine the superior physics engine claims for rfactor2. Do they do this with other laser scanned tracks?
Also on a side note I remember driving the conversion of Nord's from AC to rfactor2 and not thinking it felt underwelmingly bumpy.
Lastly what are your gut feelings on how rtsebring currently feels...to me imo..and comparing footage from real life on board cams..I'd say it's about right
Also last side note.. surely a professional locally scanned data set would represent a better feeling of the surface intracasies than lidar data ever could, regardless of the claimed 20% increase
I look forward to hearing your thoughts on this
Jim
From my perspective, I would tend to trust publicly available data over proprietary data, as a general rule.
Doesn't mean it HAS to be right, but it's very likely, IMO.
As to why s397's data would be incorrect, it could be any number of reasons.
It could be a mistake, or it could be on purpose.
Perhaps it was felt that the "real" data didn't properly convey the bumpiness of the track.
We all know that the current generation of racing physics engines are not perfect 1:1 representations of reality.
Some data need to be "fudged" in order to get a good output result, and some aspects will remain lacking no matter what.
I.E. the correct input won't necessarily result in the correct output.
As a sim-enthusiast, I'm firmly of the belief that it's the output that matters. If it behaves as it's real-life counterpart (or as close as) in the game, it's good, and if it doesn't, it's not. Regardless of how correct the input data is.
Unfortunately, that doesn't really help in knowing what would be the best approach in this case.
I would tentatively suggest that going with the, presumably correct, lower bumpiness would be advisable.
While the physics engines, as mentioned, aren't perfect, they're still robust, and I suspect an overly bumpy track
would affect the car's behavior, and/or how to setup the car for such a track, in an unrealistic manner.
Perhaps.
This is all conjecture, of course, but at least that's my two (slightly belated) cents!