J.B. and Shotglass. Excellent points on all ends 
Actually, this post was a bit snippy and defensive so I must ask you a couple questions here:
1) How sensitive to temperature is "rubber"? Which rubber? And in what ways is it sensitive to it?
2) What "rewriting of physics" would need to be done to implement my suggestion?
If you were polite I'd have left ya alone, but...
Not all tread rubber compounds are noticeably sensitive to temperature (up to the point of melting, anyway). Particularily in street compounds used in drifting. Touche..
Edit: As I said before, I very rarely drift in LFS. If I want to do some drifting I just fire up my own sim
Anyway, drifting in LFS absolutely destroys the tires and causes them to lose huge amounts of grip due to the heating in an extraordinarily short amount of time. I'm suggesting having a tire that's toned down a bit (which is closer to reality, mind you, so your "imaginary world" comment can find a place to park itself)
Edit: As I said before, I very rarely drift in LFS. If I want to do some drifting I just fire up my own sim
Anyway, drifting in LFS absolutely destroys the tires and causes them to lose huge amounts of grip due to the heating in an extraordinarily short amount of time. I'm suggesting having a tire that's toned down a bit (which is closer to reality, mind you, so your "imaginary world" comment can find a place to park itself)
Last edited by jtw62074, .
Hi, guys. Great discussion you've got going here. Lots of knowledgable people here
I read this thread last night (quite awhile ago) so I probably won't remember to hit all the points that are being discussed here.
The original poster's friend said he was setting up his car to deliberately lock the wheels in an effort to regain control of the car. What he didn't say is if he was doing this to lock all four wheels or just two of them, and he didn't say if "regaining control" meant trying to recover from a spin or avoid plowing (understeering at/over the limit) off the track. When I think about regaining control it's all about getting out of a slide I didn't want to be in, so I'll assume he's talking about the same thing.
1) Locking the fronts: This should indeed be a quick way to recover from a slide because in doing so you pretty much kill off all the side force the front tires are producing (not really, it depends on the angle of slide, but it will generally reduce the side force a whole lot very quickly). Depending on what exactly is going on, this could be a very quick way to straighten up the car and might even be quicker than countersteering. So yes, it should indeed work to some extent. However, as many of you pointed out, this isn't the best way to go about things as you can overheat and flat spot the tires. If the purpose is to recover from a spin then the fact that the grip drops off only goes to reducing that side force even more, which would straighten up the car faster. Note that this is different from trying to stop the car as quickly as possible. Here we're interested in recovering from a spin, not stopping ASAP.
2) Locking the rears: I think we all know what happens when you do that. Around you go in a real hurry. The only time I can see this being a useful thing is in a car that is understeering through a slow corner and about to go off the outside of the track. A quick jab of the brakes could get the car more sideways and keep you from going off. A setup like that could of course make for hairy moments during all normal braking maneuvers, so I doubt that's generally a wise thing to do. I sure wouldn't want it, but there could be times when you might want that to recover from a boo-boo.
3) If he's locking all four tires in an attempt to recover from a spin, well, if the balance and so on is just so then this might help too, but it's a bit unlikely. This one probably is not a good idea.
I'm betting that #1 above is what the OP's friend was describing. Indeed, this could very well save you from some nasty moments by nearly killing off all sideforce at the front tires which should indeed straighten up the car in a real hurry, albeit at the expense of the tires. A setup like that will have less braking force at the rear than you could get away with normally (otherwise they'd probably lock too when you emergency-jabbed the brakes to straighten the car), so to have this crutch you're most likely giving up a fair amount of your regular braking performance. He probably brakes earlier going into the corners than somebody with a more typical setup, and probably does not trail brake at all. All braking is done in a straight line.
Bottom line there really is he's most likely giving up performance in many areas of the track in order to have a sort of "oh crap" button at the bottom of the brake pedal for emergency spin recovery. I used to do that in GPL too, but haven't found it necessary in LFS due to my superb driving skills
Some thoughts on ABS: I am not familiar with the current ABS technology in production cars. However, theoretically it should be possible to make a system that under straight line braking outperforms most drivers. The paper one of you posted shows that for the nine or so vehicles they tested this indeed is the case, so perhaps the technology has been to that level for some time now. On the other end of the argument I must agree that perhaps Mario Andretti or the likes might outperform the ABS systems in certain maneuvers, so this is a case where both sides might really be right. ABS technology development is essentially about trying to control the slip ratios better than a human can, so it's more or less a game between the programmers/hardware developers and real drivers. Some may be better than others at it (both on the dev side and the real drivers' side).
Locking wheels and stopping distances: There may be a bit more here than initially meets the eye. First, as most of you have pointed out already, a locked tire produces less grip than a rotating one. The 11% slip ratio "rule" that Wikipedia posted is in the ballpark, but it actually varies quite a bit from one tire to the next. Anyway, that's not important. What's important is that there is some slip ratio where the braking force will be highest, and if you go any higher than that then the braking force reduces (as does the road reaction torque on the wheel). If the brakes aren't released a little bit at that point, the wheel will lock up. If you do this on all four tires then you could expect straight line braking distance to increase.
Another possible scenario is this: The rears might be at the 11% (or whatever) slip where the maximum braking force is. If the fronts are too then this is where you're getting the shortest straight line braking distance. If you then lock up either the front or the rears then the distance will increase due to the grip dropping off after the peak. Keep in mind that the grip when the tires are locked is changing quite a lot throughout time due to the temperature changing, so it's not quite as clean and neat as the force graphs show in the case of a tire that has actually locked. I've seen some old braking test data where an accelerometer was put in a couple of cars to measure the deceleration rate under a panic stop situation. It peaked before the tires locked, then dropped somewhat gradually to a constant (but lower) deceleration rate. So the tires indeed lost grip. This was done with no rear wheel braking at all if I recall correctly, so it was purely the fronts doing all the work.
