UPDATE: I had some time today to reinstall both Callaway Corvette and Sebring. I drove 10 laps with the exact same setup that seemed to cause the issue. So far, so good. I wonder what it was, but since nobody else ever reported about the problem on Sebring I'm assuming it had to be something on my end (corrupted file?). Anyway, I've changed the title to solved. -------------------------------------------------------------------------------------------------------------------------------- I stumbled across the issue today while driving Callaway Corvette on official Sebring. I don't have the recording software, so I just took some screenshots in succession. On one of the corners car dives in and is sent into the air. Happens every time. Steps to reproduce: can be a stock setup, but change the ride height to 5.5cm front, 7.5cm rear This happened during practice session, don't know if it matters. I didn't mess further with the ride heights, but at stock 5.0 front 7.0 rear it never happens. At this point I have no idea if it just happens with the Vette at that particular ride height. I don't remember it ever happening with any other car, but to be honest first couple of laps when I was running in the pack I thought I was just bumped by car behind. I didn't have more time to test different cars yet, or different ride heights. Nevertheless, 5.5 and 7.5 seems pretty good for a Callaway here with my setup. I loose easily 1s each lap because of this, that is if I don't loose control of the car, which can easily happen going into the corner with all wheels off the ground This is very sudden and violent, and if one goes full speed it's very easy to loose the car @Christopher Elliott please let me know if you need anything else from me, will try to help if you cannot reproduce Different lap Another, this one a bit less severe, still violent though
did I read that right? your problem began after you raised the ride height?(I would have thought it would have been the other way around)
Correct. However, this has nothing to do with bumpy track. The track is actually smooth in that section. Look how the car suddenly "sinks in tarmac", and then is sent off flying. I've never seen this before until today. Only happens with the Callaway at that certain ride height. I wonder if anyone can replicate this. My setup when it happened was far from stock, but I noticed it happened when I raised the height to 5.5 and 7.5. Then I tested with stock setup, just the ride height adjusted and possibly soft tires and it happened again. When I reset to stock height all was back to normal
might be a slight mis-match of something there. We saw several cars have this kind of issue on 3rd party tracks last spring or summer. Some of them would get launched waaaay up in the air. 15 to 20 feet up. I'm guessing your increased ride height allowed the tire/car to sink further than intended. Most folks start at the default ride height and begin lowering the car when building setups. You might want to repeat this post in the bug report thread.
The sinking tires happen at a few other tracks too. When I opened rF2 about an hour ago, I got a newer version, 1.01. Not sure if its related, but I was hoping they had corrected the extreme surface mismatches but its still a MotoX track.
THAT 'IS' Sebring. If you want smooth, choose Cota or any other Tilke designed track. Sebring is bouncy, Sebring is extremely hard on equipment, but that pothole AMI found was a little beyond even Sebring's legendary roughness.
While I've been away from racing IMSA since 1991, I still do a bit of evaluation for two active teams. Spent the weekend reviewing MoTec data from Feb 21-22 testing from the Nissan, thats a pretty fresh snapshot. The rF2 track is pretty similar in locations but has as much as 10X the acceleration rates of suspension motion in several areas. I have a hunch I know why (to get an accurate resolution of waypoints would be a >10GB file). In any event the severity of the bumps is so far off from the real track I don't bother. That's too bad because otherwise its one of the best in terms of renderings.
Not to worry, indeed its an abusive track, probably the best example of of why laser scans don't always result in the best track models without some labor intensive adjustments to the files. If the real track was as harsh as the S397 model it would destroy most prototype tubs in a matter of a few hours at the ride heights they typically run. We ran at 3.25" with stiff springs in the Spice, 2-2.5" everywhere else. Medium compression, stiff rebound on the dampers, low ARB rear. Made power off the exits of 7 & 5 a bit more challenging so we geared taller in 1-2 gears to preserve rear tires. Turn 16 is pretty accurate if you follow the line taken by the AI but on the track most tend to go a bit deeper to late apex; its still two turns from 15, but in those days they put cones up to force us into 15 a bit earlier, and after about 10 laps the cones were gone.
Can you provide a source that states laser scanning is not an adequate method for modeling tracks? And to be specific, the laser scans that are available today are done by the track company, so it's not like Studio 397 went there with their own equipment. The same scan and point cloud data is available for everyone who licenses the track, so presumably it is of best possible quality. In fact, laser scanning provides so many points in the 3D space that not nearly all of those can be used in the actual game. If the games were to use the full point cloud, the physics would not run in real time, hence the precision is somewhat reduced. However, even the slightly reduced road surface should still represent the bumps in their correct proportions.
Laserscanning is one thing, but the making a workable mesh of it is still 1000 x less resolution that the real world. So you need to pick what bump to show what not, most laserscanned track end up with a mesh size of about 0.5 to 1 meter. Lots of detail still hiding in those numbers. But if i look at the screenshot above is seems there is a issue with the mesh. That's nothing to with the regular bumpyness of the track.
Guys, a little update: I've edited the original post. After reinstalling both Sebring and Corvette issue seems to be gone.
