Driver styles/preferences

[timbo]

Gr8 post Ciro!

[747heavy]

747Heavy, the first paper you show use one radius for the curve. Is it some kind of elaborated joke? I did not get it at all. The second one analyzes drifting (on a bike?) to win. No comments, because either I also did not get it or those guys come from Japan. The third explains the same thing, I think.

I agree, Heavy: they are not for the "causal" fan (ha, ha! Another good one!). Burn them, I'd say.

I tried, unfortunately this forum does not support this feature, perhaps something to implement for the future.

[auto saibot]

Alonso, you said.....................

I agree, Heavy: they are not for the "causal" fan (ha, ha! Another good one!). Burn them, I'd say.

all that you wrote =D>, utter enjoyment

[Ciro Pabón]

I had heard a couple of years ago that Reubens was the only one who still relied primarily on right-foot braking. Can anyone confirm this?

Well, yes.

This is complicated here. A race is not a curve. A race is a loooot of curves.

You see, when you take a late apex curve, essentially you are:

- braking harder
- turning harder
- accelerating harder and longer on the exit

This is not good for your tyres on the long run. So, if you could take ALL curves at mid apex, at the end of the race you have more tyres than your opponents, won't you?

The problem here is that this tactic is good for races where you have a fair chance of overtaking! Rubens career proves that in F1, being second all the race, taking good care of your tyres, gives you a fair chance of... arriving in second place. This doesn't happen so blatantly at other series like NASCAR or Indy Cars, which are long games of poker...

Many brazilians are master of NOT turning the wheel. So, if you think about it, a late apex, a tap of the brakes on exit, a long braking and a smoooth continuous movement of the wheel curve after curve (like, again, Schumi) makes you feel the pain of the tyres (have you felt it? you can hear them complaining... ).

IN THE RAIN Rubens´s tactic is an excellent one. Check this video, that shows you how a large constant exterior radius is better. In the rain you do not want sudden movements, because is easy to lose the grip and you get nothing from your tyres.

Ayrton lap extraordinaire at Spa. from fifth to first in less than half a lap (that's more overtakings than the one we have today in one race). Four curves, four overtakings, as usual avoiding Schumacher "good intentions". Check the second overtake, for third place, on the outside of the curve: you can feel he's concentrating on his right foot, pure "Brazilian easy rider", "fully circular", no cross of the sideslope, taking advantage of a slightly raised external edge on this curve
[youtube]http://www.youtube.com/watch?v=9t8QMyuGhdE[/youtube]

I'll pick at random the opinion of somebody. For example, me. When I ask myself: "do you love racing in the rain" I answered: "yes". Now, I only use left foot braking intentionally.

Thanks, guys, for your comments on my previous post. You're very kind. You already knew that, I simply made a summary.

[Tim.Wright]

haha "despite Schumachers efforts to ease him out on the wun down to wedgate".

Hey Ciro,
Have you done any work on deriving a "racing line" computationally from track information, eg centreline and track width?

Its an area of interest for me at the moment. (Maybe this needs to be shoved to a separate thread?)

Tim

[Ciro Pabón]

Sure, Tim. That's why I've made restitutions of tracks (well, only once and only of five specific tracks). They are for clients that race in Historical F1. They use them for checking some racing lines, and for comparing the movements of suspension with the racing line, but I have no access to the software they use (those guys are RICH).

However, this is the beauty of racing (and the reason why I mocked slightly the papers given so kindly by 747 Heavy). You can get a mathematical solution for ONE curve. It has been demonstrated that it is impossible to find an proven optimum solution for TWO curves, even if you use the Hamiltonian (which, anyway, is an approximation to the behavior of a car, because of the tyres are moving under the rim anyway and I guess that trying to model that is a nightmare).

In the end, most software works as a microsimulation, not a macro (micro uses simplifications about the behavior of the car and then "magic curves", while macro uses, supposedly, the true dynamical equations).

However, you can try. You can use Hamiltonian mechanics, or so I think.

I used Bosch software and Brembo cards for some time here in Colombia, because the cards cost around 200 U$ and the software is free, but, hey, in the end is all about your instincts and your ability...

Anyway, give it a try, you'll learn A LOT. For Bosch software, check here: http://www.bosch-motorsport.com/content ... l/3589.htm

The documentation is here: http://www.bosch-motorsport.com/content ... Manual.pdf

And here: http://www.bosch-motorsport.com/content ... mV2007.pdf

I posted in my defunct site a scan of the article about simulations, that appeared in RaceTech magazine. Just in case, this is it:

Accesible simulations



Belatti once passed me a nice package they use in TC, I gave it a try and, yes, it works. He should explain how he does it, because now he knows more than anyone around here, AFAIK.

I'm seriously going OOT here, so sure, this deserves a thread. However, what's its name? Roads and trajectories? We lack the information about superelevation, btw.