What's important to understand is that most production cars are built more for safety than performance. For the most part this means directional stability (building in enough understeer so if you suddenly swerve to avoid something at typical highway speeds you don't spin out), but perhaps more importantly, the panic stop/swerve situation. I bet that most sim racers that are really into this wonderful hobby will react very differently to an emergency than your average joe on the street would. Nonetheless, production cars aren't designed to handle the way you and I want, but rather to keep Mom and Pop and their teenage maniac newly-licensed kids safe.
Probably the most important emergency situation that a vehicle engineer is likely to be considering is the panic stop/swerve, as mentioned. I.e., you're puttering along and suddenly there's something stopped in front of you. Average Joe does up to three things in this situation, in this order:
1) He slams the brake pedal to the floor.
2) He steers the wheel wildly in one direction.
3) Upon finding that the front wheels have locked and his car isn't turning at all, he steers the wheel all the way to full lock. Nothing new happens there.
Then, usually there's a crash of some sort
Ok, from the "engineering for safety" standpoint, you want minimum stopping distances. However, you don't want Average Joe to slam the brake pedal to the floor, steer wildly to one side, only to find himself sliding sideways into goodness knows what. One sure fire way to avoid that is to design the braking system so the front wheels lock and the rear wheels do NOT. Ideally, the rear wheels would hit their peak 11% or whatever and at the same time, the fronts lock up.
The purpose of this is to do keep the car straight, which is exactly what the original poster's friend is trying to do. What he's essentially done is designed into his LFS car the same safety feature that's built into our road cars. Does this produce the best stopping distance? Probably not, but it actually might do so indeed depending on how the braking system is set up.
Here's the exception: If the brake pedal on the above car is pushed down to the point where the front wheels are at their 11% slip, the rears will be well below that. So the fronts are making their peak braking force, but there's still plenty more left at the rear. If you now push the brake to the floor, you lock up the fronts, which reduces front grip, but in turn you increase the rear braking. Which one wins? Well... It depends on the car. The point is that from one production car to the next I wouldn't be surprised at all to find some cars that stop better with the front wheels locked and others that stop better with a "best effort" of feathering/pumping the brake.
Ideally really you don't want Joe to spin out or be forced to go straight either. You want the car to react the way Joe wants it to. However, if you have to pick one of the options, it's generally going to be better if he just plows straight ahead into whatever he was trying to avoid while slowing down as much as possible at the same time. Imagine spinning sideways into oncoming traffic or off the road into a tree. It's usually much better just to rearend somebody. Fender benders really don't hurt as much as people think as long as they have their seatbelts on, but people absolutely freak out when they're about to even rear end somebody at 10-20 mph. I saw a video of a bunch of cars sliding one after the other on slick ice into a line of cars. More than half of the people were literally jumping out of their cars to avoid the impact. Bumper cars hurt more than that, so stay in the car!
Enter ABS and other systems. As many of you pointed out, the main reason for ABS and stability control on production cars is to keep the fronts from locking, which sends Joe straight into whatever he was trying to steer around. At the same time, in a straight line, a regular braking system assisted with ABS is able to go ahead and try to keep all the slip ratios at 11%. When he turns the wheel to avoid the big bad object, without the ABS the wheels would then suddenly lock. ABS will relieve some of that brake pressure so the slip ratios stay at 11% (or whatever the ABS designer wants it too; the algos are probably much more complex than that), so he's able to get plenty of braking but also is able to steer around the big bad obstacle (BBO), while at the same time not spin out during his frantic, panic stricken effort to avoid smashing into the BBO.
This is a neat discussion largely because nearly all people on all sides of the debate here are really correct, even though some of the points seem to contradict each other
I read this thread last night (quite awhile ago) so I probably won't remember to hit all the points that are being discussed here.
The original poster's friend said he was setting up his car to deliberately lock the wheels in an effort to regain control of the car. What he didn't say is if he was doing this to lock all four wheels or just two of them, and he didn't say if "regaining control" meant trying to recover from a spin or avoid plowing (understeering at/over the limit) off the track. When I think about regaining control it's all about getting out of a slide I didn't want to be in, so I'll assume he's talking about the same thing.
1) Locking the fronts: This should indeed be a quick way to recover from a slide because in doing so you pretty much kill off all the side force the front tires are producing (not really, it depends on the angle of slide, but it will generally reduce the side force a whole lot very quickly). Depending on what exactly is going on, this could be a very quick way to straighten up the car and might even be quicker than countersteering. So yes, it should indeed work to some extent. However, as many of you pointed out, this isn't the best way to go about things as you can overheat and flat spot the tires. If the purpose is to recover from a spin then the fact that the grip drops off only goes to reducing that side force even more, which would straighten up the car faster. Note that this is different from trying to stop the car as quickly as possible. Here we're interested in recovering from a spin, not stopping ASAP.
2) Locking the rears: I think we all know what happens when you do that.
Around you go in a real hurry. The only time I can see this being a useful thing is in a car that is understeering through a slow corner and about to go off the outside of the track. A quick jab of the brakes could get the car more sideways and keep you from going off. A setup like that could of course make for hairy moments during all normal braking maneuvers, so I doubt that's generally a wise thing to do. I sure wouldn't want it, but there could be times when you might want that to recover from a boo-boo. 3) If he's locking all four tires in an attempt to recover from a spin, well, if the balance and so on is just so then this might help too, but it's a bit unlikely. This one probably is not a good idea.
I'm betting that #1 above is what the OP's friend was describing. Indeed, this could very well save you from some nasty moments by nearly killing off all sideforce at the front tires which should indeed straighten up the car in a real hurry, albeit at the expense of the tires. A setup like that will have less braking force at the rear than you could get away with normally (otherwise they'd probably lock too when you emergency-jabbed the brakes to straighten the car), so to have this crutch you're most likely giving up a fair amount of your regular braking performance. He probably brakes earlier going into the corners than somebody with a more typical setup, and probably does not trail brake at all. All braking is done in a straight line.
