Physics thread

  • Thread starter LVracerGT
  • 1,459 comments
  • 133,956 views
This is another example of something that appears to be a bug, but is actually a feature/simulation of the real deal. There are several of these minute details in the game. Another example is tire pressure: people reported a bug where the tire pressure after coming out of the pit was higher than what they set in the tuning screen. Turned out it wasn't a bug, but a result of the tire blankets heating up the tires, thus increasing pressure (tire pressure in the tuning screen is for cold tires). There's a list of things like this on the physics wiki, but I don't think it's complete.
 
This is another example of something that appears to be a bug, but is actually a feature/simulation of the real deal. There are several of these minute details in the game. Another example is tire pressure: people reported a bug where the tire pressure after coming out of the pit was higher than what they set in the tuning screen. Turned out it wasn't a bug, but a result of the tire blankets heating up the tires, thus increasing pressure (tire pressure in the tuning screen is for cold tires). There's a list of things like this on the physics wiki, but I don't think it's complete.
Additional rolling reseitance due to increased negative camber is definitely a real thing. It is however over exaggerated on GT and LMP tires. In real motorsports you would never run less camber to gain straight line speed. The resulting less mid-corner grip would outweigh the benefits of less rolling resistance. Just like the good old saying of running more pressure at high speed tracks like Monza. This does not give you any noticable advantage in lap time. As you said it isn't a camber bug, it's just the tire model design for those specific tires. I've been told to try camber on a specific Formula tire (which I don't remember) and the results I'm looking for would show.
 
If you feel it's over exaggerated, best report/discuss it on the official forums/ask Doug. I think I remember a discussion about this, but will have to look it up.
 
Additional rolling reseitance due to increased negative camber is definitely a real thing. It is however over exaggerated on GT and LMP tires. In real motorsports you would never run less camber to gain straight line speed. The resulting less mid-corner grip would outweigh the benefits of less rolling resistance. Just like the good old saying of running more pressure at high speed tracks like Monza. This does not give you any noticable advantage in lap time. As you said it isn't a camber bug, it's just the tire model design for those specific tires. I've been told to try camber on a specific Formula tire (which I don't remember) and the results I'm looking for would show.

Yiro Formula B. I'm sure some other formula tires are similar
 
If you feel it's over exaggerated, best report/discuss it on the official forums/ask Doug. I think I remember a discussion about this, but will have to look it up.
I'm not a member and what I said about contact patch over camber thrust is what a dev told my friend who asked questions on my behalf. I'm not sure of his name at this very moment but it was one ofthe guys who made the tires.
 
If you feel it's over exaggerated, best report/discuss it on the official forums/ask Doug. I think I remember a discussion about this, but will have to look it up.

I'm going to report it on the official forum.

OK, so it's just an over exaggerated effect with GT/LMP tyres, on formula tyres it should work fine.

I do not know how much camber GT/LMP cars run in real life. If they indeed run almost no camber, why is the stock setup different in that regard?
 
I did some additional testing. Again on Nürburgring GP, this time with Formula C and Formula B. I used the stock setup (except brake pressure which I raised to 100% and camber, of course), fuel load may vary a little bit.

Formula C:
Lap time with stock camber settings (-3.0° front, -1.9° rear): 1:49.815
Lap time with adjusted camber settings (0.0° front, 0.0° rear): 1:48.306 -1.509

Formula B:
Lap time with stock camber settings (-3.6° front, -1.9° rear): 1:42.118
Lap time with adjusted camber settings (0.0° front, 0.0° rear): 1:41.007 -1.111
Lap time with zero camber in the rear ( -3.6° front, 0.0° rear): 1:41.570 -0.548

So I am afraid this is not an issue only with GT/LMP tyres. It appears with other kinds of cars and tyres, too, and even to a greater extent.
To determine if the problem may be dependent on either front or rear tyres I ran a few laps with Formula B and set only the rear tyres to 0.0° camber. Obviously, the time gained is half as much as with all four tyres set to zero. So the front as well as the rear tyres' impact is pretty much equal and not onesided.

This leads to the conclusion that the camber issue is a general issue with the tyre model/physics in the game and not dependent on front or rear, formula car or prototype as the effect is always there. The magnitude may vary but it is always noticeable and clearly too much.

