Performance Ratings [TS040 etc - see post #7]

The P1 has a higher PI than most of the race cars but in reality is should be much slower, going off the top of my head at Spa the P1 did a 2:32, the WTCC Civic was in the 2:23/24s, the GTE/GT3 are now in the 2:16s at Spa and GT1s used to get to the 13s.

Chris Harris has the best time around Portimao an the P1 at 1:51 on the track rubber and the other were around 1:52/53 mark, yet GT3 cars where in the 1:45s and the old WTCC cars (S2000) which are a second or 2 slower than the current BTCC cars could do 1:54s.

The Caparo is in low P class the GT500 cars should be, yet around the Top Gear test track the DBR9 was 2 seconds quicker.

And at Rockingham Gordon Sheddon lapped a McLaren 12C 3/4 seconds off the NGTC Spec Civic (BTCC).

Here are some times from Laguna Seca.
918 1:29
P1 1:30
AMG GT, 12 C, Aventador 1:35

BMW Z4 GTE 1:22
the GTD cars ( GT3’s with clipped wings and poor tyres and are around 911 cup pace) 1:28
IMSA GS (like GT4 and as fast as the BTCC cars) 1:35

Sorry about all the random times, but is shows that cars like the WTCC Civic and GT500 GT-R are massively far from the real thing, the STCC Volvo too but I don’t have any data, but its just a Megane Trophy and should be rated higher in both speed and handling. I Know people will say its the driver but I can get the same time as Randy Pobst around Laguna Seca, yet the race cars find them selves around 5-10 seconds off the IRL times.

My take on this is that if cars were upgraded/tuned to their real world counterparts when delivered there would be nothing for the tuners to do. So the initial PI is not the end all/be all for a car, but the starting point to then work with. Where is the fun in getting a car already built ready to go as is without any planning and experimenting? Forza has to appeal to a large community and that community has varying interests. I’m not into designing liveries but respect that it is in the game for the people that do, same with tuning, multiplayer, etc. Part of the fun is that the game is not the real world and so there are more possibilities that can be explored. The 787 was poor in real life but it is a powerhouse in the game because of that open potential the game provides. I’m probably naive in my view and I don’t know a lot about racing in the real world, but that lack of knowledge allows me to take the game at face value and just enjoy it as the hobby it is for me.


I would agree with this is there was any semblance of consistency and rationality behind the “stock” PI of the cars, but there’s just absolutely no rhyme or reason to some of these figures. There are plenty of cars which relative PI far exceed “built and ready to go”. I’m all for tuning, but without a realistic/fair baseline what’s the point? It’s all predicated on false or made-up figures.

The appeal is realism and difficulty to me. Anyone can take the restrictor off the 787 and put new tires on, that’s not really tuning in my eyes.


Commenting on the bold part here. I guess I’m one type of player you don’t know yet. :slight_smile: I’m bad at tuning and I personally don’t care about it at all. I just see it as a necessary evil that comes with the game. What I care about is the driving/racing experience and the collector in me enjoys collecting all cars and filling up his garage. I want stock FM cars to be as accurate as possible to their real life counterparts, behaviour and lap times. I have no interest whatsoever in tuning and tweaking a certain car. What I do love though is compare different stock cars with each other and see where the differences are situated. Last week I spent several hours test driving the stock versions of all 60s Grand Prix cars, all untuned and no assists. If a car is hard to drive IRL, I want to experience that in FM6!!! I don’t want to tune it away. I’m the type of guy that even looks up cars at Wikipedia to find out if it has a stick shifter, pedal shifters or has an automatic gearbox and I set my ingame transmission to the car I drive. I’m ALWAYS in cockpit view as well. I’m also the type of guy that looks up authentic photo’s of classic cars and then I give my FM car the same colour. I want the driving experience to be as real as it can get… without caring of being forced to care (sigh :frowning: ) about tuning.

I have all respect to tuners and I know there is a lot of you out there. But just know there are people like me who just want to collect cars and drive them.

Taking the restrictor off and putting tires on is ‘upgrading’ rather than tuning but I get your point. The question would be are the PI false or made up and if so by who? Are they what the manufacturers give Turn 10 to work with, are they created by Turn 10 for their own reasons or some combination of both? Realism is a relative term. For example, if realism was actually applied you could only race in cockpit view and all the people that love to smash their way around would have broken cars that generated massive repair bills and violations and banning, etc. etc.
Forza is not a hard core simulation, no matter what the marketing department tries to say and so there will always be trade offs between reality and playability for the largest group possible.

1 Like

We’ve explained our PI system somewhere in these forums before, but I’ll try to sum it up as succinctly as I can:

Forza uses a classification and “Performance Index” scoring system to fairly balance cars across the game. This PI system refers to a ranking number based on a car’s best lap time around a simulated test track. The slower the lap time, the lower the PI number; the faster the lap time the higher the PI. This test track is mathematically generated to be representative of all the race tracks in a specific game. We call this a “PI test track.”

