Scawen, values are the same when the bug is not occuring, when it does, I get two click releases, the first is wrong (and presumably is the one used) and the second one is correct. Right clicks do not trigger the messages at all.
Not all cards will run 3D in 1x PCI-E slot though. I had to upgrade to an SLI board in order to get a second 16x PCI-E slot so that the card would run in 3D mode, and thus do more than merely extend my desktop onto the third display.
More points for a longer pitstop? Interesting. Mixes strategies up for sure.
Aha, I think this is the issue I've been having. If you don't delete that file, what happens if you just repeatedly mash the left and right mouse buttons instead of just single clicking? That seems to get it to work for me.
An interesting idea, but remember that a car takes a lot more physics and rendering processing, not to mention memory, so you'll never be able to add 512 cars instead of hay bails, and I'd imagine that for every car you added, you'd be able to have one less person join the server. There's also the issue of the dedicated server, as I don't think it runs the physics, so either it would start having to, or somehow get this offloaded to one of the guests.
Tweak does not work for demo users anyway.
Would definitely like to watch a decent video made with the mod. 
Shows promise though.
Shows promise though.
Wolf is a type of dog. I nealy went with Panther, as that's a cat. The domesticated versions are boring as hell. A good pet should either eat me, the postman, or ideally both.
You can still download the normal mod from my site, then install the I-SCAR patch from my earlier post in this thread. Not quite as convenient but at least it's online.
This is related to the way trailing arm suspension creates jacking forces which cause this unusual behaviour. I'm not sure if anti-dive or anti-squat geometry works in LFS at all.
Please use existing threads on this topic.
And what evidence do you have that your account was taken by someone you know? Anyway, since you only have a demo licence atm anyway, there's nothing stopping your from creating another account for the month, if you feel it was definitely not you who used the hack.
Enough of the "my team > your team" dribble.
Did the trick! I was able to copy the file without the hard drive going bonkers, but presumably the data was already lost, as I get a few frames of corruption in the video (at the point where it would normally freeze). Still, better than not being able to play it at all.
Thanks, I'm downloading the demo now.
Compare tyre forces, if they are equal at all times*, then diff is open.
*Steady state, of course
With diff model B stated above, tyre forces stay within 5% of one another, which is pretty good given the massive integration error involved at the time step.
One wheel will spin more than the other, unless resistances are exactly equal, in which case they will both accelerate at the same rate, which is only ever going to happen when they are in the air.
So why is it then that keeping the felt torques equal creates an open diff, and sending equal torque to each wheel creates a LSD?
There is a difference between the torque you try to apply, and the torque you could read if you were to somehow get a torque wrench into the system.
If a single wheel (of 10m radius) is connected to an electric motor which creates 1000Nm of torque, but that wheel is in the air, then 1000Nm of torque goes into spinning the wheel, and 0Nm when be read at the wrench.
If the tyre is on the ground and able to produce 750N of force, then you will read 750Nm at the wrench, and the other 250Nm increases the angular velocity of the wheel.
Likewise when the tyre can produce 1500N of force, then all 1000Nm can be read on the wrench, and the wheel does not accelerate.
It's this figure, at the wrench, that is actually felt by the diff, and should remain equal on both shafts during steady state cornering. Hence why if you just try to send an equal amount of torque to each shaft, the actual torque that could be read will differ, and you don't have an open diff.
Or at least, that's the conclusion I've come up with.
I suspect that, although method A doesn't seem to work as well, it is just more prone to integration error and oscilations, as I think the tyre which produced more force alternates each frame, so on average it might actually be correct. I might try to investigate a bit deeper later on, but for now it seems to work, so I'll stick with what I've got, unless somebody here can come up with something better.
You can edit your own thread titles, at least for a while after it is made.
Also, "lol" is not a reasonable substitute for brackets.
If you find a bug with a test patch, please double check with your regular patch install, so that you can compare. This forum is correct for existing bugs (i.e. they exist in patch Z), else you should use the test patch thread for new bugs. If you could compare and report your findings, I'll merge these posts in with that thread if necessary.
It decides what the maximum torque is that can be applied to any wheel, sends this to both, then sends the extra torque to the wheel of least resistance, where we hava already sent exactly enough torque to match the resistance, so the extra torque is what results in angular acceleration.
Less so than the first method, actually.
Both methods seem to give an open diff behaviour, i.e. the inside wheel spins up, but the second method is actually better at keeping the tyre torques equal, and was the one given to me to use, whereas the first method I invented.
Unfortunately, I understand the first method, and how I could then adapt it to make an LSD, whereas I'm clueless about what to do with the second method. I have some code for a TBD but I haven't got my head around it yet.
Yes, LFS uses a simple line collision, and does not use a a multi-rigid body approach, so the wheel instantly moves to follow the contour of the road.
FGED GREDG RDFGDR GSFDG