Grip

[zenrael]

Having real life race drivers involved in testing/development means nothing. Ben Collins helped make pCARS what it is today. See my point?

Also, you can adapt to driving conditions, it doesnt mean they are correct to begin with.

Come on, are you really going to insist that having real life racing drivers involved in testing means nothing? Seriously?

How then, do you expect to accurately model your environment?

Unfortunately, todays computers are not capable of modelling real-world physics on the fly to the kind of precision that render such testing useless. The best approximate models are used. They are deemed accurate if they yield expected results. So what then, are the expected results?

Statistically, the best way of verifying the accuracy of your model is to trial it on a sample population of people familiar with the modeled environment. I have never driven on a track, I have never driven a car more powerful than a road-legal 1.6. I have relatively little authority on track grip levels when contrasted against an experienced racing driver, with hours under his belt in various cars of a similar size, weight, and characteristics.

Please explain how you would have the developers verify the accuracy of their model, without calling on professional drivers.

 

[Esteve Rueda]

I didn't post any videos, it was Wajdi you are reffering to, just to be clear.

I was only concerned about the grip in tigh low speed corners on a green track. And as I said with Corvette I felt felt disconnected from the road, but not with GTR.

You can't really "push" in those corners, can you? Or if you do then you spin, but this wasn't the case in my driving, floating with no throttle and floating with 5-10% throttle.

This said, Overall grip levels in rf2 feel realistic to me. I just haven't seen anybody proving my point above wrong. It seems every time someone brings this up somebody goes "it isn't easy to drift", "rf2 doesn't feel like on ice", "warm up your tyres and don't push over the limit" etc. so you are missing the point, I do agree with those comments but my question still remains unanswered.

btw, I think I haven't seen anybody just cruising for 3-5 laps warming up their tyres Usually people don't do hotlaps right away because there isn't enough grip to push hard on a green track.

I end with a statment that I like very much how rf2 drives atm, just would like to see small changes in tight slow speed corners.

Ooops, sorry

 

[Sofo]

Real life drivers are the only who knows how racing feels. Having them involved in development must be great, if they can communicate their knowledge

 

[88mphTim]

I've read elsewhere that with a constant and high FPS (100+) and generally reducing input lag, car handling improves, so catching the slides is easier. Maybe this can explain why opinions are so different.

A difference in FPS affects the human perception and reactions a lot more than it would affect the software... lol

 

[88mphTim]

Come on, are you really going to insist that having real life racing drivers involved in testing means nothing? Seriously?

How then, do you expect to accurately model your environment?

Unfortunately, todays computers are not capable of modelling real-world physics on the fly to the kind of precision that render such testing useless. The best approximate models are used. They are deemed accurate if they yield expected results. So what then, are the expected results?

Statistically, the best way of verifying the accuracy of your model is to trial it on a sample population of people familiar with the modeled environment. I have never driven on a track, I have never driven a car more powerful than a road-legal 1.6. I have relatively little authority on track grip levels when contrasted against an experienced racing driver, with hours under his belt in various cars of a similar size, weight, and characteristics.

Please explain how you would have the developers verify the accuracy of their model, without calling on professional drivers.

It COMPLETELY depends on the type of involvement, and the driver involved. Some real drivers just don't 'do' racing sims.

It is absolutely possible to verify information without professional drivers, using data, and as long as you have a sim racer who understands the complete concept of driving a race car (few do, or are even capable, including me). For example if you have a sim racer who can be matched on telemetry and be seen to drive a car the same way a real world counterpart does (again, few do). This is why paying some driver to hold your logo and say how awesome it is, doesn't help a product quality, just it's sales. Saying that though, the feedback of a real driver is actually incredibly important for things like grip on a green race track. I'm frankly amused by this discussion, it's like nobody watches F1 FP1 sessions all season.

 

[sg333]

Come on, are you really going to insist that having real life racing drivers involved in testing means nothing? Seriously?

.....

Please explain how you would have the developers verify the accuracy of their model, without calling on professional drivers.

I worded that badly. I mean its not guaranteed to end up with a quality similation, as in the example I gave . As someone else said, maybe they cant correctly communicate the problems, or are too keen to just give a thumbs up and take the cheque/get out of there.

 

[tjc]

We consistently have professional full-time racing drivers giving us the opposite feedback, that actually a green track can have less grip than we simulate due to dust and debris...

Good to know Tim.

I'm not really concerned by that.

lol

Most of the test drivers are sim racers, and we organize them into regular races. We have quite a few race drivers (incl. world champions, if you feel the need to know what levels we're talking about) who have access, they tend to email us directly.

We're not in the habit of using endorsements so we can put a name to the comments like you might see in a couple of other products. With us you can take or leave the info posted already, and can disagree all you want about the grip levels. We'll just work towards what we know to be right.

 

[Dennis May]

We consistently have professional full-time racing drivers giving us the opposite feedback, that actually a green track can have less grip than we simulate due to dust and debris...

I can't see any problems with grip. At Sebring with the Nissan GTR, once I felt it was too much grip on the track after 30 cars rubbered the track.

Think everything is fine here, beside the 60s cars should not produce that much marbles :X

 

[Ronnie]

Im quite happy with how the FIA F2 grips. Same with the Megane. Its mostly the Corvette and Nissan suffering from these problems.

" FIA F2 grips. Same with the Megane." Huge downforce. ", "Corvette and Nissan" not many aerofoils. Depending more on mech grip.