I will post on the official forum later today. Apart from that it would be nice if someone is kind enough to test the camber effect on some other types of cars in order to confirm (or to disprove) my recent findings. :)
 
Apart from that it would be nice if someone is kind enough to test the camber effect on some other types of cars in order to confirm (or to disprove) my recent findings. :)
Yep,there is problem with camber.
This is why my tune has zero camber.
I found this about two months ago while I tested two @Ridox2JZGTE BMW 1M tunes.
He asked me to share this information in physics thread ,but I refused.

Без име.png
 
He asked me to share this information in physics thread ,but I refused.

View attachment 438246
:lol: But in all seriousness, what were you afraid of? Camber is either working as intended or it's not. SMS should explain either way (or perhaps there are some other people around here that have different experiences, personally I would expect a lot more issues to pop up if it were a global bug).
 
How much camber does the BMW 1M in real life have? It is a street car so I guess there is probably close to zero camber anyway.
Street cars do run very low camber. GT racecars can easily have 2.5 f 1.5 r and then some depending on track. The problem with camber isn't like gt6 where it removed grip. It most certainly does add grip with roll. It's just the whole rolling resistance side of things. You should look at mph as well as your lap times from your testing, you'll be surprised. I'll see what's what when I tune the m3 GT for spa here shortly.
 
How much camber does the BMW 1M have in real life? It is a street car so I guess there is probably close to zero camber anyway.
Ask @Ridox2JZGTE
:lol: But in all seriousness, what were you afraid of? Camber is either working as intended or it's not. SMS should explain either way (or perhaps there are some other people around here that have different experiences, personally I would expect a lot more issues to pop up if it were a global bug).
I had big problems criticizing GT5 physics and since then I am more careful.
 
How much camber does the BMW 1M have in real life? It is a street car so I guess there is probably close to zero camber anyway.

Factory ( OEM alignment spec )

Front Axle: Caster from 6°50' up to 7°50', Toe from 0°02' up to 0°18' and camber from -0°55' up to -1°45'

Rear Axle: Toe from 0°02' up to 0°18' and camber from -1°33' up to -1°58'

This from a considered as safe factory alignment, most 1M owner would add more camber at both axle when taking the 1M at the track.
 
I had big problems criticizing GT5 physics and since then I am more careful.
I understand. Haven't seen that behavior in this part of the forums, though people will ask you to back up any claims. I know most WMD members and certainly all of SMS are very much interested to fix any flaw/issuess we discover, or are quite happy to explain if it's actually intented behavior. 👍 I'll raise the question at WMD, see what I can find out. :)
 
Generally on road cars with street legal tires which are more likely on softer springs/damper suspension setup will need some camber as the car will roll during cornering, depends on how much flex the car goes through, how much grip the tire has, and how rigid the car overall. I think people who put slick on road cars while still having softer suspension/chassis will need more camber than slick tire on car with super stiff suspension/chassis setup.

Some say the rake of the car and caster also have an effect on how much front tires camber is optimum. Higher caster adds more dynamic camber when turning, and most Pcars cars are set with high caster. Back in the old days of Group A cars, one example, the Supra MK3 MA70 had very high caster similar to the E30 M3 Gr A, at about 10 degrees, most setup were never more than 3 degree camber on the front wheels. The Supra also had super stiff springs at front at over than 70kg/mm set by TRD :eek:, the Aussie team who ran the MA70 used lower springs when raced at Bathurst, but still ball breaking 26+kg/mm at the front with half the caster set by TRD, and they reportedly never run more than 3 degree camber.
 
Last edited:
Just to throw this fact in the discussion. Project CARS has rigid chassis modelling, so there is no flex of the chassis that might induce some roll.
Also the only think I can think of that might be 3 things. Spring rate of rubber compound or carcass being too low, causing increased rolling resistance (beyond normal/realistic spec) during straightline driving. More likely is the lack of camber thrust at the edge of grip, the grip type that you hang on during drifting as well.

Apart from that I do not have any idea at this point in time why camber might be working incorrectly.
 
I'm certainly not a computer/programmer guy, but in my opinion it seems to simply be the contact patch deformation from negative camber (as Casey described) is adding too much drag. I know the coding side probably isn't that simple. Unfortunately all we can do is wait for a dev to hopefully clarify the situation.
 
http://forum.projectcarsgame.com/sh...er-questions&p=1103598&viewfull=1#post1103598

"Camber sensitivity in our models isn't what i would like either. The race tires and most modern radials, should make more and more grip as camber goes up with the downside being more inside heat and more wear. So you find the compromise for you situation, race,length ets.. But admittedly this blancing act dosen't exsit. Not sure it its more code work AJ needs to do or if Casey and myself need a different approach with the paramters AJ has given us. I have tried to make more camber sensitive tires with but the results were not that great, and in the end a tire that works and feels right always wins out. More to investigate in pcars 2."

