Wich S397/ISI cars have REAL AND ACCURATE data VS real car in real life

Hi everyone, Hi S397.

The ISI /S397 car content is exceptional (thank's S397/ISI). For me that is the content which is really at the "standards 2017".

Question for S397 or/and rf2 users which are connoisseurs (more informed) :

In which cars is there REAL AND ACCURATE data VS the real car ?

I know for GT3, I don't know for others ? Perhaps the radical because the is a license... I don't know.

I think it is one of the most important thing for realism in a 2017 racing simulation, + improvement of physics (and if it is possible + laser scanned track with real and accurate datas).

Thank you

 

Try them all and give us your opinion. You seem to be an expert on that. I prepared the chart for you.

McLaren 650s GT3
Physics: ***** FFB:*****

Mercedes AMG GT3
Physics: ***** FFB:*****

Bentley Continental GT3
Physics: ***** FFB:*****

Radical SR8 GT3
Physics: ***** FFB:*****

Corvette Callaway GT3
Physics: ***** FFB:*****

Radical SR3 (use RH version)
Physics: ***** FFB:*****

USF2000
Physics: ***** FFB:*****

Honda Civic (avoid M tires)
Physics: ***** FFB:*****

Renault FR3.5
Physics: ***** FFB:*****

Formula ISI
Physics: ***** FFB:*****

AC Cobra
Physics: ***** FFB:*****

Corvette C7
Physics: ***** FFB:*****

Formula 2
Physics: ***** FFB:*****

Nissan GTR
Physics: ***** FFB:*****

Nissan 350Z
Physics: ***** FFB:*****

Renault Clio Cup
Physics: ***** FFB:*****

Renault Megane
Physics: ***** FFB:*****

 

Why are you so sarcastic ?? My question is very interesting I think.... you seems to have a personnal problem with me, if it is the case just stop to answer to my posts. It would be more clever.

EDIT :
And I absolutely not a expert, I never said that !!!!!!

 

The only expert on what data is used is S397, and I don't think you will ever get an exact answer from them, because it's all under NDA from the manufacturer. The only thing sims typically tell us is that "we got the most accurate data available for this car bla bla bla", which is just marketing and means nothing. The real details, they can't share with us.

 

As I understand it, it's a simulator. They acquire real-world data with manufacturers and/or real racing teams, and input such data wherever it must go
.
And contrary to other simulators, the data input in rFactor 2 default cars is correct and realistic, no need to fudge with the numbers so that the physics feels right and users don't complain about broken cars.

-----------------------------------------------------------------------

However, when talking about MOD's, it's a different story. They may have way less access to real-world data when it comes to the precise parts like tyres, and some do what is called "balance" which also exists in real life, but they do it with unreal numbers. In real life, BoP (balance of performance) happens by many forms, like adding or removing weight, giving teams authorization to use larger fuel injectors, etc.

Some MOD's get a few cars really close to the real-world counterparts, like the ApexModding GT3 Ferrari 458 and Z4, but other cars from the same MOD are either broken or are way slower than the rest of the cars, which is unrealistic since in most competitions the cars from same class must perform very close to each other. In the same example, the Apex MOD has the C6R which is capable of doing 2:15:1 on Spa (unrealistic), while the Z4 does ~2:18 (real-world lap time) and the Ferrari 2:16. The SLS is completely broken, which is why almost nobody uses it for online or offline races.
So in this particular MOD winning or losing is a matter of choosing the right (or wrong) car. To me, that's really bad.

Other MOD's, like the Simtek GT3 World Series, have really consistent cars that perform really close to each other, which is a good thing. However, they used some "non-real data to balance for online racing". It's a very good MOD nevertheless, really worth checking it out.

Then there are the really good and almost perfect MOD's like the GP3 Series and thePalatov D4, which as I understand both had real data from real-world teams. In fact, the Palatov was actually made and released by the racing team (talked about here in the forums) and had help from ISI to work on the tyres.

 

This is really a hard question to answer and I was seeking for an answer since 2012. What I know is, the simulation model is good enough. There are papers showing that even rF1 simulation model is good (I still search the paper of an racing engineer comparing rF1 output to real messurements with great results when the car is modeled accurately, I will post the link when i found it) . It really depends on how accurate cars are modeled, but I trust that ISI and S397 do a great job in modeling realistic.

And then there is advertisment. E.g. Sony writes everywhere: "the most realistic racing simulator" and posting interviews with real drivers saying that. We all know this is not true, but most of Sony GT users probably believe in this (and i did too when i was a Sony GT user 10 years ago).

It is a difficult question probably nobody can answer, except ISI/S397! But then how do you know it is not false advertisement like Sony does? But sometimes there are indications when people own simulated cars. I've read a post from an AC cobra owner comparing ACs and rF2s model and he said, out of my memory (sorry no link), that rF2s model feels exactly like his AC Cobra does.

EDIT: This is the link i was seeking:
https://drracing.wordpress.com/2014/01/03/how-close-is-close-enough/
Some interesting reading

 

I think they used real data for the mods. But to use real data doesn't mean to have something close to reality. These datas needs to work in a finctional world so it doesn't mean that we can expect absolute fidelty with the real world. Expecially in a hardcore physic engine...it could be more easy to have more similitudes using a less powerful physic engine.

 

I didn't own an AC Cobra but something a little simular to a lesser degree in weight/horse power feel. A V8 Vega. About 300 horse 2500 pounds. I can testify to the craziness of the rF2's Cobra for sure.

