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

Every word you wrote was like you are reading my mind, especially the 2nd paragraph

Thats getting back to the lack of information/feel/ffb/visual cues. Yes I completely understand lack of information problems . I am talking about pure physics of how the car reacts and acts based on what is happening, based on your throttle, braking and steering inputs based on what is going on the the car, its yaw, pitch, roll, what the tyres are doing, how all this is affected everything and affeting what the car is doing in game comparaed to what it SHOULD be doing if the exact same thing was happening in real life.

But yes, there is lack of information, some people seem to feel more info in rF2, some seem to feel more in NetKar Pro. That is down to how all of our different brains and bodies work , but this argument keeps jumping back and forth from physics/handling characteristics in certain situations to FFB/information the sim translates to the user, 2 different subjects in my opinion.

In defense to James, I believe he is saying all this as a form of helping out the developers of rFactor 2 (ISI) in order to make a better game, and definetely not as a way of "flaming" the product. For this we should greatly appreciate his thoughts, opinions and suggestions.

 

I am sure James is just trying to get rF2 to feel like he wants so it will be better in his mind than NKP. I can respect that, but still, I can have a difference of opinion on where the sim should go

 

Though this conversation has been very interesting with the back and forth, I find I must leave for now, Will read more and follow up tomorrow. TTFN

 

You missed the point of what I said and that is that ultimately its fake regardless of if you put real tire data into it or other data its still abstract.

Another point is that the tire data might be perfect but who is to say which part of the maths is at issue. Or if ultimately a dynamic active model is better than a model using lookup tables ( who is to say that the data for the lookup tables is generated right !!! isi needs to make a documentary of there whole development process, though physics programming is way way beyond me I just nod and smile when coders explain the math behind physics to me)

Its the end result that's important, its all very well having the most consistent or measured data in the simulation if the output does not match that of real life then the data put into the system is meaningless.

I don't think anyone knows for sure where the future of real time simulation will lead or what the best method to use is for this current generation of hardware is.

RF2 does seem to be getting progressively better mind you and I'm dying to see what AC is like then of course who knows Pcars might actually put out something with half decent physics.

 

This is how I felt when I tried netkar recently, if I were to do things I can do in NKP, at real track and car, then I would end up fence real fast.

Different cars are different things, there are no general rule about them, I'm creating same car for rF2 that I have used on track and there is lot more similarities between the two that there was with rF1. In reality, rear end does bite and you will get snake trailing or whatever that is called in english when tail is going left to right, only way to avoid that is overpower it, but one needs enough engine power to do that, if one lifts, well it is spin, push down clutch and hope best is easiest way to regain control.

For me it does work very similarily in rF2.

For very tight hairpins, I think there might be something to improve.

If GT1 car would be easily moddable, I would be tempted to put new engine model type engine file to it and compare what happens, I suspect that hairpin issue would be lot less then as that is what did happen with Firebird when I moved to new engine model, low throttle wheelspin changed a lot, much more natural.

I did remember that I have fraps 2.1, it works much better than new one (like with everything, old beats new?), but fails to lock fps for recording so I get bad sound issues when fps drops below 60. Also I end up 500MB video for a lap and 0.5M internet is not good for uploading such, need to test editing or something. But with that GT4 car I can exit corners sliding at full throttle, better driver certainly much better.

 

Yes and you as a user should not be judging the end result in terms of realism, which one is more realistic in terms of handling, car reactions, the way its driven, how the car in game reacts from all the millions of variables that affect a cars handling characteristics millions of times a second etc. by on how good it feels in your hands. Thats like saying a car is faster than another car because its louder or sounds better. Sorry it doesnt work that way and its wayyyyyyyyyyyyyyyyy deeper and more complex than by going by your hands' feeling.

Well you can create a car in Net Kar Pro and give it crap unrealistic physics, itll still feel good in your hands because of the awesome ffb software of NKP. So the car feels fabulous because of the great ffb system of the game and how it translates to the user, but the car is still completely all messed up in how it acts and reacts to all the millions of variables that make cars act in certain ways. But to you it will feel fabulous because the FFB is so great so now you will be "blinded" by that great FFB, that great controllable feeling in your hands and think the car is great, but really its all messed up physics wise.

