OK.
In summary. This thread was started by a number or posters who had issues with the RS01 GT3 in one specific place - the hairpin at Suzuka. Having found this issue too, I found this thread hoping to find a solution. Sadly there wasn't one so I embarked on some testing myself and used input from various friends and posters to help zero in on the issue.
Test Conditions:-
BoP On - Issue seen in daily races and FIA event
Default settings for "Expert" (I have switched ASM on/off when exploring the issue)
I have tested this issue repeatedly driving back and forth around the hairpin using various approaches in terms of line, speed, gear, throttle, brake and steering applications. I have incremented and turned on and off the various aids available as part of this.
Video of issue being seen:
Notes from video:
This is one example of many I could use, the issue is similar under various conditions.
In this example I approach at race speed using normal braking points and lines. As you can see from the video I pick up the throttle early with a small application and the car spins.
You can see on the video that even a light throttle application causes TCS to cut in mid corner after which point the whole throttle bar flashes red.
Findings:
The issue is real and I can repeat it at will.
Following testing I can also switch it on and off.
The issue can be contained by use of ASM but at the cost of overall lap time.
The issue is exacerbated by tyre wear which would have an obvious effect on traction, grip and in-game physics and TCS.
My thoughts on this:
I believe there is an issue with the RS01 GT3, specifically related to the interaction between it and the driver aid TCS which, when set at Level 2 or above, cuts the power leaving the car susceptible to spinning due to the basic physics model and/or the diff setting. Once TCS is triggered releasing the throttle, accelerating or braking will have the effect - the rear end breaks away.
Given that it only happens in this specific hairpin the issue may also be triggered by the camber profile of this corner.
Further testing:
To validate this theory, I have repeatedly run with the TCS set at 1 or 0 (All other settings per "Expert default" i.e. no ASM) and found the car takes the corner more consistently. See video below where I deliberately throw the car in and am sharper on the throttle to invoke the issue and it does not occur. TCS cuts in, but the lower setting isn't so intrusive as to cut the power to trigger the problem.
Call it a bug, a characteristic, an issue, a glitch whatever - I personally think it shouldn't happen and there's a code issue which causes this to happen and could be fixed.