[timbo]

Ayrton lap extraordinaire at Spa. from fifth to first in less than half a lap (that's more overtakings than the one we have today in one race). Four curves, four overtakings, as usual avoiding Schumacher "good intentions". Check the second overtake, for third place, on the outside of the curve: you can feel he's concentrating on his right foot, pure "Brazilian easy rider", "fully circular", no cross of the sideslope, taking advantage of a slightly raised external edge on this curve

Ciro, I'm a little bit embarrassed here, this was Donington
But more on topic, I read somewhere that his lines at the outer edge of the corners during that lap were chosen because he knew that rubbered in surface would be more slippery in the rain.

[Ciro Pabón]

Yeah, you're right, Timbo. I actually looked for Donnington 1993, but I got a different movie at YouTube, then I finally found the video and I thought I heard Spa. Sorry.

Yes, you're also right about the rubber, we've already talk about it. You also avoid the oil in the racing line, all that gives more speed. Thank you very much, the embarrassed one is me (although I don't care about being embarrassed, that's the idea of a forum, together we can do more).

[timbo]

Yeah, you're right, Timbo. I actually looked for Donnington 1993, but I got a different movie at YouTube, then I finally found the video and I thought I heard Spa. Sorry.

No prob=)
Can you comment on what I posted about having more grip on the "green" surface in the rain?

[Ciro Pabón]

I think you already wrote about it, when we talked about Turkey, when the circuit opened. Am I mistaken?

Anyway, yes, as I said, you avoid the oil. When it starts to rain, the mixture of rubber, oil and dust (specially dust) gets pretty slippery, as anyone going out in the first summer storm knows. Besides, all the water concentrates in the inner part of the curve. Is there anything else to say? I know you also like to race... so, please, go on, master Timbo, by all means. We hear you.

[Tim.Wright]

Yea I think road and trajectories would make a good new thread name. I'm going to go on here assuming you will push this into a new thread later

I have done a lot of work with a simulation package which has a race driver controller. Once I found out how important the trajectory in terms of laptime simulation, I have had somewhat of an interest in the calculations of deriving it.

This particular driver model did a pre-processing of the track data and derived a trajectory based on a "minimum strain" goal. To me, his suggests a trajectory to minimise curavature by using all of the track. Then there are modifiers to change the apex point and curvature factor of the trajectory. There are other calculations that go into deriving the trajectory but I can't get any of that information out of their engineers. Of course thats all proprietry stuff.

I was doing a lot of lap simulations on Monza and found that by accurately modelling the track width (as opposed to running a 12m wide track everywhere) was worth 1-2 seconds a lap. This is what really made me realise the importance of trajectory when assesing the accuracy of a simulation. Suddenly my steering traces and velocity profiles matched the telemety much better!

Speaking of elevation, I found this particular driver could not handle 3D track geometry in a predictable way. The simulation software could model it, but when I ran the driver on it, it completely looses the plot.

I really think that there is a gap in the market with respects to a good race driver model. Only thing we have found is that there is no-one willing to pay for it.

Tim

[timbo]

I think you already wrote about it, when we talked about Turkey, when the circuit opened. Am I mistaken?

I'm not sure I was registered back then.

Anyway, yes, as I said, you avoid the oil. When it starts to rain, the mixture of rubber, oil and dust (specially dust) gets pretty slippery, as anyone going out in the first summer storm knows. Besides, all the water concentrates in the inner part of the curve. Is there anything else to say? I know you also like to race... so, please, go on, master Timbo, by all means. We hear you.

Oh, please don't take offense in anything I've said, by no means I was trying to refute anything you said there! Surely he was the master of the throttle, and his steering input looks smoother than Schumacher's and of course Alonso's (but damn wasn't he fast correcting the car, when it slided!).
My point was, there's so many things going during that lap, and as we know him being VERY analytical, so he went from 5th to 1st because he was a)smooth and choosey about his lines b) knew where to seek the grip in wet conditions.

[luca]

Button, who supposedly prefers an understeery car, said Pirellis suit his style.
Yesterday Sutil said Pirellis tend to oversteer.
So, which is it? It's not like it could be both at the same time (beside serious issues with the setup of the car).

[Tim.Wright]

Style is such a subjective word. There could be a million other reasons as to why he likes the tyres. Theres more to a vehicles feel than weather it breaks away at the front or rear first. Much more.

Not to mention the fact that vehicle design has a large effect on the OS/US tendancy of a car. Granted it is somewhat limited by the mandatory weight distribution.

Statements such as "driver X likes a car that over/under-steers" annoy me because they are generally taken out of context, propogated further by armchair engineers, and then taken as gospel by the forum community at large.

[/rant]

Tim

[PlatinumZealot]

I agree. Style and Preference are two different things. Just like with clothes. You can have style and no particular preference, or a preference and no style!

And to think, the preference can change and to some extent the style too.