I never felt safe in my V8 Vega from the second of turning the key. There was torque steer right away just like the rF2 Cobra. I remember twisting a finger backwards with my Vega the same way I do in rF2's Cobra. 100 miles an hour plus are very simular feel of heavy wheel to nothing at all.

And then the new GT3's came out and blew me away. I don't think they are more realistic exactly but have more physics detail over all. I think the Cobra could have more of this. All the cars really and I would pay a lot for it.

And then throw in all your hardware and setting preferences and only god knows what we all have individually.

I would like to drive a racing teams setup is all I can say for sure!

 

Me too. And then compare rF2's lap times and video with the real car the team used <3

I guess if you drive the Palatov you're driving their race setup? The GP3 as well?

 

In case you haven't figured it out, setups are dynamic and adjusted to suit driving style and weather and track grip. There is no such thing as "a race setup", only "a base setup" plus changes to match conditions.

 

My understanding is that he would like to race with the setups used by one or more real-world racing teams. I'd like that too, to see what the driver likes, how the real teams setup their cars and why, feeling the difference from my setups, etc.

 

I doubt real setups would work that well in rF2. I know that at least ISI used the baseline setup provided by the real team for some of their cars, there was some screenshot showing FR 3.5 setup page in rF2 had many of the same values as in real setup sheet of that car. But still that same setup did not work very well in rF2...

 

It's not really news to anyone, that both ISI and S397 have used real data as reference for all their cars. How well it all translates in the sim is highly dependant on the physics evolution of the sim. Many older cars could prob. use a revision.

 

A year or so ago someone was accusing Assetto Corsa of false data. I guess they had come to the same conclusion and just made it work best they could, which I can appreciate.

I think this line of thought started with my comment about sitting in a race teams sim setup that had been tuned as close as they could get it to their real cars. I didn't mean all data matching.

 

People (generally... some here much less so) have the wrong impression about how accurately things can be simulated. The games/sim industry as a whole has itself to blame for this to some extent, because each iteration of a game (or part thereof) is said to be the best and most realistic ever. I suppose it's bad marketing to say that you've done your best but had to fudge a few figures to make it playable.

 

You are right there, tires in RF2 as in all sims are the more difficult thing to simulate. For most parameters of a car you can validate them against real life data, for tires many parameters are guessed, missing, not taken into account, then by trial and error adjusted and impossible to validate against a real tire.

In RF2 we have fantasy tires, very complex and sophisticated fantasy tires, in some cases this fantasy tire may behave close enough to what we think is reality for us to have fun.

I fly RC planes, old timers scoff at some of us that train on sims, saying the models and the sim are not accurate enough (they are right on this point).
But what is also true, is that an experienced pilot who tries a good RC sim, flies without problem after a short adaptation time. New pilots that train before on a sim, have a faster learning curve, and crash less planes in doing so. Some of the best aerobatic pilots in our club train on sims, even if they are not perfect.

Cheers.

 

That's interesting to hear. All of our pilots (some are in their 80's at our club) all agree on sim training. It's the first piece of advice we give to new pilots and I don't remember in the past 10 years hearing anyone ever disputing using a sim to train on.

But I think model airplane sims are easier to code. Cars are much more complicated for a simulator. One thing for a model plane, real or simulation you still just have the controller in your hands. No FFB to worry about.

One thing I have heard more than once from real pilots is the model is harder to fly then the real plane. Probably because most models have way more power with a fraction of the weight and then you have to think backwards when the plane is coming back.

Getting back to car sims I think when we look back from 2027 the new sims will have even more physics detail over all. That's what makes rF2 and now PC2 feel so good. Now S397's GT3's come out with even more physics detail. I'm amazed at how much the GT3's feel simular in rF2 and PC2. But it takes a few laps to get used to AC's physics again. It must be the missing tire flex physics?

I think SMS took a hard look at rF2?

 

If GT3 drive similarly in RF2 and PC2, may be it is because both have reached a good enough level of realism.
Racing tires are very stiff laterally (may be not so much for GT3 tires), and don't flex much under load.
I fear it is easier and cheaper to have automotive engineers help you refine your physics model than have software engineers bring your outdated graphics engine to parity with the bests.
SMS having one the best graphics in racing sims choose for PC2 to improve the main flaw of PC1, car physics.

Cheers.

 

I think tire flex physics translates to more of a jerk in the wheel, which I like opposed to just a light wheel. Also there is less to work with in the absence of tire flex physics.

I think I understand what I like about rF2 now. When the tire rolls over the physics have more traction because it doesn't just let go the way it does in AC so Kunos had to increase the over all traction to make it work. And without the pull on the wheel as the tire folds over the same as a real car there is just a light wheel forcing you to drive more by memory instead of reaction.

I guess that's why older drivers with more experience like rF2 more? Which would also explain the immature behavior over there as well.

So I would say rF2 over all is most realistic, which I've heard real race car drivers say in the past.

I think the new GT3's are most real closely followed by the 917K but I have not drove any of these cars in real life.

Be nice to hear from a real driver on a race team.

 

Here you are for the well regarded USF2000. https://forum.studio-397.com/index.php?threads/usf2000-baseline-setup.58167/
I know that a former GT3 Europe Champion tested a lot the URD C7R GTE (video on YT) before it's release.
Cheers.