What I believe to be wrong with the RF2 tire model / grip levels / lack of control

Probably supposed to represent street legal high performance sports tyres rather than your average Nokian summer tyres.

 

I found these:
Toyo R888 4.76mm Semi slick racing tire.
Michelin Super sport 7.9mm
Michelin pilot sport 3 6.75-7.54mm varies quite a bit with sizes
Michelin Pilot Sport Cup 4.76mm same as Toyo as tire is pretty close to semi slick tire intended for racing, it has hybrid of semi slick and sport tire thread.

I use Continental, Nokian will not last as long.

Of course only ISI will know for sure what kind of tire they have wanted to simulate, but I would think that it is racing tire, they usually have racing tires in their vehicles, even ZR from rF1 which had 'street' tires had them more like semi slicks than street tires. Of course even sport tire like pilot sport 3 will not take long for heavy abuse, so as they make their vehicles to be race cars used for racing, it does make sense to use something better than typical sport tires. But would be nice to see how they would make V speed rated common street tire too, preferably 185-60-14 size

 

The rTrainer is based on the Formula Skip Barber right? So I think the rTrainer tires should be similar to Michelin/BFGoodrich g-Force Sport (fully threaded road tires):
http://www.tirereview.com/Article/6...p_dealers_make_realworld_recommendations.aspx

 

Thx, that looks similar to my poor eyes at least, for that tire I found 7.9mm thread depth:
http://www.bfgoodrichtires.com/tire-selector/category/sports-performance-tires/g-force-sport-comp-2/tire-details#

ISI perhaps made tires that are run in, I remember reading how in lower class track racing they did use some machine to run in the tires, it takes bit of thread out and makes rubber to handle better of that abuse, so that pieces are not torn out from thread. Don't know exact english words for that.

Maybe that is why thread depth is less.

Still don't know about traction coeffs, I need to play with them bit more. I know that Toyo AR1 had CoF of 1.452 measured, but how that relates to rF2 parameters can be of course bit different from 1:1.

 

I have got to say after driving the GTR for a few days I think its great fun, especially at Mid Ohio. I have been doing a few 10 lap races against a full field of AI, with them at 120% 100% aggression and 100% damage and have had some great races, would love to race online but I haven't seen anyone driving the GT cars when I have checked. Maybe most are too scared of them after reading this thread with some people saying they are so terrible to drive.

The AI are still a little slow here in spots and I noticed the more rubber that goes down the more trouble they tend to have but they do run laps around 1:15-1:19 so its still a decent pace. They are much slower in traffic but one on one they can do low 1:17s as long as your not hounding them too much where they get a little defensive. They just brake a little early into a couple of corners but surprisingly have dive bombed me on the inside in the same corners a few times.

Edit I changed the video as I noticed I had my old setup assigned and this video show the tire model pretty well I think. Should be encoded in a few minutes.

Here is a video of just a 5 lap race with AI at 120% and 100% aggression, I ran out on to the marbles on lap one and got passed by 5 cars, also a few pretty big corrections. Many have different opinions to mine but even this early beta to me feels a long way ahead of any other sim Ive ever driven. Grip levels are fine and I dont have any lack of control so I think the tire model is going in a great direction.

 

http://www.mazdamotorsports.com/wcsstore/B2BDirect/Motorsports/skipbarber.htm

They're not only run on treaded tyres.

 

Well, at 100% aggression I'd want a bit of diving. That's what some humans, do after all, isn't it? However, it seemed a couple of times they backed out when they had already just started heading for the apex and you weren't along side yet. At 100%, I'd expect them to stand up for their rights and take the racing line. And then hurl abuse at you for diving and spinning them out. Maybe some do, some don't(?)

 

Interesting discussion from the start - even if it gets a bit OT from time to time. Much the same thing was said about rF1. Basically if a standard car got a bit out of shape then all was lost. Having done a bit of work with rF1 tyres I can offer a couple of observations.

1: From a theoretical perspective, the tyre slip curve in the first post has three regions.
- A region of positive stability - the rising section, in which grip increases with increasing slip angle.
- A transition region
- A region of negative stability, in which grip falls as slip angle increases.

If this is all that is controlling the car behaviour then once the tyre reaches the third region, "all is lost" and the car is out of control. One has to observe that this simplistic model does not apply on its own in Real Life, as real cars do not (always) behave like that.

All sorts of things might apply. Lifting off can cause the car to slow, and that may reduce speed enough that the car comes under control as the forces fall with the speed. Self-centering torque as a consequence of proper steering geometry and the various trails in the tyre-force locations also contributed.

