Yep, we’re going to make it three articles in less than a month regarding the tire physics in Assetto Corsa.
Dating back to the title’s original release on Steam’s Early Access platform in the fall of 2013, a big selling point of Assetto Corsa – one which captivated many sim racers across the community – were the claims Kunos Simulazioni had made in regards to how each car had been developed. Assetto Corsa was said to stray from the formula that allowed enormous car counts in both Gran Turismo and Forza, instead focusing on real data to ensure the driving experience of each and every car was as authentic as possible. Even though the double digit vehicle roster was overshadowed by both triple and quadruple digit lists found in the mainstream car collecting titles, the lack of diversity was allegedly a trade-off for the most accurate physics ever seen in a consumer racing simulator. And for a period of time, many of us believed them. Virtual lap times were fairly accurate, setup techniques mirrored what had been occurring on the real race track, and the game was for the most part a genuine pleasure to drive.
But of course, people started finding shit. Third party modding teams in the process of building open wheel cars for Assetto Corsa began running into physics engine issues, with rumors of Kunos personally calling them and requesting the teams to keep quiet about obvious shortcomings on the official forums, which had once been an open resource for discussions. Individuals curious about how Kunos were able to pump out DLC packages featuring brand new GT3 cars so quickly ended up discovering many key components and unique geometries of certain cars had been hastily copied and pasted from one vehicle to another. Certain work-in-progress mods that were once advertised as a free download from a talented sim racer would disappear without any explanation given, only to resurface as official Kunos content in a future paid downloadable package. Slowly but surely, the smoke and mirrors show Kunos once used to push Assetto Corsa as the definitive PC racing simulator began to fade away, revealing a product vastly different than what their marketing department had described.
Two weeks ago, I penned a pretty basic article discussing something that your average Assetto Corsa player might not know about in regards to car setup – Optimal Tire Pressures. Despite all of the hype surrounding Assetto Corsa’s continuously evolving and highly complex tire model – which will be updated to Version 10 in the coming weeks – the internal files for each car told a much different story. Busting open the tyres.ini file for any car residing within your Assetto Corsa folder allows you to hunt down a value labelled PRESSURE_IDEAL, and this number dictates the exact air pressure you want in your tires while the car is turning laps on track, as it’ll help you achieve the maximum grip possible from each tire.
It’s not rocket science by any stretch of the imagination, and most modders already knew about this, so it’s not exactly a revelation either, but what it did do was significantly cut down on the amount of time spent fiddling with tire pressures in the garage menu. If a car’s ideal pressure was 33 PSI, you worked backwards and adjusted your cold pressures inside the setup menu to ensure once the tires would warm up from driving, it would hit exactly 33 PSI. I’m aware there are a few third party setup applications that convey this info in a much more user friendly manner, but at it’s core, it’s equivalent to looking up the answers in the back of your math text book. Forget about feel, forget about heat cycles, forget about anything you’ve learned from watching real life auto racing and how teams find the right pressures to run for each track, just look up this number and adjust accordingly.
So during my afternoon spent decrypting the content in Assetto Corsa created by Kunos Simulazioni themselves, and building a mini database for the article, I discovered that all GT3 cars in the game – regardless of the year they first hit the track – essentially ran a spec tire that used 33 PSI as the ideal pressure value. And I didn’t really have a problem with this, as it ensured fairness among all the cars. Sure, the cold setting was definitely on the edge of what both Michelin and Pirelli suggested to be safe values, but I can’t exactly get mad at a single PSI difference. However, this was only one piece of the puzzle. Pirelli and Michelin gave us the minimum value to ensure a customer’s safety. Nowhere on their website did it list an ideal operating pressure for their respective racing slicks, so naturally I assumed Kunos must have gotten this magic 33 PSI number straight from a GT3 team – falling in line with their whole “we use real data for each car” thing.
I was pretty shocked when I hit up the team manager of GAINESCO/Bob Stallings Racing Terry Wilbert, and received a much different value.
A one or two PSI difference between a modern PC simulator and the real thing is fine, as even the most advanced computer software can’t replicate everything down to the final near-insignificant detail, but a 7 PSI difference is way off the mark for a simulator which has claimed to use real world data. And obviously, there are questions which arise from this discovery. I don’t doubt that Kunos Simulazioni obtained legitimate data for some of the cars featured in Assetto Corsa, but this is a pretty big gap.
So I’d like to know what’s going on here, and obviously I’ll leave this up to the readers of PRC.net to discuss. It really wasn’t all that difficult to pick a North American GT3 team at random from the most recent set of results, visit their website, find an email address from somebody vaguely related to the team, and fire them a quick email both introducing myself and inquiring in regards to what tire pressures they’re running in their GT3-spec McLaren 650s. Physics don’t change when you cross state lines, nor do they change when you run races on different continents, so I’d like to know how a developer who made their access to real data a selling point, ended up being 7 PSI in the wrong direction. Is it a European thing to over-inflate your slicks? Was the team Kunos received data from experimenting with a drastically different setup approach? Or, in a situation that will upset the Assetto Corsa fanboys, is the real data advertised by Kunos just a marketing gimmick operating on a technicality or two?