-Doug Arnao

I'm getting the best results with the Yiro slick on the Formula B with camber. Is it the only tire in the game like that? Perhaps.
 
Last edited:
I have tried to make more camber sensitive tires with but the results were not that great, and in the end a tire that works and feels right always wins out. More to investigate in pcars 2."

-Doug Arnao

More to investigate in Gran Turismo 7 and Live for Speed new tire model next year :D:D:D:D:D:D
 
Last edited:
http://forum.projectcarsgame.com/sh...er-questions&p=1103598&viewfull=1#post1103598

"Camber sensitivity in our models isn't what i would like either. The race tires and most modern radials, should make more and more grip as camber goes up with the downside being more inside heat and more wear. So you find the compromise for you situation, race,length ets.. But admittedly this blancing act dosen't exsit. Not sure it its more code work AJ needs to do or if Casey and myself need a different approach with the paramters AJ has given us. I have tried to make more camber sensitive tires with but the results were not that great, and in the end a tire that works and feels right always wins out. More to investigate in pcars 2."

-Doug Arnao

I'm getting the best results with the Yiro slick on the Formula B with camber. Is it the only tire in the game like that? Perhaps.
Now we officially know it's not a bug. Just a byproduct of the tire model/coding.
 
http://forum.projectcarsgame.com/sh...er-questions&p=1103598&viewfull=1#post1103598

"Camber sensitivity in our models isn't what i would like either. The race tires and most modern radials, should make more and more grip as camber goes up with the downside being more inside heat and more wear. So you find the compromise for you situation, race,length ets.. But admittedly this blancing act dosen't exsit. Not sure it its more code work AJ needs to do or if Casey and myself need a different approach with the paramters AJ has given us. I have tried to make more camber sensitive tires with but the results were not that great, and in the end a tire that works and feels right always wins out. More to investigate in pcars 2."

-Doug Arnao

Great to see one of the devs responding to the questions.

With the very complex tyre model of the game it is obviously very hard to get every aspect right. But highlighting his last sentence, it is a little letdown for me if we are not going to see any improvement regarding this issue in Project CARS 1 at all.

I'm getting the best results with the Yiro slick on the Formula B with camber. Is it the only tire in the game like that? Perhaps.

Might be worth a try. IIRC, I was using the Faretti Soft on the Formula B.
 
Last edited:
I've been waiting to test PCars physics against a RL car. So I decided to give PCars another try today with the Old vs. New DLC pack so that I can try the Yellowbird. Since I do motorsports in 2 cars with a chassis and suspension very nearly identical to the Yellowbird, this is a chassis I know.

And there is still almost ZERO lift-off oversteer. I can do things in the YB in PCars at that I absolutely guarantee you would get you killed - things I don't even do in my RL cars at motorsports events unless I have a huge run-off area for safety, because even though I'm VERY good at lift-off oversteer in these chassis, that one little screw-up could cost me big in the wrong place.

The YB in PCars simply feels NOTHING like it should. I don't know how else to explain it. The balance is all wrong. It refuses to oversteer off-throttle but has a very classic "light-rear-end" FR-style oversteer when you get on throttle too much. I feel nothing of the nimble front-end, nor of the massive rear-grip these cars possess IRL. Honestly, it reminds me a LOT of the crappy physics in GT5 & 6 (in which RR cars or more-or-less completely borked). I'm not a fanboy of any product, period, and even though I like a LOT of things about PCars, it's incredibly disappointing to find out that their simulated physics don't match real-life in this case at all. It's just as wrong as wrong can be.

Very, very disappointed.

(If you want to feel what a real Porsche of that chassis type feels like, try AC, which models it brilliantly, or go clear back to GT4, which actually does a very good job of it also.)

And while I didn't post this on the PCars board, I can post it here: As far as I can now tell, PCars physics are crap, like GT5 and GT6. They've done a very good job of talking the talk and adding some great weather effects and gameplay, but if they can't get something as fundamental and basic as rear-engine physics correct then they either don't know what they are doing or don't have a physics engine capable of doing what it should be doing. Either way, it's a crushing disappointment.

Go play AC, it's got the physics down so good it's scary. (Now all they have to do is wrap a better game around their fantastic physics and FFB.)
 
