Reader Submission #32 – Modern Sims need to re-invent how we approach Car Setups

A fantastic Reader Submission has come in today from Ruben Lopez, who after reading our lengthy article on The Evolution of Broken iRacing Setups, was inspired to write his own piece on car setups relating to Project CARS, and sims that will inevitably be released in the future.

attachment.phpSome days ago I read here on PRC a great article about iRacing setups. I have never played iRacing, but still the stuff I was reading sounded familiar enough. It seems that, in every racing game you play, there’s a God-mode setup that gives you an absurd speed boost. And that setup is usually far from similar to what that particular car needs in real life. In your article it was not a pretend racer like us but an insider with actual real life setup experience confirming how the ideal values on iRacing would make a particular car terrible in real life, and how ideal real life values did simply not work on iRacing.

Without having any setup experience myself at all, I can point to similar flaws in other games. In Project CARS I’ve seen cars that go faster with extremely low pressures and overheated tyres than with sensible pressures and the correct tyre temperature. Or cars that transform into infinite grip spaceships by lowering and stiffening to the max every suspension part. Adding negative camber and toe to the very edge of the bars also seem to have no downsides and you just get faster and faster.

I don’t know if we still don’t have powerful enough hardware to simulate this stuff, or if it’s developers lacking “know how”, but this seems to be present -to different extents- in pretty much every racing game, be it a simcade game or something like iRacing that aspires to mimic real life as much as possible. I don’t know where the problem lies and I don’t care too much either, but while the situation doesn’t improve, the current way racing games handle their setup sections is just not ideal.

Instead of being given a reasonable baseline setup that the player can try to improve to get a bit better balance and a couple of extra tenths, we’re always given ridiculous default setups that are miles off the pace. Then you have to start a long trial and error process to see what’s broken in that particular game and car and exploit it once you find it. So you start driving a car that is clocking laps similar to its real life counterpart, and you end up with a spaceship beating the real times by huge margins, and with a driving experience that has just nothing to do with the actual car. I’m talking about stuff like an unlicensed GP2 car going flat out through Campsa Corner in Montmeló circuit at Barcelona and clocking laptimes good enough to lead a F1 race.

Another common problem is that the ranges for some parts are poorly selected, and the ideal setting is in one of the extremes of the slider… If I can race a car with 20% brake duct cooling opened in Bahrain International Circuit at 45C degrees, it means everything above that level is just useless and should not have been included. You can be sure the default value will be way above that for no good reason. So you have a slider that just has a “fast” side and a lot of useless values.

I have discussed this with other sim racer pals and some say that the game should reward you for driving hours and hours to prepare a race, and that it is OK default setups being seconds off the pace as they have to be easy to drive for people the first time they use a car. I just don’t agree at all. These default setups are usually WAY harder to drive too on top of being slow, and I don’t think unlimited practice time should decide the race winner. Sure, you can’t expect to be fighting for a win if you just jump in the car for qualifying. But the opposite situation is ridiculous too, you shouldn’t have to drive 10 race distances to find what’s broken and capitalize on it. 

I DO like having to drive to tune my car and I enjoy the preparation for an event. It’s nice finding by yourself that small change that improves the car balance and makes you gain some confidence. But when it all becomes a contest to find the broken parts through extensive trial and error, that’s just not enjoyable.

I was gobsmacked to read in your article that there are guys being paid to create iRacing setups. That alone tells you all you need to know about how ridiculous the situation can get.

I feel bad moaning about a problem I don’t have a solution for, but I’m sure there are better ways to handle this. Just by reducing the range of each part to useful values the situation would improve a lot. I mean, why you give me a huge slider for the spring rate if in your game stiffer means faster everytime, with every car? Making double sure the ideal values are close to the real ones would also make everything far more intuitive, but this is probably harder than it sounds… Still, if you can’t achieve decent levels of real life fidelity, the setup feature is not adding anything good to the game.

