Car set-up wisdoms

It is not always like that, I can't say I am good driver or fast let say because good is not the right word here.
If we get spa and gt 2 I can do 2.19 with softs which is slow. But I do it with my own setup, I just did some laps with the default and I can't get it under 2.23!! If I change the ratios I managed to 2.21 but not steady. I can feel when the car is wrong and does not suits me but the hard thing here is know how to tune it in order to make it feel good to me.

No matter how fast you are you can make the car at least easier to drive so you can make steady laps and be able to learn easier I hope you get what I mean.

 

that is why setup gods gave us telemetry and suspension histagrams

 

True. But that's not the whole thruth because you don't have to mess with every setup option. You'll find very soon some changes you want to do to make your car better for your driving style. For example the first this kind of setup change for me was to loosen up ARB's a bit. I don't like car to be too sensitive for my inputs.

With others setups I always felt frustrated. I still haven't found any working setups from the forums..

 

But as I say, that doesn't tell you why you might want it stiff soft or whatever.

It's just purely descriptive rather than pragmatic. It's like reading a dictionary to learn how to speak English.

Aris wrote some stuff that wasn't - that was actually "increase the front camber on road cars / lower the tyre pressure" - although I noticed no perceivable difference at all when I tried it on a few of the cars a while back. No measurable effect at all.

Perhaps I could create something where I gave the AI setup changes and measured the difference that way - maybe that would be convincing. But TBH, I kind of lost interest as setup changes didn't really alter or improve the gameplay in any meaningful way.

TBH I think mostly the setup differences are either blindingly obvious because they are huge (for things like gearing and aero or changing the tyres) or they simply do nothing - and people will change things and any change to their lap time is just coincidental.

It's like you could adjust the tyre pressure in AC and see how that affects tomorrows weather - and then spot a non-existent pattern.

I think it's rather like track grip too, as Stefano was saying, putting extra grip on the racing line, in a realistic way, wouldn't really make any difference to the gameplay. The only way it would is if they made the effect unrealistically larger. I feel the same must be true of setup changes - at least those on the road cars.

 

What's the question here, read the OP and honestly... confusing.
Yeah it takes tons of reading, driving and making setups before every time you leave the pits, exchanging setups and tips with the fastest drivers, changing your driving style to suit the car/track/setup/race-type/weather/strategy/... combo.

A setup for a sprint race is different from a setup for endurance race. Maybe not in AC, haven't seen any proper endurance here due to lack of driver swapping but in LFS there were differences and even whole different setups based on race length.

If you want to point to some places I'm sure Google can dig out the info. It's often found around LFS, some rFactor and real cars. But most you will have to learn yourself by driving.
If you cannot feel weight transfer then you have tons and tons to practice as a driver before getting to making setups. And no you don't need FFB or something fancy to feel weight transfers, it's all right on the screen.

There are some tips that can be written down but they aren't universal and often may be personal preferences. It's one thing to make a good setup in general, it's another to make a fast setup for a fast driver. These things are often tailored to a specific person and their driving style.

Also you have to be able to tell the difference when you change a setting by it's smallest amount. As well as test on a combo that you know and are consistent on.

A simple test, guess a car's weight distribution in % by driving it. You should be able to guess +-1%. For some cars in AC it can be calculated for verification, for others search the real counterpart to see how good your guess was.

For some games there are online setup sharing places but to be honest, setups from there are often crap

 

The struggle for me was that changes make no perceptible difference when I tried them.

You know, I wrote in the last thread on this topic that I would have been happy if my setup changes had made the car worse to drive, or slower.

But they didn't - they didn't, so far as I could tell, change anything for good or bad - and, no, I wasn't 4+ seconds off the pace.

Note however, I'm talking mostly about the road cars - I know you can change the gearing or aero, or switch tyres and get some huge differences (differences that don't require anyone to be a "good driver" to see, nor indeed "years" to learn)

But you know, I'm skeptical that anyone really knows anyway - that's why the guides are so lacking in any detail or substance. People adjust the things that make a big difference - kinda obvious the difference, for e.g, that aero makes on Monza or softer tyres, or getting the gearing so the car isn't hitting the redline halfway down the straight. These things are trivial to know and change. They adjust other stuff do a few laps, and if their time is quicker convince themselves that the setup was better. I'd argue that Kunos could add a bunch of sliders that really did nothing (in fact I think they already have )and I bet some would convince themselves that they'd improved their laptime by adjusting them

 

Spa is uncharacteristic and will highlight major flaws in the default sets more than elsewhere.