Bottom line there really is he's most likely giving up performance in many areas of the track in order to have a sort of "oh crap" button at the bottom of the brake pedal for emergency spin recovery. I used to do that in GPL too, but haven't found it necessary in LFS due to my superb driving skills
Some thoughts on ABS: I am not familiar with the current ABS technology in production cars. However, theoretically it should be possible to make a system that under straight line braking outperforms most drivers. The paper one of you posted shows that for the nine or so vehicles they tested this indeed is the case, so perhaps the technology has been to that level for some time now. On the other end of the argument I must agree that perhaps Mario Andretti or the likes might outperform the ABS systems in certain maneuvers, so this is a case where both sides might really be right. ABS technology development is essentially about trying to control the slip ratios better than a human can, so it's more or less a game between the programmers/hardware developers and real drivers. Some may be better than others at it (both on the dev side and the real drivers' side).
Locking wheels and stopping distances: There may be a bit more here than initially meets the eye. First, as most of you have pointed out already, a locked tire produces less grip than a rotating one. The 11% slip ratio "rule" that Wikipedia posted is in the ballpark, but it actually varies quite a bit from one tire to the next. Anyway, that's not important. What's important is that there is some slip ratio where the braking force will be highest, and if you go any higher than that then the braking force reduces (as does the road reaction torque on the wheel). If the brakes aren't released a little bit at that point, the wheel will lock up. If you do this on all four tires then you could expect straight line braking distance to increase.
Another possible scenario is this: The rears might be at the 11% (or whatever) slip where the maximum braking force is. If the fronts are too then this is where you're getting the shortest straight line braking distance. If you then lock up either the front or the rears then the distance will increase due to the grip dropping off after the peak. Keep in mind that the grip when the tires are locked is changing quite a lot throughout time due to the temperature changing, so it's not quite as clean and neat as the force graphs show in the case of a tire that has actually locked. I've seen some old braking test data where an accelerometer was put in a couple of cars to measure the deceleration rate under a panic stop situation. It peaked before the tires locked, then dropped somewhat gradually to a constant (but lower) deceleration rate. So the tires indeed lost grip. This was done with no rear wheel braking at all if I recall correctly, so it was purely the fronts doing all the work.
What's important to understand is that most production cars are built more for safety than performance. For the most part this means directional stability (building in enough understeer so if you suddenly swerve to avoid something at typical highway speeds you don't spin out), but perhaps more importantly, the panic stop/swerve situation. I bet that most sim racers that are really into this wonderful hobby will react very differently to an emergency than your average joe on the street would. Nonetheless, production cars aren't designed to handle the way you and I want, but rather to keep Mom and Pop and their teenage maniac newly-licensed kids safe.
Probably the most important emergency situation that a vehicle engineer is likely to be considering is the panic stop/swerve, as mentioned. I.e., you're puttering along and suddenly there's something stopped in front of you. Average Joe does up to three things in this situation, in this order:
1) He slams the brake pedal to the floor.
2) He steers the wheel wildly in one direction.
3) Upon finding that the front wheels have locked and his car isn't turning at all, he steers the wheel all the way to full lock. Nothing new happens there.
Then, usually there's a crash of some sort
Ok, from the "engineering for safety" standpoint, you want minimum stopping distances. However, you don't want Average Joe to slam the brake pedal to the floor, steer wildly to one side, only to find himself sliding sideways into goodness knows what. One sure fire way to avoid that is to design the braking system so the front wheels lock and the rear wheels do NOT. Ideally, the rear wheels would hit their peak 11% or whatever and at the same time, the fronts lock up.
The purpose of this is to do keep the car straight, which is exactly what the original poster's friend is trying to do. What he's essentially done is designed into his LFS car the same safety feature that's built into our road cars. Does this produce the best stopping distance? Probably not, but it actually might do so indeed depending on how the braking system is set up.
Here's the exception: If the brake pedal on the above car is pushed down to the point where the front wheels are at their 11% slip, the rears will be well below that. So the fronts are making their peak braking force, but there's still plenty more left at the rear. If you now push the brake to the floor, you lock up the fronts, which reduces front grip, but in turn you increase the rear braking. Which one wins? Well... It depends on the car. The point is that from one production car to the next I wouldn't be surprised at all to find some cars that stop better with the front wheels locked and others that stop better with a "best effort" of feathering/pumping the brake.
Ideally really you don't want Joe to spin out or be forced to go straight either. You want the car to react the way Joe wants it to. However, if you have to pick one of the options, it's generally going to be better if he just plows straight ahead into whatever he was trying to avoid while slowing down as much as possible at the same time. Imagine spinning sideways into oncoming traffic or off the road into a tree. It's usually much better just to rearend somebody. Fender benders really don't hurt as much as people think as long as they have their seatbelts on, but people absolutely freak out when they're about to even rear end somebody at 10-20 mph. I saw a video of a bunch of cars sliding one after the other on slick ice into a line of cars. More than half of the people were literally jumping out of their cars to avoid the impact. Bumper cars hurt more than that, so stay in the car!
Enter ABS and other systems. As many of you pointed out, the main reason for ABS and stability control on production cars is to keep the fronts from locking, which sends Joe straight into whatever he was trying to steer around. At the same time, in a straight line, a regular braking system assisted with ABS is able to go ahead and try to keep all the slip ratios at 11%. When he turns the wheel to avoid the big bad object, without the ABS the wheels would then suddenly lock. ABS will relieve some of that brake pressure so the slip ratios stay at 11% (or whatever the ABS designer wants it too; the algos are probably much more complex than that), so he's able to get plenty of braking but also is able to steer around the big bad obstacle (BBO), while at the same time not spin out during his frantic, panic stricken effort to avoid smashing into the BBO.
This is a neat discussion largely because nearly all people on all sides of the debate here are really correct, even though some of the points seem to contradict each other
Last edited by jtw62074, .