So can't really compere them.

 

[Kknorpp001]

More people saying the same, the cars in rf1/2 are too icy, when someone gonna listen?

As soon as they forget everything they know about physics.


Sent from my iPad using Tapatalk HD

 

[Rich Goodwin]

Most of the test drivers are sim racers, and we organize them into regular races. We have quite a few race drivers (incl. world champions, if you feel the need to know what levels we're talking about) who have access, they tend to email us directly.

We're not in the habit of using endorsements so we can put a name to the comments like you might see in a couple of other products. With us you can take or leave the info posted already, and can disagree all you want about the grip levels. We'll just work towards what we know to be right.

That's all good mate. I just like to know about testing processes and how you guys get feedback etc.

I respect not name dropping and understand what you're saying on that.

Don't get my initial statement wrong, I love rf2, just want to learn more

Cheers for the reply

Sent from my Nexus 7 using Tapatalk HD

 

[Jameswesty]

The problem is not the levels of grip of a given road , Its that the cars don't drive like real cars when they lose grip or are at the limit of grip.

ISI seem mostly oblivious to this , don't care , or dont know how to design there cars to get around this issue.

If you play GTRE or game stock car ( both games use ISI technoligy) you can instantly see that the cars behave far closer to what you see with real world cars when driven on the edge of grip. (Mind you other aspects of the simulation are way off in GTRE)

Net kar pro FVA and the AC tech demo are also very good examples of how cars drive at and on the limit as well as at low speeds with high cornering velocities .

So in the end even if ISI made the track more or less grippy you will still find there cars are fundimelty flawed to drive at the limit.

In the end the ISI technology works very well within its defined parameters but outside of them its just pot luck.

Simracers that drive to top level Drive RF1 and RF2 buy memorising the limit and then staying under that limit not by feeling out the state of the car through the FFB or easing onto the tire grip.

 

[Minibull]

It is absolutely possible to verify information without professional drivers, using data, and as long as you have a sim racer who understands the complete concept of driving a race car (few do, or are even capable, including me). For example if you have a sim racer who can be matched on telemetry and be seen to drive a car the same way a real world counterpart does (again, few do).

This is the part I'd love to know more about

I understand about the endorsement thing too. Guess some racers may just send you their impressions if they play it in their spare time, but are in no contract or being paid anything.

 

[Kknorpp001]

I can't read any more of this crap. You guys are insane if u think is ice. Learn how to drive for crying out loud. And u also have to look at actual speed in corners. Duh.


Sent from my iPad using Tapatalk HD

 

[88mphTim]

The problem is not so much the levels of grip of a given road , Its that the cars don't drive like real cars when they lose grip or are at the limit of grip.

ISI seem mostly oblivious to this , don't care , or dont know how to design there cars to get around this issue.

If you play GTRE or game stock car ( both games use ISI technoligy) you can instantly see that the cars behave far closer to what you see with real world cars when driven on the edge of grip. (Mind you other aspects of the simulation are way off in GTRE)

Net kar pro FVA and the AC tech demo are also very good examples of how cars drive at and on the limit as well as at low speeds with high cornering velocities .

So in the end even if ISI made the track more or less grippy you will still find there cars are fundimelty flawed to drive at the limit.

In the end the ISI technology works very well within its defined parameters but outside of them its just pot luck.

I don't believe anybody is oblivious to your issues. Perhaps they just don't agree.

 

[Miro]

In the end the ISI technology works very well within its defined parameters but outside of them its just pot luck.

Which parameters are you talking about exactly?

 

[baked bean]

Its easier to call it a flaw than to adapt & overcome, its the same grip level for the dude in front & behind you

 

[Rony1984]

I feel that a lot of people are forgetting that a good tire model can produce good results when it's fed with the right data. Even with the most advanced, comprehensive model, if the inputted data concerning adhesion/cohesion, slip angels etc is not correct, so will be the result.
ISI/Tim has stated in some post that also ISI, like everbody else, has to learn by doing to produce the best possible tires, and hence these will be improved upon all the way during rf2 development.
And then there is the tire model at itself, which as can be seen in the known issues file, is not finalized..

 

[88mphTim]

This is the part I'd love to know more about

I understand about the endorsement thing too. Guess some racers may just send you their impressions if they play it in their spare time, but are in no contract or being paid anything.

Most of us (me included) do almost everything completely wrong. I'm perfectly capable of going fast when I put enough time in, and staying on the track, and always have been with every sim. My personal flaws are highlighted, multiplied and shoved in my face right now with this product, and I'm essentially trying to teach myself to drive from scratch right now. I am hoping that eventually there can be a step by step guide to the simple things.

We have had offers. Drivers who use our software offering to be a spokesperson, we might do it eventually, but I don't think I'd ever want to do anything paid, so we can truthfully say we didn't pay them anything when they endorse us.

 

[88mphTim]

I feel that a lot of people are forgetting that a good tire model can produce good results when it's fed with the right data. Even with the most advanced, comprehensive model, if the inputted data concerning adhesion/cohesion, slip angels etc is not correct, so will be the result.
ISI/Tim has stated in some post that also ISI, like everbody else, has to learn by doing to produce the best possible tires, and hence these will be improved upon all the way during rf2 development.
And then there is the tire model at itself, which as can be seen in the known issues file, is not finalized..

We will always be tweaking. However, belief right now is that you are on the nicer side of both grip and tire wear right now compared to the worst in real life.