I've been waiting to test PCars physics against a RL car. So I decided to give PCars another try today with the Old vs. New DLC pack so that I can try the Yellowbird. Since I do motorsports in 2 cars with a chassis and suspension very nearly identical to the Yellowbird, this is a chassis I know.

And there is still almost ZERO lift-off oversteer. I can do things in the YB in PCars at that I absolutely guarantee you would get you killed - things I don't even do in my RL cars at motorsports events unless I have a huge run-off area for safety, because even though I'm VERY good at lift-off oversteer in these chassis, that one little screw-up could cost me big in the wrong place.

The YB in PCars simply feels NOTHING like it should. I don't know how else to explain it. The balance is all wrong. It refuses to oversteer off-throttle but has a very classic "light-rear-end" FR-style oversteer when you get on throttle too much. I feel nothing of the nimble front-end, nor of the massive rear-grip these cars possess IRL. Honestly, it reminds me a LOT of the crappy physics in GT5 & 6 (in which RR cars or more-or-less completely borked). I'm not a fanboy of any product, period, and even though I like a LOT of things about PCars, it's incredibly disappointing to find out that their simulated physics don't match real-life in this case at all. It's just as wrong as wrong can be.

Very, very disappointed.

(If you want to feel what a real Porsche of that chassis type feels like, try AC, which models it brilliantly, or go clear back to GT4, which actually does a very good job of it also.)

And while I didn't post this on the PCars board, I can post it here: As far as I can now tell, PCars physics are crap, like GT5 and GT6. They've done a very good job of talking the talk and adding some great weather effects and gameplay, but if they can't get something as fundamental and basic as rear-engine physics correct then they either don't know what they are doing or don't have a physics engine capable of doing what it should be doing. Either way, it's a crushing disappointment.

Go play AC, it's got the physics down so good it's scary. (Now all they have to do is wrap a better game around their fantastic physics and FFB.)
I remember you bringing up lift off oversteer months ago and everyone said they felt it except you. It seems like something on your end is wrong weather than the game. Even 50/50 cars in this game can be made to have ridiculous lift off oversteer. I genuinely don't know where your coming from here. Maybe you can record your gameplay and post it.
 
http://forum.projectcarsgame.com/sh...er-questions&p=1103598&viewfull=1#post1103598

"Camber sensitivity in our models isn't what i would like either. The race tires and most modern radials, should make more and more grip as camber goes up with the downside being more inside heat and more wear. So you find the compromise for you situation, race,length ets.. But admittedly this blancing act dosen't exsit. Not sure it its more code work AJ needs to do or if Casey and myself need a different approach with the paramters AJ has given us. I have tried to make more camber sensitive tires with but the results were not that great, and in the end a tire that works and feels right always wins out. More to investigate in pcars 2."

-Doug Arnao

I'm getting the best results with the Yiro slick on the Formula B with camber. Is it the only tire in the game like that? Perhaps.
Nice that they communicated with the truth, but still disappointing nonetheless. I haven't played much since summer and golfing season hit, but in my early tuning efforts I found that increasing camber had very little, if any effect on tire heat like I thought it was supposed to. One of the things I looked forward to the most was this "balancing act" between camber/grip/heat/traction/braking and to find it really doesn't exist and may not until PCars 2...well...:(
 
Interesting @Johnnypenso, because I've found that it does work. Lowering the camber results in lower tyre temperatures for me, so I have to adjust camber and pressure together. I've also found that a small amount of camber for a twisty circuit will give me better lap times than running at zero camber.

I'm still learning the finer details of car setup, but so far everything I've tried works as it should.
 
Interesting @Johnnypenso, because I've found that it does work. Lowering the camber results in lower tyre temperatures for me, so I have to adjust camber and pressure together. I've also found that a small amount of camber for a twisty circuit will give me better lap times than running at zero camber.

I'm still learning the finer details of car setup, but so far everything I've tried works as it should.
It's not that camber doesn't work. It just adds a lot of rolling resistance. Like 2-3mph for 1 degree of rear camber.
 
The YB in PCars simply feels NOTHING like it should. I don't know how else to explain it. The balance is all wrong. It refuses to oversteer off-throttle but has a very classic "light-rear-end" FR-style oversteer when you get on throttle too much.
Here's something interesting that you can try:
Take this tune put in front Masculin All Season tires,rear Masculin Street-Medium tires and enjoy lift off oversteer in mid corner and big grip under acceleration,you will only have to imagine that you are driving Porsche :lol: (This is not joke,try it)
 
Back