Didn't read the whole thread here so sorry if I say something that's been said already. Since the tires in LFS are very sensitive to temperature and wear it'd be nice for the drifters to have a tire that is not nearly so sensitive, but produces a bit less grip than the others so people don't find they are better for racing than the road supers are.
I don't drift in LFS very often at all, but a +1 here anyway. The more happy people the better
I don't drift in LFS very often at all, but a +1 here anyway. The more happy people the better
Sorry for dropping out of the other thread. I do want to clarify a couple of things:
1) The curves I showed were taken under braking rather than acceleration, so the slip ratio of 1 is a fully locked tire at 20mph. (And yes, it was a real tire, not "something like" a real tire
) Traction curves can look a little different than that, but the total amount of drop off in that range is pretty similar. I've got data at three or four different speeds on that tire and the curves all look a little different. So what do the curves look like at 0 mph and slip ratios 0-10? I couldn't tell you for sure. Maybe somebody with a G-Tech or accelerometer and a powerful car could make some tests. Just show us the acceleration rate for the first few seconds with a lot of wheelspin versus trying to modulate it. That'd give some indication of it at super high slip ratios if you wanted to know.
Just to clarify, there will indeed be a drop off. The question is how much? I would be cautious about making an attempt to extrapolate the data I showed for braking at 20mph out to a slip ratio of 10 and assuming that's going to be the same as 0mph under traction... It's likely to vary quite a lot in reality. The curves at 5 mph are not the same as 20, 40, 60, etc., in reality.
2) Actual traction data is pretty rare, really. Most data is shown under braking as tire companies and automobile engineers have historically been interested in much larger numbers in road safety than getting quick starts off the line. I.e., braking performance gets a lot of attention.
Another reason traction data (slip ratio under acceleration) is pretty rare is that you need a machine with quite a lot of power in order to generate it. In contrast, for braking you just need a really powerful brake, essentially.
The link below shows some different tire testing machines as well as some measured results for one tire out to 28 degrees slip angle, and other tests for traction and braking slip ratio (where you can see the curves look pretty similar). In this case you can see a bit of a force drop off at high loads and high slip angles. I've seen another one like that all the way out to 35 degrees that was absolutely pancake flat, so there's some variance there:
http://rclsgi.eng.ohio-state.e ... Model%20-%20ME%20794G.ppt
Concerning the F120002 curves:
http://www.racer.nl/images/pac_f12002.jpg
I highly doubt this is real data from an actual F1 tire. I'd be surprised to find slip ratio +2 data available even to Ferrari F1, let alone find that the data was handed out to a video game company. They're unbelievably secretive. I could be wrong of course, but it's not safe to assume that's a real curve for any tire at all.
Concerning the LFS curves in slip ratio:
http://www.lfsforum.net/attach ... id=11060&d=1149423472
Looks like there already is quite a drop off there. It'd be a good idea to run in the car park and just keep an eye on the acceleration (F9 display) over the first couple of seconds to see what happens with a lot of slip versus more optimal slip.
One other source of info you might look into is 0-60 foot times for drag racing cars. Be careful here though as wrinkle wall drag slicks are likely to be very different from road racing tires, so make sure the data you find is done with something somewhat comparable to a road racing tire. A lot of drag racers log nearly every run they ever make, including their 0-60 foot times and so on along with weather conditions, whether they had a good launch or a lot of wheelspin and so on. If you can find somebody that has logged data like that you could get a pretty good idea how much of a drop off you should see.
Just some ideas
1) The curves I showed were taken under braking rather than acceleration, so the slip ratio of 1 is a fully locked tire at 20mph. (And yes, it was a real tire, not "something like" a real tire
) Traction curves can look a little different than that, but the total amount of drop off in that range is pretty similar. I've got data at three or four different speeds on that tire and the curves all look a little different. So what do the curves look like at 0 mph and slip ratios 0-10? I couldn't tell you for sure. Maybe somebody with a G-Tech or accelerometer and a powerful car could make some tests. Just show us the acceleration rate for the first few seconds with a lot of wheelspin versus trying to modulate it. That'd give some indication of it at super high slip ratios if you wanted to know. Just to clarify, there will indeed be a drop off. The question is how much? I would be cautious about making an attempt to extrapolate the data I showed for braking at 20mph out to a slip ratio of 10 and assuming that's going to be the same as 0mph under traction... It's likely to vary quite a lot in reality. The curves at 5 mph are not the same as 20, 40, 60, etc., in reality.
2) Actual traction data is pretty rare, really. Most data is shown under braking as tire companies and automobile engineers have historically been interested in much larger numbers in road safety than getting quick starts off the line. I.e., braking performance gets a lot of attention.
Another reason traction data (slip ratio under acceleration) is pretty rare is that you need a machine with quite a lot of power in order to generate it. In contrast, for braking you just need a really powerful brake, essentially.
The link below shows some different tire testing machines as well as some measured results for one tire out to 28 degrees slip angle, and other tests for traction and braking slip ratio (where you can see the curves look pretty similar). In this case you can see a bit of a force drop off at high loads and high slip angles. I've seen another one like that all the way out to 35 degrees that was absolutely pancake flat, so there's some variance there:
http://rclsgi.eng.ohio-state.e ... Model%20-%20ME%20794G.ppt
Concerning the F120002 curves:
http://www.racer.nl/images/pac_f12002.jpg
I highly doubt this is real data from an actual F1 tire. I'd be surprised to find slip ratio +2 data available even to Ferrari F1, let alone find that the data was handed out to a video game company. They're unbelievably secretive. I could be wrong of course, but it's not safe to assume that's a real curve for any tire at all.
Concerning the LFS curves in slip ratio:
http://www.lfsforum.net/attach ... id=11060&d=1149423472
Looks like there already is quite a drop off there. It'd be a good idea to run in the car park and just keep an eye on the acceleration (F9 display) over the first couple of seconds to see what happens with a lot of slip versus more optimal slip.