As it is now, in most cases the setup feature causes more trouble than the enjoyment you can get out of it, and not a lot of games give support to those wanting to escape this bullshit (like lobbies where custom setups are not allowed). I’d like to know how you feel about this topic guys! 

6885855018_32359f7570_bAt some point, Maple has told me he’s going to write an in-depth setup article on here, so prepare your collective anuses (anii?) for that.

All real life setups are exploit setups; real world car setups are constantly evolving and designed to get around whatever rules the sanctioning body has laid out. Coil binding in Stock Cars is essentially an exploit setup – designed to get around the ride height rule during tech inspection, only for the nose to be pinned to the ground by the wall of air going over the nose of the car at speed.

coil bindingThe reason the extreme exploit setups work is because of poor coding. The NASCAR games by Eutechnyx are a prime example of this. On Daytona and Talladega, both in real life and in video games, you want the front end of the car sitting higher than the rear end of the car so the huge rear spoiler is taken completely out of the air and doesn’t slow the car down. In the real world, it’s possible to go too low and bottom out the car, which will cause a wreck sooner rather than later even though you’re at two tracks where handling isn’t a priority. The Eutechnyx games never punished you for bottoming out on plate tracks, meaning you’d see absolutely ridiculous setups where the front end is pointing almost skyward.

The reason you could get away with this? Nobody coded in any elements to make you wreck when bottoming out like in real life. Most dev teams don’t have any proper mechanical engineer as a staff member, which is pretty counterproductive, as when you’re dealing with cars, you need someone on the crew who understands the physics of a car in order to model it properly. The stuff that dev teams miss by not having a mechanical engineer on the staff is eventually what gets exploited.

You’re right about baseline setups. They should indeed be in the ballpark of an ideal setup to begin with, and devs really need to get on this. This is one of the main reasons I’m so quick to share setups on here; the default garage menu values and setups for almost all racing sims are atrocious, and unless you fully know what you’re doing (I’m 80% there), you’ll have no clue where to start. It’s really something most dev teams could outsource to the community – find a couple fast guys and give them a full list of what they need to develop. Heck, for some games like R3E, I don’t even have track specific setups. I run the same setup everywhere and just tweak the sway bars based on how it feels out of the box. Nothing stopping the Sector 3 guys from using that setup across all the different cars it works for, and just making it the default for each track.

Fixed setup lobbies are definitely the way of the future judging by iRacing’s participation levels, but, as someone who is comfortable in the garage menu, it’s very hard to sit down and say we need to lock this adjustment and keep these adjustments open. To me, all adjustments you can make in the garage area are equally important and locking down the setup aside from just downforce levels or brake bias doesn’t really work. It’s like trying to solve a math problem only using addition – you need the other three operations.


18 thoughts on “Reader Submission #32 – Modern Sims need to re-invent how we approach Car Setups

  1. Rear ride height at plate tracks is actually limited by spec shocks and springs, they used to run some really dumb qualifying setups


    1. That isn’t what he meant, he was talking about the fact that in those games you can actually bottom the car out to the point that telemetry wise you are dragging the frame rails around the whole track cause there is no downside like IDK friction for example to slow the car down

      Liked by 1 person

    2. Also under current NASCAR rules the springs and shock are mandated but that is all, you can exploit certain things like track bar height splits to get more or less rear travel and that’s what current NASCAR engineers regularly are trying to figure out at superspeedways is how to exploit the rules as much as possible to get minimum rear travel through other means besides springs and shock, so they can use a lower static height at all times while still keeping the rear end from scraping to a point that is detrimental to overall speed


  2. Any setting that has its ideal value in one of the edges of the sliders is indeed broken. If you can’t fuck up both having too much or too little of something, that means either the range offered is wrong or the downsides of using a extreme value are not present.

    Like the “no bottom down” example. If that effect is not present then ride height becomes an area to exploit.


