I am having a critical issue with FFB that I started to notice a few months ago. Initially I though the problem was with a mod track that I was working on. As I was busy working on this track I was not driving at all in AMS for a couple of months. When I am in the gravel, and hit the brakes to lock up the wheels, I get a very strong high frequency oscillation in the steering (to the point of damage). If I don't release the brakes my computer will hard lockup so that I need to shut off the power to reboot. In many cases my bios has been reset, some times the system clock was reset. I have been trouble shooting the problem, I have tried every possible controller FFB setup combination... the only thing that works is to turn off FFB. The problem happens with stock cars on stock tracks. I don't think its my hardware because of the very specific in game conditions that trigger the issue, that is in the gravel with the brakes on. There is no problem with wheels locked on the grass, or the road, or any other surface type. I found that the violent oscillations under braking in the gravel can be eliminated by reducing the "Resistance=" value in the track TDF. Standard RZ values are 25000... reducing this to the 5000 - 8000 range seems to prevent the oscillation. Does anyone else have this issue? Hardware is in the signature... I use a TM 500 RS that is a couple of years old now but seems to work perfectly. Cheers I managed to capture a bit of telemetry data the other day while I was trouble shooting this issue. The car is an F309 on a mod track Algarve. The track has no bearing on the result... I can reproduce the problem on stock tracks (Imola is a good one, lot of gravel traps). However on stock tracks I always ended up with a computer lockup. Another car I tested with was the StockV8, I have not tested with every car but I think the track / car combination is not an issue. One more note... The TM wheels have in the controller setup utility a feature to demonstrate various FFB effects. A few of these (Bong and Force Field) produce a similar oscillation to the one I get in the gravel traps under braking. I can work my wheel until I smell the motor heat by repeatedly running the Bong / Force Field tests but I don't end up with a system lockup like I have in AMS. I was just testing a bit more with the F309 at Imola 2016, AMS Beta... I just went wide at high speed into the gravel trap at T6... no brakes on, no violent wheel vibrations, but I got a system freeze and needed a hard reboot, this time with my BIOS reset.
Which track--any stock one or the mod one you are working on? If it's a mod track, you already have your answer. If it's stock, give us a specific track and gravel trap location so we can help test this. Again, if it is every gravel trap, then the rest of us are skidding through gravel all the time (I'll speak for myself ) without lock-ups. Severe oscillation means that your wheel is not configured properly. Check the Smoothing Level in Realfeel.ini for the car(s) you are driving. Or, there are more basic damping settings out of whack.
No not at all. The wheel is not responsible for the signal sent to it by the simulation. You can see that AMS is generating a strange high frequency high energy oscillation when the front wheels are not turning from reading the telemetry log. Forget the system lockup for now. Explain the strange steering arm force... only seen when on the gravel surface.
Still looking for answers to my problem, I stress tested my system with prime95 x64 for a few hours and found no issues with hardware. How could I confirm my controllers. Any ideas?
Is this "Steering Arm Force" also generated when there is no ffb device? If so, maybe you could disconnect your wheel and try to drive with your keyboard and have a look at those values. If it behaves the same you can at least be certain that it has nothing to do with your wheel
Very helpful. I reinstalled directX and the VB runtimes from the redist folder in AMS, reinstalled the TM drivers and the firmware but the lockups continued. I have reset the controller.ini and the realfeal.ini to default values. I added an exception for AMS to my AV (ESET NOD32) just in case but I have never had and AV problem with AMS. If I turn off the FFB effects then I don't get any of the strange oscillations in the steering arm forces when the wheels are not turning on a gravel surface. I set the FFB strength to 80% and I found that my system would not lockup, which could be a work around for this problem as the overall FFB feeling seems ok compared to 100%. That said there is definitely a problem with AMS. Have a look at this telemetry. I simply drive straight into the gravel with the brakes applied. At the higher speeds the FFB is ok and seems natural... as the car slows the weird oscillations begin and increase in intensity frequency as the car come to a stop. Something is wrong here.
Good idea, cheers. I can use my wheel with the FFB turned off and when braking in the gravel there are no strange steering arm forces in the telemetry log. It seems to be an issue with the FFB.
If it's anything like rF2, the gravel and some other off-road areas were not built with precision with the philosophy that you wouldn't be driving there, only wrecking, so don't put a lot of time or attention to an accurate profile. Some of these areas were programmed with simple repeating oscillating shapes to provide crude bumps that probably felt plausible in a Logi G. When you run an accurate wheel over them, you get an accurate reproduction of the crude proxy bumps. Look carefully at what the wheel is driving over. Does it look like accurately modelled gravel, or a systematic pattern filled-in quickly and unrealistically? Remember the olden days? We used to have this for graphics, too. Repeating road and grass textures that would strobe at speed because they were just repetitious geometry. It's my best guess. I am not the expert on this.
Sorry if i did not help much, i answered what i could. Steering arm force is force present on Steering rack. It's only output to ffb to "drive"wheel. Another thing, that vibrations you recorded as i can tell from graphs, wheel speed there is zero? So vehicle standing still? Steering arm force is not output to ffb wheels ar speeds under some 2-3mph. Steering arm forcee wheel gets under that speeds is zero. Substitute smooth force is used then, while when accelerating steering arm force is blended in up to some 5mph. Have you swapped real feel that is in game to some other version? And i repeat no one did answer as we simply cant tell why your pc is crashing in given situation.
No the brakes are on and the car is skidding in the gravel. So the wheel rotations are zero, but the car is still moving forward. As the car slows... at some point the vibrations begin and the frequency of these vibrations increase as the car comes to a stop. If its useful I can send the Motec log file for your convenience. It seems that reducing the FFB strength to 80% has stopped the system lockups. Reducing the "RESISTANCE=" value in the track TDF stops the vibrations when skidding in the gravel. There is some feedback issue in the FFB code I would guess. I don't see how hardware could generate this signal. Skidding on a road, runoff, rumble, or grass surface is fine. I could have some issue with my hardware for sure, everything seems find to me. The vibrations are strange, and since the Beta program began I have never had this issue. Cheers
Oh while locked up, that could be a hint, perhaps somewhat related to zero speed. Perhaps you already tried, but try using Pure ffb only, no effects. Since you mod tracks try zeroing all BumpAmpLen in tdf file There is some other parameter too but can't recall which one (not the area i know fully).
No AMS Beta is pure so that I can participate in the Beta program. I think that the abnormal steering forces recorded in the telemetry is the issue to be addressed.
I have tried pure360 / pure180 / and pure + effects1. They all produce the same vibrations when skidding in the gravel. The bumplength values seem to have no affect... its the resistance value that makes the difference.