One other source of info you might look into is 0-60 foot times for drag racing cars. Be careful here though as wrinkle wall drag slicks are likely to be very different from road racing tires, so make sure the data you find is done with something somewhat comparable to a road racing tire. A lot of drag racers log nearly every run they ever make, including their 0-60 foot times and so on along with weather conditions, whether they had a good launch or a lot of wheelspin and so on. If you can find somebody that has logged data like that you could get a pretty good idea how much of a drop off you should see.
Just some ideas
Quote from your link:
"Anyway, all of this is now simulated in X-Plane, and even though you
do not have the ability to actually DRIVE the wheels with engine-
torque, the physics are there,"
I didn't read the whole post on account of its length, but from what I did read it appears that they've included a new tire model and he used anecdotes relating to how they would work on cars. I.e., he showed some of his thinking about tires and gave a little insight into how the model works. I didn't see anything in there about actually including cars in 8.5 though. I could have missed it though, but if you can't drive the tires with engine torque I doubt there'd be much of an interesting "auto" aspect to X-Plane in this next release
The discussion on tires was a little disconcerting though. Lots of talk about static and dynamic friction coefficients, which don't pertain to tires at all. One quote:
"Then, as in reality, the tire shifts from static to kinetic
coefficient of friction as the total required force exceed the
available friction and you enter a skid."
And:
"Now here is where it gets
complicated... the direction of this bond is NOT simply equal to the
sliding direction of the tire!"
That's just flat out wrong in the case he described and violates Newton's laws. Lots of stuff like that in there...
Anyway, it's sure to make much better landing and ground handling behavior for the airplanes either way. I know Flight Sim 9 doesn't do a very good job of it...
The intention to develop X-Auto was announced by Austin at least as early as May, 2002. There was a buzz for a little while then it appeared to sort of fizzle out as he took on other projects. Given their very public way of discussing what they're working on it would suggest they haven't touched in years. Maybe I missed something though over the past few months.
It would be cool, but I'm not holding my breath about seeing it anytime remotely soon. From what I've read on their forums on discussions about cars they don't appear to have a very high level of knowledge in that area, unfortunately. That could very well change though of course.
Last edited by jtw62074, .
X-Auto? Long time ago.... Not happenin' I think..
Yes, that's about right for % slip nowadays. The data I showed here are on very old tires. Back then they used to be a lot more flexible, so peaks came in at very high slip ratios and slip angles. A bit like GPL tires really
Yeah, I know what you mean. For a long time I was a joystick driver and did rather well with that. Once I switched to a wheel I lost a lot of time, but eventually got faster than I was before. With a joystick it's a bit tricky to keep the steering in one spot while adjusting throttle/brake. Much easier with a wheel of course, so eventually I got quicker that way.
I played LFS quite a bit in Holland using an RC radio hooked up to our USB adaptor since the Serpent engineer I was staying with didn't have a place to hook up his wheel. It was hard at first, especially since I had to use automatic shifting, but after a couple of weeks was still running pretty quick.
All just a matter of adjustment I think. You could drive your current times with a trackball or VR glove probably given enough practice
I played LFS quite a bit in Holland using an RC radio hooked up to our USB adaptor since the Serpent engineer I was staying with didn't have a place to hook up his wheel. It was hard at first, especially since I had to use automatic shifting, but after a couple of weeks was still running pretty quick.
All just a matter of adjustment I think. You could drive your current times with a trackball or VR glove probably given enough practice
I wholeheartedly agree and this is an arguement that has been brought up since the beginning, and is brought up in every sim forum in town.
A sim really needs to advance through some stages of realism before this becomes a really valid argument. Before the latest patch I really felt the combined force stuff was way, way out of whack. Didn't want to criticize it too much or too harshly though because I'm a pretty nice guy and wouldn't appreciate that from another dev in my forum either, but now that it's been fixed up a great deal I can say it.
Prior to the combined stuff being drastically improved I argued against the common view Woz is stating here because in my opinion LFS had not advanced to the point yet where this point matters a whole lot. In other threads I'd talked about how I'd fixed up the combined stuff in my sim and it was worlds different and much easier to drive than LFS was. That was very true, and a common response was that I didn't have bumps in the road and that sort of thing. Well... Now that you have that area fixed up in LFS you can see how much easier it is. Here's a case where realistic actually = easier
Now that the tires have been improved so much LFS is really at the point where the lack of physical feedback imho is a valid argument. There are still a couple of little things I'd critique, but so far nobody has noticed or mentioned them so I'll remain mum
LFS right now in several areas has the best tire model and realistic output data of any sim out there, and in some other areas there are shortcomings, but the strengths more than make up for that. These other areas aren't too tough to work out though and I'm sure we'll see it eventually.
Last edited by jtw62074, .
Key word here is "significant." Is there a drop? Yes, there usually is, at least in the data I've seen.
Here are a couple of examples of real tires:
http://www.PerformanceSimulations.com/files/longslip.JPG
http://www.PerformanceSimulations.com/files/longslip2.JPG
These are old street tires rather than racing tires, but at least it's something concrete to put numbers too. The tests were done on a trailer on concrete, although the results on asphalt look pretty similar. The curve we're interested in here is the top one circled in red which shows longitudinal force vs. slip ratio with no slip angle. (The other lines are showing lateral and longitudinal force at varying slip angles as a function of slip ratio. That's the "combined force" stuff that needs to be right and was goofy in LFS before the latest patch.)
Opening up these in Paint and counting pixels it can be seen that the first curve drops 9.5% from the peak out to 0.8 slip ratio, which is BIG slip. These tests were done at 20 mph, so you're spinning the wheels up to the equivalent of 36 mph at the far right of the graph. The second test only shows a 3% drop.