  3. I just tried this theory with project cars with the BMW Z4 GT3 around Donington park, my pace is less than a tenth of a second off the current world record for that track with that car ( the only GT3 car in front of that is the Bentley, which isn’t used in proper leagues as it’s known to have a big advantage )

    I decided to go all out and just stiffened everything up, all the way to the end of the sliders, including what would be considered crazy camber and tyre pressures etc, the works.

    What i found was that it didn’t make me any faster, i was on average a couple of tenths slower, but what i think is worse is the fact it barely changed the cars handling at all, it bounced around and ‘looked’ like it should be struggling a bit more with these crazy settings, but i wasn’t feeling it through the wheel.

    It could perhaps be down to the car being less sensitive in the game to these changes than other cars, maybe if i tried it with the F1 style cars it may of made more of a difference, but i couldn’t be arsed.

    With these cars in my experience the only time gains are really made through the basic shit.

    Liked by 1 person

    1. Yeah, the particular car is probably less likely to exhibit changes as much as others might. That said, running intentionally jacked up values on a z4 gt3 SHOULD cause severe issues for setting competitive laptimes.

      It’s better to have ‘setups’ moderated if things are already rather inconsistent in the first place, so with pcars, I would rather not see very effective changes, at least for now.


  4. A friend of mine who spent a LOT of time racing Skippy trainers in iRacing and a LOT of time racing real Skippy cars on real tracks, made really cogent suggestion: if IRacing (or anybody else in the sim biz) wants to go all stark realism on everybody’s ass, what they should do is ensure a one-to-one relationship with the real steel. That is, if -2.1 degrees of (negative) camber on the LR is exactly the right setting for the real Lime Rock, than that should also be the optimum setting under the same circumstances in the appropriate sim. Nobody (that is to say, least of all Dave Kaemmer, who has had a LOT of real-world time at Lime Rock, say) took him up on it.

    Liked by 2 people

  5. It’s definitely a lot harder to get the real world setups to work correctly than to get the correct response to each setting. Just having front aero increase/decrease as you change the wing angle is simple. Maintaining the exact same aero balance as the real world, through every situation you experience in a lap of the circuit (track surface heat, dirty air, adjustments to other wings, brake ducts sealed or not) would require a huge amount of data collection and probably just end up overfitting behaviour at that particular circuit when you inevitably missed a variable.

    Some things that affect sim handling you can isolate, like putting shocks or an engine onto a dyno, others you can’t measure without the real car and expensive tools, like chassis flex, inertia. As for tires (what’s the optimal camber at every pressure/temperature? How much lateral grip does it provide?) the real world guys don’t measure it at all, they just do what works. The manufacturers do measure some aspects but others like “how does grip change with tire wear over 20 laps” it’s pretty much word of mouth, so if someone told iR that grip just drops off a precipice after 20 laps, that’s what happens (even though that’s entirely ridiculous and is causing massive crashes)


    1. Of course temps and pressures are measured IRL mate. That’s the first they do with these “pencils” they put on the surface. They do some laps and measure temps at different points of the surface. And pressure is constantly monitored, that’s how teams spot slow punctures before drivers.

      That stuff is critical. It’s not a game, if a tyre overheats it will blister and stop gripping.

      Liked by 1 person

  6. I didn’t say it was do-able; I said nobody’s done it. And yet…and yet…at the end of the day, every time a real-world car leaves the pits, it’s set up with real-world values. If they’re right, then the driver has a shot. And if they’re right, somebody could transfer them to a sim. Not me…and not thee, apparently.

    Liked by 1 person

    1. It can at least be sensible and somewhat reasonable. That’s the thing.

      Some of the titles are starting out with completely fudge values, simply to make the car drive ‘correctly’. Once they add setup tuning on top, it just exacerbates the problems.

      Setup tuning in titles that allow unaddressed exploits are a negative in terms of features for people that simply want a good race.


Ratio of vowels to consonants will be monitored. Post at your own discretion.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s