If a sim was more realistic, but the other one did a better job on making the feel and control better, I would choose the one that is more real, not the one that has better ffb because the guys came up with a better ffb software. I want the car to be acting and reacting as close as it does it real life, than if my wheel sucks, or the developer didnt do the greatest job with ffb or the visual cues then so be it, everyone will have a different opinion on which feels better, in the end I want the most realistic driving simulation.

If God himself came down from heaven and made the most perfect physics/vehicle dynamics/tyre model etc etc etc for the game, but said sorry there will be no FFB, or sorry the FFB will only be as good as rFactor2s. I would still choose that sim over another sim that may have better FFB but not God's perfect physics. I want the car to be acting and reacting and doing exactly as close to what it would do in real life for every given situation, then after that we can work on the FFB and try to get it to be as good as we can.

 

Again the balk of what I have said is outside of FFB

This is the issue with RF2 the cars become unstable and lose control in places that real cars don't.

NKP is awesome outside of its FFB the FFB is just the icing on the cake , though the single seaters suffer from a lack of overall grip outside of that the model is incredibly consistent so although they are to slippy its as if you are driving a real car but with a less gripping compound tire.

RF2 on the other hand is as if the tires are prity much right for grip levels ( to slippery in some ways) but then they have massive holes in the simulation when the tires spin up or the cars get into evan small slides ESPECAILY AT LOW SPEEDS or on the gass.

 

THANK YOU. Ive always felt this too. Like its almost too easy to just tear any car hard around the track, corner after corner, lap after lap. Been trying to explain this to people for months, years. Just because a sim has great FFB and can make you feel very in direct and in control, which makes you feel like a very good driver, which makes you enjoy the experience more, it doesnt mean the sim is more real, it doesnt mean the car is doing things that its supposed to be doing, it just does a fantastic job of giving you that controlling a racecar sensation, which is totally different than actual physics.....

for example

Is the back of the car squatting down enough when appling 60% throttle in a given specific situation, and when it does squat does it give the rear tyres the extra grip properly because of the extra load? How is that rear weight transfer affecting all the thousands of variables that are being calculated by the sim" but it doesnt stop there, now, how are all those calculations affecting how the car is moving, sliding, acting, reacting, relative to the surface that its on and reletive to all the complex variables including driver input. THAT is the stuff that makes a car handle, act and react as close in game as possible to how it does in real life (example was very very basic but you get the point).

Once you get the car to react to the environment, driver inputs and all the thousands of variables as realistically as possible (having as realisitc physics as possible), THEN you might get a different feel on YOUR END (not in game) depending on how the game makers programmed the ffb, but that isnt affecting physics though, its just sending out information to your wheel.

Basically, when I apply 55% throttle at 235 km/h, on this turn, with this much load going through my tyres, with this much roll, with this exact ride height at this exact moment, well, I want the car to be doing exactly, or as close to as possible, what it would be doing in real life in the exact same situation, regardless of FFB feelings, lack of feelings behind a computer, slight lags etc etc etc etc

No way, theres way too often where NetKar Pro seems tooo planted and stable, the base physics I dont feel are as realistic as rF2's, REGARDLESS of how amazing/direct/consistent/in control the steering and ffb may feel. Its almost like they were trying to replicate how everything looks so much easier from a person watching for eg. an onboard cam, and replicating that into the actual driving part. Ok thats going a little too far I know im exaggeratting a bit, Netkar Pro is up there in the few top sims category, but its like they slightly over did the ease of correcting and controlling the cars.

 

ive already linked, in this thread, to a video of me driving in real life on a wet track, in a relatively fast car - literally chucking it around like i can in Netkar pro..

you continue to ignore this fact....documented in video evidence.
I'll link it again, in case you missed it.

I could also link you to a TON of youtube videos onboard, showing people driving cars with plenty of oversteer.......name a racing catergory and i'll show you a youtube video of a guy throttling it and throwing it around and retaining control - F1, Touring car, Anything.
Netkar pro allows you to control cars on and over the limit properly, like in real life (if you have the skill).

Rf2 simply doesn't, you have to Learn the game, and cannot rely on any kind of instinctive feel or feedback.
This is most apparent in the ISI f1 car.

 

Continue to ignore facts .......... lol


Just point me to the track.

I be happy with however rF2 finishes up for one !

Wishing and hoping and shoving your view down peoples throat does no one any good.

 

FFB is related to the physics (unless it's fully based on canned effects), so if Netkar FFB is good means that its physics are good. Stefano said he wrote around 20 lines of code only for the FFB.

20 lines of code out of 250.000

 

I'd say he is saying something quite similar to the OP.