Is massive wheelspin going to reduce acceleration? Yes, it could depending on the tire, but generally it's not a whole lot or some massive drop off like most people seem to think where the acceleration dumps down to 60% of what you'd otherwise get or anything like that.
Now, again, the key word here is "significant" drop off. Keep in mind that all tire data you see like this has been processed somehow. It's not the actual raw data read from the machine during the test. The real data points are very noisy in reality. I.e., the forces jump up and down quite a bit instead of just following along those curves perfectly. If the noise is a 10% up/down variation from one moment to the next, then for all practical purposes there are instants in time where at 0.8 slip ratio you could be getting more grip than at some instants of time at the peak. So on that one curve where you see a 3% drop off, once you're on the actual track you could very well see very little difference in acceleration at all.
There are other tires that drop off more than this, and others that drop off less and are essentially flat (until you start melting the rubber of course, that will change things). I've got one here that rises to a peak, then drops off maybe 1-2%, then rises again to the same level as the peak at 0.8 slip ratio. So anything can happen.
Unfortunately I don't have any data like this on a modern racing tire. I've been told though that there are indeed tires that really don't have any significant drop off at all. Drag slicks could be a major departure from these curves though and in all likelihood are, but I won't go into the reasons for that here.
On wet pavement nearly all tires drop off much more than this, so if you're testing with your poopometer in your own car, make sure to do it in the dry.
Something else interesting is that any data I've seen on big truck tires all have very steep drop offs after the peak, probably even more so in general than regular car tires do in the wet. And again, this is a design issue in tires. You can do more or less what you want. It depends what you're trying to design the tire to do and where you'll allow it to sort of, well, not work so well that separates one tire from the next.
Here are a couple of examples of real tires:
http://www.PerformanceSimulations.com/files/longslip.JPG
http://www.PerformanceSimulations.com/files/longslip2.JPG
These are old street tires rather than racing tires, but at least it's something concrete to put numbers too. The tests were done on a trailer on concrete, although the results on asphalt look pretty similar. The curve we're interested in here is the top one circled in red which shows longitudinal force vs. slip ratio with no slip angle. (The other lines are showing lateral and longitudinal force at varying slip angles as a function of slip ratio. That's the "combined force" stuff that needs to be right and was goofy in LFS before the latest patch.)
Opening up these in Paint and counting pixels it can be seen that the first curve drops 9.5% from the peak out to 0.8 slip ratio, which is BIG slip. These tests were done at 20 mph, so you're spinning the wheels up to the equivalent of 36 mph at the far right of the graph. The second test only shows a 3% drop.
Is massive wheelspin going to reduce acceleration? Yes, it could depending on the tire, but generally it's not a whole lot or some massive drop off like most people seem to think where the acceleration dumps down to 60% of what you'd otherwise get or anything like that.
Now, again, the key word here is "significant" drop off. Keep in mind that all tire data you see like this has been processed somehow. It's not the actual raw data read from the machine during the test. The real data points are very noisy in reality. I.e., the forces jump up and down quite a bit instead of just following along those curves perfectly. If the noise is a 10% up/down variation from one moment to the next, then for all practical purposes there are instants in time where at 0.8 slip ratio you could be getting more grip than at some instants of time at the peak. So on that one curve where you see a 3% drop off, once you're on the actual track you could very well see very little difference in acceleration at all.
There are other tires that drop off more than this, and others that drop off less and are essentially flat (until you start melting the rubber of course, that will change things). I've got one here that rises to a peak, then drops off maybe 1-2%, then rises again to the same level as the peak at 0.8 slip ratio. So anything can happen.
Unfortunately I don't have any data like this on a modern racing tire. I've been told though that there are indeed tires that really don't have any significant drop off at all. Drag slicks could be a major departure from these curves though and in all likelihood are, but I won't go into the reasons for that here.
On wet pavement nearly all tires drop off much more than this, so if you're testing with your poopometer in your own car, make sure to do it in the dry.
Something else interesting is that any data I've seen on big truck tires all have very steep drop offs after the peak, probably even more so in general than regular car tires do in the wet. And again, this is a design issue in tires. You can do more or less what you want. It depends what you're trying to design the tire to do and where you'll allow it to sort of, well, not work so well that separates one tire from the next.
I don't use FF either very often, even though I've got the red Momo. I just use the self centering most of the time. I can feel what the car is doing purely visually and often forget in our physics discussions that a lot of (most?) people are using FF, which has the potential of confusing what's happening.
In any sim, not just LFS, the FF quality can really change the impression of what's happening. You could have a perfect tire and physics model and all that, but if the FF isn't up to par or done properly or lags or anything like that, people will think the physics model is somehow wrong because the car doesn't feel right. So you'd get the FF guys saying the sim stinks while the non-FF guys that go purely visually and/or with sound cues saying it feels very much like the real thing.
RC racers frequently talk about how the car "feels" even though they have no FF at all and are of course not sitting in the car. They're 50 feet away on top of a driver's stand but can feel when the car gets loose or is pushing and so on, and will come in to make tiny setup changes to improve the handling. Purely visual and of course absolutely realistic since, well, it's real
In any sim, not just LFS, the FF quality can really change the impression of what's happening. You could have a perfect tire and physics model and all that, but if the FF isn't up to par or done properly or lags or anything like that, people will think the physics model is somehow wrong because the car doesn't feel right. So you'd get the FF guys saying the sim stinks while the non-FF guys that go purely visually and/or with sound cues saying it feels very much like the real thing.
RC racers frequently talk about how the car "feels" even though they have no FF at all and are of course not sitting in the car. They're 50 feet away on top of a driver's stand but can feel when the car gets loose or is pushing and so on, and will come in to make tiny setup changes to improve the handling. Purely visual and of course absolutely realistic since, well, it's real
Why not just hit F9 and use the longitudinal G meter instead of making long acceleration runs? Just see how high you can get that off the start in first gear before the acceleration drops. Make sure to control the camber and temperature though as much as possible because that effects things quite a lot in LFS. It'd probably be best to use fully stiff suspension as well in order to transfer weight to the rears as quickly as possible and minimize the dynamic effects.