Still, honestly, when I was 4 seconds off pace in early LFS days, I didn't try to change a so-so setup to suit my "style". My style was probably wrong. I drove the hardest- meanest-on the f**ckin edge WR setups until I was able to manage them. I never tried to build up speed by making sets suit me first. Maybe it works...I have no idea. So yeah I get what you mean, it's just not a path I can recommend.

I assumed that people interested in setup want to get close to ultimate pace more than stick around midpack and feel confortable there. Maybe I'm totally wrong.

Pretty sure almost everyone has a hard time feeling differences on road cars setups tbh. I use the delta bar app, so if a setup change gives me 0.05 extra lateral G, I see it on the screen even if I don't feel it. And if it costs me 0.03 seconds on the next straight, I see it too.

 

I guess you are wright but for the basic stuff like ratios aero and differentials and brake bias even car height you don't need to be alien to setup them, also to feel the understeer or oversteer or unstable braking you don't need to be fast you can adjust that even if you are not that fast. It can help you to drive cleaner with less mistakes and the laptimes will start to improve from there not from the setup itself.

 

Yeah, well I reached that conclusion that they don't make any difference. Thanks for confirming it.

You see this is good because it shows my point. Instead of saying "use telemetry" or "use the developer apps" you actually say what specific app I might use and what it might tell me of interest. Something specific and pragmatic. However, whether I deduce from your post that 0.05 extra lateral G is good or bad, I'm not sure because you don't actually say. No doubt "it depends". Presumably it makes you faster through the corner but perhaps at the expense of being slower down the straight?

But this is the kind of thing I meant - watching or listening to someone actually creating a setup from scratch - someone who actually understands it and who can point to a number or drive around a corner and show you that he understands it and demonstrate what his change did would teach something (rather than the myriad people who just slide stuff around and see if they are quicker or not - which, to be Frank, any buffoon can do, myself included - it's the point when you ask me why I made the setup changes I did that you'd see I was a buffoon - albeit I'd say "I changed it to suit my driving style...it's more of an art than a science...I got under 59" with it" to fit in)

The other thing that can teach you is, as you say, if you make changes and experiment and spend years doing it. Perhaps I started with the wrong cars. I thought using the cars that had significantly fewer sliders was reducing the complexity but it seems really these cars have things to adjust that make next to diddly squat difference anyway.

I don't believe any of the guides really do this. They might teach you if you didn't know what "camber" or "toe" was but none of them I've read have been very useful for people who know what camber means, but really have no clue why they might want to change it nor how they might go about measuring in, if not a scientific way, at least reasonably repeatable way what effect their changes had.

Really I'd like it if the python api could let us load ai cars and give them a setup.

Then you could do something like this pseudo-code:

ai_driver = load_ai()
laptimes = []
for tyre_pressure in range (15,40):

for camber in range (...):

for toe in range (...):
setup = create_setup(tyre_pressure, camber, toe)
ai_driver.apply_setup (setup)
lt = ai_driver.do_hotlap()
laptimes.add(lt, setup)​

print min(laptimes)​

Maybe could do something outside of the game like this? Can't you put a setup in a folder for the AI?

Of course people are saying "driving style" - but it'd be interesting to leave the above running and seeing which setup, if any, makes the AI faster.

 

I notice the difference with setup tweaks.

 

So what you are saying is that people are adjusting something ranomly and making fast setups? Really? Is this where we are now? Everything that you change makes a difference if you know why you changed it. Do you have too much power oversteer? More front rebound, does your car understeer trough corners under power? Less rebound. Knowing that more wing doesnt always mean more grip. Every single thing can improve your laptime/make the car more suitable for you. Does it take years? probably not if you have someone to help you, but if not, there isnt much literature out there for you and you have to do a lot of testing yourself. You can make fast setup if you have enough experience with the car and you know what works and what doesnt. 1 tick +/- rear ARB meant 3 seconds for my laptime around nordschleife. Did I notice it? Not that much, but i guess it was a big change over 24km. I know exactly what im doing when im preparing my setup, I know why i changed the things that i did. There hasnt been a press random buttons and hope it works for me. And im not even that good at it...