As well as that, at any given time the lateral forces may not be at the same region on the curve as longitudinal forces, so one may be in the stable section after the other has reached instability. The DropOff Factor in rF1 tyres gave an opportunity to play with that interaction. We found that choosing to have the longitudinal forces stretched out more than the lateral forces gave us a more drivable car.

2: The amount of time a driver has to respond is also important. The contributor to this thread who observed that he was a better driver because he learned to drive on ice is on the money. He has developed better skills for responding rapidly to the signals of a car about to spin, so he has a better result.

The tyres contribute to this through the shape of the slip curve as it shifts from positive stability to negative stability. The sharper is that transition then the sharper the response. And the overall amount of grip has an effect as well. For the same shape of slip curve, high-grip tyres are apparently more sudden to respond than lower grip tyres. Which is why "older" cars with less grip are easier to recover - you have more time.

3: The car itself has a role. The apparent suddenness of the response of a car is dependent on the polar moment of inertia about the vertical axis and that is (roughly) proportional to the square of the length. For example, short mid-engined cars are harder to drive fast than NASCAR stock cars. Go-karts are more challenging again, which is why they are a good indicator of later driver success.

4: I wonder whether the slip-curve lookup tables in rF1 went far enough. The doco in the tyre file sort of implied that the horizontal axis of the curve needed to go to around 2 to cater for some circumstances. But on the road the cars sometimes seemed to have no grip at all, as if the slip calculation had suddenly gone to zero grip because it ran out of entries in the table. Wheelspin can produce conditions which are outside the range of the standard rF1 slip curve tables. Pushing the range out to 4 or more improved standing-start behaviour among other things.

The new tyre model may have rendered all this redundant...

 

Good post !

I think the key thing is that Its fine having harder to drive cars or cars that are far sharper but as it is in rf2 No matter the skill level of the driver the cars simply don't handle like real cars when the rear wheels spin up or when the tires get to the limits of grip.

I presume that's why they are still working on the contact patch.

"The DropOff Factor in rF1 tyres gave an opportunity to play with that interaction. We found that choosing to have the longitudinal forces stretched out more than the lateral forces gave us a more drivable car."

The key is to try and get that Drivable car ( in that it behaves in a progressive way like a real car) whilst still having a car that's a beast and wild , If you look at Game stock car and some of the better RF1 mods all the cars that have a more realistic stability to them lose some aspect of how lively the real car might be in some specific situations.

Maybe all the models are overlooking some aspect of simulation I don't think any simulator has really got things spot on for a good range of fast modern cars.

It will be intresting to see what Assetto Corsa is like with laser scanned tracks As it certainly is the case that the tracks in NKP are way to flat making it hard to tell if the cars are overly passive or if that's a result of the road surface.

RBR certainly had very lively cars off road and offerd fantastic car stability but then those cars were always in a slide and off road in a 4wheel drive is not comparable to a RWD race car.

Mind you on the road stages RBR feels very much like NKP.

 

No comments about my test? Jameswesty, you requested me a video, and I gave you that, and a replay, with a GTR handled near limit sliding with a race setup in Mid Ohio... and I have no response.

I have a lot of replays with most of cars of rFactor 2 going near limit, with corrections, saves, microsliding. Some cars more than others, and Historics are easy to slide without crash (is awesome to see a replay in Spa with BT20, fast chicanes scares me), but with FR3.5 you can do it like with GTR. The only car that I need to test thoroughly is the ISI Formula Master, and I think I will get the same results.

 

Your video demonstrates the exact issue with RF2 with the grippy cars and that is the binary nature of the slides.


Sure you can catch the car and get a bit of angle that's never been the issue

Its the depth and ability of being able to control the angle that is very limited. When racing fast in RF2 say for a proper event all you will ever do is snap the car back as fast as possible you don't ever chose between having a bit of angle for a specific line at a comprise to tire ware or a to achieve a certain exit point.

In a real car the transition from no slide to slide back to no slide should be near seemless , Granted it might happen very fast and in a very aggressive way but the point is the seamless nature which is totally lacking from RF2.

Have you played NKP ?

Like others have said Some aspects of NKP are to smooth with not enough track detail but at least the point at which cars gain and lose grip is essentially seamless and this is what makes it so compelling and addictive to drive.

Game stock car is also far closer to this along with RBR.