I.e., try launching at different rpm/wheelspin levels and see what the effects are. That would eliminate the shifting stuff and also minimize what happens due to steering as well. I've done this several times and IIRC have found I get the same acceleration from run to run down to 0.01g, so that's probably a more accurate way of measuring it.
I.e., try launching at different rpm/wheelspin levels and see what the effects are. That would eliminate the shifting stuff and also minimize what happens due to steering as well. I've done this several times and IIRC have found I get the same acceleration from run to run down to 0.01g, so that's probably a more accurate way of measuring it.
Glad you guys liked the posts.
Here's an example of a couple of things I was talking about:
http://performancesimulations.com/files/tire1.JPG
Check out that top right picture. Those are two different tires that have the exact same dimensions. I.e., width, aspect ratio, etc., are identical. Even though the peak grip is very close to the same with both tires (at least out to the slip angle max they tested to there), the shapes of the curves are quite different. This is strictly due to internal differences in the cords, so really, a tire engineer has a great deal of control over the shape of the curves. Nowadays it's all done with FEA on computers of course, and those simulations give quite accurate results.
In the bottom left picture, and again this is real tire data, you can see another pair of tires that are remarkably different. The radial rises very linearily and then rolls off into the peak very quickly, very much like the LFS tires do (that's the most extreme example I've seen though). The bias one is much more progressive. This data is probably 30 years old so the peaks come in at greater slip angles than modern tires do, but it does illustrate the large flexibility a tire designer has in controlling the shapes of the curves.
If you were driving that radial tire you'd say it was far less forgiving than the bias one. It's not that the grip drops off after the peak, but rather the force suddenly stops rising. In the seat of your pants that feels like you lost grip, but really it's the shape of the curve that's responsible for that. The force suddenly stops rising over a very small slip angle range, so it feels like it breaks away suddenly, so it's not really a "losing grip" phenomenon at all, but it sure can feel like it
Here's an example of a couple of things I was talking about:
http://performancesimulations.com/files/tire1.JPG
Check out that top right picture. Those are two different tires that have the exact same dimensions. I.e., width, aspect ratio, etc., are identical. Even though the peak grip is very close to the same with both tires (at least out to the slip angle max they tested to there), the shapes of the curves are quite different. This is strictly due to internal differences in the cords, so really, a tire engineer has a great deal of control over the shape of the curves. Nowadays it's all done with FEA on computers of course, and those simulations give quite accurate results.
In the bottom left picture, and again this is real tire data, you can see another pair of tires that are remarkably different. The radial rises very linearily and then rolls off into the peak very quickly, very much like the LFS tires do (that's the most extreme example I've seen though). The bias one is much more progressive. This data is probably 30 years old so the peaks come in at greater slip angles than modern tires do, but it does illustrate the large flexibility a tire designer has in controlling the shapes of the curves.
If you were driving that radial tire you'd say it was far less forgiving than the bias one. It's not that the grip drops off after the peak, but rather the force suddenly stops rising. In the seat of your pants that feels like you lost grip, but really it's the shape of the curve that's responsible for that. The force suddenly stops rising over a very small slip angle range, so it feels like it breaks away suddenly, so it's not really a "losing grip" phenomenon at all, but it sure can feel like it
Wierd... What's GRC?
I was using a tyre rolling resistance coefficient of 0.012 there, but I doubt it will make much difference at those speeds. Just make sure it's geared so the engine is actually running at the power peak at top speed. Also, not all programs are created equal, so that could be part of it too...
I'm a big fan of the AC Cobra. I just love loud, abnoxious V8's with side pipes and all that fun stuff
Chris West has a blue Cobra with a Jag V-12 in it. First time I've ever heard of anything other than a V8 or V10 in one of those
Last edited by jtw62074, .
I think Shelby is still designing cars, aren't they?
http://www.shelbysupercars.com/
That ultimate Aero is killer. Over 1000 HP and it should do 273 MPH (according to wind tunnel calculations anyway).
Eat your heart out Bugatti Veyron and Ferrari Enzo!
EDIT: Just out of curiousity I ran the car through Straightline Acceleration Simulator (one of my proggies). With the aero and engine data they provided it would indeed do 273 MPH provided the drivetrain efficiency was 87%. Not unreasonable at all. Can you imagine going that fast though? lol
Last edited by jtw62074, .
Yeah, I'd have to agree with that. The only time a car is really hard to control when it's sliding is when it's set up for final oversteer, which is not something you typically want in the first place. Pieter Bervoets told me in his Donkervoort that full opposite lock drifting, even on racing slicks, was actually very easy to do once you practiced in a parking lot for awhile. And this is in a 650Kg car with 280HP... He demonstrated this to me going through a corner once. Normally he'd use a lot of throttle at the apex and simultaneously straighten the wheel and just steer out of the turn with the throttle. Then, he went ahead and used a lot more throttle and we went quite sideways. Full opposite lock for a second or two, then bang, straighten the wheel and let off that excess throttle and the car straightened up immediately.
Fear was plentiful in my mind throughout that, even though there was nothing to hit but a couple of small ditches on either side of the road
Edit: Oh, one more thing. Typically even on public road ways when there's an accident or near accident involving a spin, the car generally does not just spin around. Rather, it goes very sideways and the driver overcorrects. I.e., the slide left, turn right, then don't begin to straighten the wheel until the car is pointing forwards again. By then it's too late because the car is yawing even faster now in the opposite direction. Slide left, then spin around to the right. I was a passenger in a '71 Firebird as a teenager that did several such oscillations before finally spinning around three complete times and stopping dead in the middle of the freeway. 120mph off road excursions need to be dealt with carefully
Excellent lol
Last edited by jtw62074, .