 

Yes and no. You have to understand that supposedly we're led to believe in this thread that you can get within 1% of the WR time with the default setup. So, for most drivers that makes every setup "a fast setup" - indeed I'm sure one post says not to even bother with setups unless you've done this and if your laptimes aren't all within 2 tenths of each other.

Now, obviously for some setup changes, as I've already said, this is demonstrable nonsense - there are clearly bigger differences in laptimes from some of the tyres, for example, or by changing the gearing - but equally there's no need to find these "randomly" they are obvious.

The low hanging fruit of the setup world if you will.

So, we assume they mean the more subtle changes. And yes I've no doubt at all that some people are adjusting stuff and it makes no difference at all to the time the car can lap. Except when they do a lap their time will be different from another time they do a lap - and hey, if they are already a fast driver that will be 'a fast setup' except as the guys says another guy driving the same car in the same conditions gets the same laptime with a completely different setup and they conclude its "driving style" or something.

So, by yes and no above - I mean, yes they are adjusting things randomly and the setup may make a fast lap, but equally no - the setup isn't what is making the car fast or not.

 

Agree 100% on the basic stuff. I felt that the discussion was going in too deep and that's what confusing most beginners more than helping them. I felt the thread was trying to close that big gap with text, it doesn't work. That's more what I was referring to. Advice on basic stuff works, can unlock the first couple of seconds and make the car feel ok if that's what people are looking for.

 

It is what makes the car fast... without setup they wouldnt be nearly as fast as they are. What you are talking about is that different setups make same laptimes and that is the beauty of it. At the end of the day, driver is still important, but when the driver gets comfortable with his setup, he is faster than he would be without it. Think of it as different brands of shoes. For me, Nike is the most comfortable, i have no blisters and i have excellent ball control. With Joma i still have the ball control, but after the match i have blisters and they arent really shoes for me

 

Those aren't developer apps. The names of the apps in-game are "Performance Delta" and "G-Meter". If you have even scrolled through the list of apps that should be self explanatory. Not everyone uses the game in English, so saying exactly what each app is called isn't really possible. He shouldn't have to say whether more G force is better. If you know enough to be messing with those sliders you should know that. It isn't some obscure racing art, it is common knowledge. Asking if being able to pull more Gs through corners is better is the same thing as asking if going faster is better. Generally speaking in order to get more grip through corners you have to trade straight-line speed, and those tradeoffs are easy information to find. For a some of them, like that specific tradeoff with aero, you don't even need a setup guide because just mousing over the value in the setup screen will tell you.

 

Learning how to set up cars falls most of the time into having confidence with the car. Having confidence with the car lets us push instead of babying it, trail braking and earlier throttle on exits at the same time hugging the kerb and not worry about spinning.

That is set up for me. I mean how are you going to trail brake if you have huge turn in oversteer let alone be on the kerbs while braking hard. When I started playing AC these were unknown to me and naturally understeer is what was suited for me. I liked understeer because I was not spinning all over and finishing races, I got really good on finding the most grip on the exit but the drawback is that I am no good at trail braking. Now I set the car up for more oversteer because I realized I don't have to fight the steering wheel and not be on the steering lock every time I turn.

Everytime I see fast guys join in I'd watch them how they get it done, sure I can download a setup online but anything I get is no good for me (this was around 300 hours for me and AC is the only proper one I ever tried) so I decided it was time for me learn how to set up a car.
Ex arb f=max, r=min then drive until I get used to it, once I'm comfortable and consistent with time I do arb f=min, r=max. It is a long process but you do get a feeling out of the details. Some things to remember is that arb are considered as springs, tire pressure as well, dampers not so much but once you get how they behave you will notice the subleties.

 

Click stuff and go fast.

 

It's a minefield! A wealth of information all tailored to an individual's tastes.

I'm .1 away from 2:17 @ spa, I know I'll find that point one, I relish that challenge. Is it setup or practice? Probably both. One helping the other. I've got to know the car pretty well I know what I'm looking for regarding corner exit speeds, top end speeds on the straights. Great fun trying to put it all together for a quali lap. More fun trying to hang on the shirt tails of a similarly paced opponent to keep the pressure on.

 

Wait. Wut?
I don`t have buttons nor sliders "stuff" and "go fast" ... is that part of the last update?

 

oups....

Finn's griphax.exe experimental update. But don't tell anyone.