Edit: full interview: http://www.radicalsonline.com/conte...peaks-about-his-life-simracing-and-netKar-PRO

 

Gosh popular thread, but very important subject matter

Ive never really tried netkar pro as it dosnt support motion feed (for motion my motion rig)

I'm sure ISI has copies of netkar pro and would at least being trying to better it?

 

Sooo, are we comparing wet real world physics to dry sim world physics now?

Wet weather is very much more forgiving in real life, but also it is challenging in it's own way, but it is easier to overdrive without loosing control when it is wet, that is in reality my experience at least.

rF2 is first sim for me where I can be scared but where backing off I'm still in control and not scared, only sim that truly has managed to get that thrill of limit in game. GT4 car is really well made, imo.

Remember to check how much you apply throttle in sim, it is usually much more than one thinks, problem is that engine load is not felt and not even heard most of the time, so it is easy to apply more throttle than one intends to. I find that my throttle usage is 50-100%, quite rarely below that, but in reality lower 50% get used too.

1:26.5x GT4 lap for example:
View attachment 3562

 

+1

 

View attachment 3563 very funny Durge- your avatar is a guy from my home town ....here is one of my buddies playing guitar with him

 

OK, watched the video a second time seeing as you took the trouble to post it again. I have a couple of questions for you.
1. How close were you to maximum grip level and maximum slip angle for those tires in that condition?
2. How much throttle were you giving the car on exit of each of those corners?

And nice one Spinelli, you get a +1 for saying the same thing basically that I did a few posts before, LoL

 

erm.........is that not clear on the video.....?

try turning your sound on.
also try looking at the steering wheel in the video - its right there near the centre.

i mean...at 6 seconds, im literally bouncing off the rev limiter, whilst applying opposite lock, powering out the corner.

how can you.....not see/hear that?

i just can't believe whats happening here lol.

and just as a footnote - that was literally my first ever time in that car, on that track (which was damp and wet in places), and we only had 15 minutes of track time.
so, digest that.

im going to relink what domi posted on another thread, as a perfect example (and more proof)
this time, from a more established driver...in a much more powerful car.

 

ATTENTION.

What we've got here is what we've always got on the internet, any number of people who reject obvious evidence posted before their eyes{could be that they're religious}, then we've got another issue of those who seem to favour mathematics simulators over racing simulators, me, I favour racing sims, and if you look at that schumi video above, that level of control is what GSC 2012 Camaro has, and for that and other reasons, it's probably the finest simulation of a tin top, with the rf2's F2 being the best OWer.

 

Hey all,

Still relatively new posting here; I've been over at LFS since 2003 but came to RF2 when the beta was released, hoping for something new (any LFS folk here will understand that ). Really enjoying RF2.

Also enjoying the continuation of this debate which apparently spans all sims of all time. I spent some time with iR as well for almost a year before I became disappointed with their progress in terms of physics. Seemed like I was investing in a lot of content but no return on the most important part for me - the physical simulation. I probably fall into the "math physics preferred" here, since I believe that in the end, that will ultimately be the best "racing simulation" as well simply by nature, even if it's a longer path.

In any event, one thing that hasn't been mentioned in the thread is the area of aero drop off with yaw angle. Driving your street car at 45 degrees through a turn whether on the street OR the track doesn't relate very will into RF2 partly because of aerodynamics. As soon as a downforce equipped vehicle begins to yaw there is a non linear (in a bad way) reduction of downforce which does help, to some degree, to explain the perception of the OP.

In my mind, this is enforced by the fact that everyone (pretty much) loves the old cars, and the slower ones. I'm sure the bias ply tires help on the old cars too though.

Personally, I think the posts describing nKP as too forgiving are spot on. After years of LFS I feel the same way about LFS. nKP for example, it seems counter-intuitive that I can drift the entire Trento Bondone in the Vintage GT at silly angles after a 6 pack with no issues at all, because I'm just not that good

But I don't want to continue the sim vs sim thing, just wanted to throw in my two cents.

Enjoying RF2 and looking forward to a bright future, and the incoming turbocharging sim attempt! LFS, to date, has been my only date when I need to satiate the need for some turbos; dubiously implemented or otherwise

Cheers

Edit: I just have to say though I read earlier in the thread, the OP felt that the 350Z drifting through that turn was still in the "green" zone on the graph he posted. That's a bit of a silly comment IMO