Real curves are measured on a variety of machines such as these:
http://www.ocp.tudelft.nl/tt/vehicle/testing/trailer.htm
http://www.carsafety.com/tire.htm
The tire's load and camber is controlled (as well as possible anyway, there's usually a bit of noise) while the slip angle is swept back and forth throughout the whole range while the forces are recorded. Many machines do slip ratio too (traction/braking) which is how you can get the full friction circle type of plots.
Doug Milliken's company web site discusses the procedure they use at Calspan for tire testing, along with a mention of the cost. It ain't cheap, unfortunately.
http://www.millikenresearch.com/fsaettc.html
For Virtual RC Racing 4.0 we've had a machine just like this built (except MUCH smaller) that measures forces in the same manner. However, in our case with the little foam tires it's all much, much cheaper to do.
The force curves shown in this thread for LFS look quite realistic. What needs to be remembered is that you can take two tires of the exact same width and dimensions from the same manufacturer, but of different "families," and the curves can vary remarkably due to the inner construction of the tire. The number of plies, how the cords are wound, the cord angles, thickness and shape of the rubber in the sidewalls, and so on have an enormous impact on the shape of those curves.
Essentially tire designers can make the curves look pretty much however they want, so as long as you get curves that look vaguely like what you see in this thread for the LFS tires then you're quite realistic in that regard. Since the tires in LFS are fictitious (correct me if they aren't) then it's a bit odd or perhaps a little pointless to say whether they match a real tire or not. They probably do match up very closely with some and not so closely with others. If I had to say yes or no to "are the tires realistic in regards to the shape of the lateral force vs. slip angle curves" I'd say "yes" though.
There are tires that look pretty much just like that. Is the load sensitivity "right?" I couldn't say for sure, but I imagine it's quite well within the ballpark to give the model a big thumbs up in that regard.One thing to throw into the discussion here is that you can have two tires that peak at the same friction coefficient and slip angle, but have different shapes up until that point which has a profound impact on how the car feels (I've spend a lot of time playing with that in my sim). The LFS force curves shown in this thread rise at a nearly constant slope until you're at around 80-90% of peak grip, then they roll off into the peak very quickly. That will feel very different from a curve that rises more steeply in the beginning and starts rolling off (decreasing the slope) earlier. The latter will feel more forgiving and be easier to drive (which will make a lot of people scream "arcade" even though that's very common for real tires. "Realistic" does not automatically mean "hard." It did when you went from Pole Position to Indy 500:The Simulation or GPL, but most racing games have been past that point for many years now...)
That latter tire will also feel as though it does not regain grip as suddenly as the first tire does. That's a driver perception thing though. In reality the tires aren't "losing grip" at all in the first place, so there's nothing to regain, but that may only make sense when you're looking at the curves when thinking about what the car is actually doing. If the curve rises quite linearly and rolls off suddenly into the peak (like in LFS) it will feel like the car breaks away more easily and "regains grip" faster on recovery than it would if the curves rolled off more gradually. But again, tires can be made to have force curves that look however you want, so the ones you have in LFS are perfectly realistic in that aspect
The location of the peak itself has an even bigger impact on how the car feels. I.e., if it peaks at 15 degrees the car will feel more squishy and forgiving. I've run my sim with peaks at or past 20 degrees and it's quite fun, a lot like GPL where you go fastest through the turns with big drift angles (which was totally wrong for '69 F1 cars, but a whole lot of fun). As you start increasing cornering stiffness so you peak at 5 or 10 degrees, the car usually gets harder to control, even if you increase the overall grip at the same time. It's more touchy and hard to find the edge. When I run 3 or 4 degree peaks in my sim, I don't really spin out often at all, but instead end up sliding, then over correcting and spinning back the other direction rather violently. It feels like the car has "regained grip" extremely suddenly with peaks at low slip angles. So much so that it's hard to catch a slide without overcorrecting.
One other thing to point out is that you can take two identical tires from the same manufacturing plant and the curves can vary considerably. I saw a study on this once and was quite shocked at the difference. In the end, the manufacturers are just working with rubber and wire essentially rather than anything rigidly formed, so there's quite a lot of variance from one tire to the next, even if they're supposedly identical. My PC with the scanner is down at the moment so I can't show the graphs from those tests to you guys, unfortunately.
I'm curious about the "Deformations of rolling tires that affect handling" column. All those sims listed are using slip angle/ratio based models, i.e., lateral force is a function of slip angle. In that case, every sim shown there is modelling tyre deformation.
Perhaps that column would be better titled "visual tire deformation?" Keep in mind the physics engine can be doing one thing while the rendering system is doing something else...
Perhaps that column would be better titled "visual tire deformation?" Keep in mind the physics engine can be doing one thing while the rendering system is doing something else...
Probably locking the wheels indeed, or one of them at least. Why not run it through one of the LFS lap analyzing programs and check the slip ratios in the areas and times where it occurs?
That's fine. A Ph.D. physicist friend of mine just read this thread and had himself a good laugh, and agreed with what I've been saying. This isn't the first time a student has started a discussion of this nature in the LFS or RSC forums and insisted to know more than everybody else about something they've never done Maybe what you're suggesting is simply that if you don't keep the time step small enough the system will blow up. That doesn't take any analysis to figure out though and can be explained in much simpler terms.
Anyway, we'll see what you come up with
Maybe what you're suggesting is simply that if you don't keep the time step small enough the system will blow up. That doesn't take any analysis to figure out though and can be explained in much simpler terms.Anyway, we'll see what you come up with
Last edited by jtw62074, .
Exactly. It's just a bunch of handwaving at this point.
How about a basic physics math example then? A falling object or something. How does frequency domain and all that fit into the simulated motion of a falling ball or something equally simple?
How about a basic physics math example then? A falling object or something. How does frequency domain and all that fit into the simulated motion of a falling ball or something equally simple?
Last edited by jtw62074, .
FGED GREDG RDFGDR GSFDG