I want RF2 to do something better than both RF1 and NKP and It will be amazing if the final tire model produces something as progressive as NKP but as tight as RF2

After all I have no doubt that RF2 will always do things like tire ware tire temps and how this affects the car WAY WAY WAY WAY better than any other simulator on the market.

However that is all useless from the point of a simracer unless the core handling of the cars is right.

 

Wow, what a good graphics. I run almost everything maxed out at 1920X1080 but doesn´t look that beautiful. Could you put yor system specs and settings on youtube?

 

You should know that GTR is a GT1, stiffnes and with slick tires, I don't know how It feels in real life (I think you not like me), but I think is normal to see that behaviour. Fast slides and some long microslides... you will see that in real GT1 races, they are not drifting. Only sometimes in corner exits some slides but nothing exagerate that you can do in NKPro.

If tyremodel is too wrong like hiohaa says, you couldn't drift with other cars like 370Z, all historic cars, rTrainer... but you can, I said, and repeat, you can literally dance with Historic Formulas without that binary behaviour you say... are completly different cars with different tyres, suspensions, weight and behaviour.

Show me a video of some GT1 real car saving an slide with more angle than me in my video, or FR3.5... I'm waiting a drifting Osella in a downhill like video I showed you about NKPro too... I can´t find It (only some drivers losing grip in corner exits).

GT1, little grip lost, and going to grass (3:11), I don't see much more slides, drifting or epic saves... only some like with GTR in rFactor 2, see in 5:32. And if you want more, you can keep seeing the video, and wait to some onboard with GTR, It looks very stable and gripy car :

I don't want to flood this post with videos, there are lots of them, entire races of GT1 if you want... you can search videos about GT1 cars, and see how It looks like, and how some drivers can lost the car suddenly without chance of recovery when he reach some limits.

 

oh look, its a pay driver backmarker in a minardi, catching some oversteer and POWERING through it.

can do this in netkar pro, can't do this in rf2.

 

LOL, don't you understand you have to compare the same cars?! oh... and in my replay I have one or two saves like that (not too elegant, GTR is a GT1, not a 1999 F1...), with FR3.5 you can do that with very, very good ands, and with very low downforce setup. And with historical cars... Is easy to do in each corner ...

I'm not saying is impossible to do a epic drift with GTs, or other cars... but is very difficult to see something like that in every corner like in NKPro, is a lucky move, a reflect act in the exact moment. You can do that in rF, but you have to be very lucky, like in real life.

I'm still waiting to see a lot of videos of drivers drifting easily with GT1s, FR3.5...

Edit: You can compare other cars... try to drive Brabham BT20 (fast laps, trying non sliding), save the replay, and compare with a real race of these cars, yo can see how much It looks like real cars, dancing I said not on off grip

 

Historical cars are using bias ply tires...

Obviously it's not easy to find a lot of videos of drivers drifting easily with GT1s or Formula cars, simply because it's slower and you can damage the tires. But I have see Alonso drifting into Barcelona pitlane entry easily, lots of times (and different years... 2008, 2009 and 2010), I don't really need videos

 

In the gt1 zolder video you linked above you can literally see that the driver is drifting out of allot of the corners only its not a crappy RF2 save the back end drift its just simply how cars drive in a 4 wheel drift with the driver controlling the power.

In RF2 if you drive anything other than all 4 wheels fully locked in then the simulation falls apart and it just becomes a case of "saving it" rather than "driving it"

Its very subtle but most race drivers in most cars will be doing a slight drift as they power out of corners if they wanted to they could transition into a full on drift or do whatever they wanted to.

RF2 general its just a case of the back end coming out or just catching the car because it goes wild.

RF2 as it is totally lacks the subtle characteristics of race driving.

Maybe how it is now is good enough for you, it probably is for allot of people.

you can drive the car like an idiot and collect the slides with 1 hand

It does not require that much skill the problem is that its litraly JUST A CASE OF COLLECTING THE SLIDES there is no depth to the way the cars lose grip or handle when getting into a slide or sliding.

 

You can enter "crossed" in pits with any car if you have hands ... a matter of dosing the throttle. You can do donuts if you want, or heat rear tyres sliding tires without spin. We are talking about grip in limit conditions, when you lose rear in fast corners going to wall. You can se a little example in the video I posted. Or when you apply too much gas in corner exits... videos are there, I said, entire races of GT1 driving in the limit, sometimes looks like rFactor, suddenly, grip off and wall few times I saw a save when a GT1 reach certain angle.

Edit: this post was a response to Domi