PI ranking, however, is different for each game. It is relative to each game’s car list and race routes. Each new Forza title gets another revision of this test track to be a good representational average of all tracks in the game. This, in addition to the changing and growing car list with each title, is why class boundaries and PI #s are not the same with each new version of the game, or between Motorsport and Horizon.

Forza Motorsport titles use an entirely different PI test track than Forza Horizon titles, because race tracks (like Silverstone, for example) are not representative of the types of race routes found in Forza Horizon’s open roads. Horizon’s roads tend to have longer straights, faster turns, and less hard braking and cornering in general. For this reason, the actual class letters and PI values for a given car are not the same between Horizon and Motorsport games. Additionally, Forza Motorsport has Formula 1, Indy, and modern prototype race cars at the top of its PI system, while Horizon 2’s top-end cars are “hypercars” like the McLaren P1 and LaFerrari. This means a PI of 998 in Forza Motorsport is a much higher level of performance than a PI of 998 in Forza Horizon. It also means less granularity is needed in Horizon’s car classes than we have in Motorsport. Again, it’s relative to each game’s cars and race routes.

Once we have PI test track lap time data for every car in the game, we analyze it to determine class boundaries. Most of these land exactly where we would expect, based on the fact that all the performance data for each car is based on extensive real world research, including data provided to us directly by car, tire, and aftermarket parts manufacturers as well as race teams.

We rarely hand tune PI, and if so it is to maintain fair game balance across career, multiplayer, and leaderboards, or to keep cars of similar type and performance in the real world in similar PI ranges or classes, and never to a degree that a car’s performance characteristics are significantly compromised compared to its real world counterpart.

Lastly, of course we allow players to upgrade and tune their cars to be competitive with cars that have a much higher PI in stock form, within the bounds of real-world limits. In multiplayer or rivals competition, it’s not uncommon to see cars which, in stock form, would have no chance of winning, but with a great tune (and driver!) can smoke the competition.


Well,so with the return of GT500 and Group 7 cars and some ultra-advanced hypercars like T1 and One:1, can we expect to get another PI class sit between R class and P class (like R2 class at XBox 360 Motorsport series) to justify the performance difference between this cars? I mean it maybe too late for FM6 but can we expect to get this special PI class introduced into the next Motorsport series title?

I never could figured out why T10 decided to cut the PI class between GT3 and Prototype with FM5 but since we lose all of DTM/GT500/GT1 of late '90 and most of GT1 after 2000 and Track-Only cars like Zonda R it seems fair enough to not put another PI class in as no car may use it but with the ever-growning car list we have now it’s time to reconsider add this PI class back.

I like when Jon explains the inner workings of Forza :slight_smile:

Last time PI calculations were discussed, I believe Dan Greenawalt mentioned that the test track was vaguely similar to Silverstone.

If you could choose one of the tracks in Forza Motorsport 6, which would have the nearest characteristics to your current test track?

1 Like

I haven’t seen the current test track used for FM6 myself, but I imagine it is still somewhat similar to the older one (vaguely similar to Silverstone, as Dan had said), perhaps with some new sections or adjusted sections to represent unique characteristics from new tracks we’ve added since then.

^added to the FAQ thread:

In all honesty I think T10 overall did well in balancing the classes when compared to other iterations. There’s still a handful of cars that probably should have their PI manually adjusted in an effort to balance the classes, but that would be an exception rather than a rule. The 787b would be an example; however, some prototypes can somewhat keep up this go around which was nice to see.

1 Like

As long as the PI value in the game accurately indicates the in-game performance, I don’t see this as a huge issue. After all, there are cars from multiple eras in almost all the race classes that would never compete with each other in reality, but have to be competitive with each other in the game.

Sine T10 is commenting on the PI system, I’d be curious how PI increases due to mods are calculated though, since it sounds like ‘stock’ PI is based on a calculated simulated lap time around a theoretical test circuit. The effect of every mod in every combination on every car would be excessive to compute as a unique test lap (it would seem), so there must be some sort of less complicated calculation for how a given mod influences the current PI when doing upgrades.

1 Like

Career Mods are gameplay modifiers which work outside of the PI system (PI is entirely driven by real-world research data). Mods don’t change PI; they use gameplay scalars that add or remove hp, weight, grip, brake friction, etc. on top of the car’s current performance, whether stock or upgraded. That’s why performance mods don’t let you post clean lap times, and why they aren’t allowed in MP or Rivals. You can, however, see the effect mods have on a car if you bring up telemetry.

Sounds like he meant parts, not “mods” as such.

Indeed I did, but that’s completely my mistake for using the word ‘mods’, which has a specific meaning in FM6 career mode. Either way, the response from T10 is interesting and appreciated!

I get the PI, but there are some discrepancies that can’t be rectified via building or tuning. Case in point would be the Ford GT40 and Ferrari 330P4. As you can tell by my signature, I’m a bit of a fanatic for vintage racing. I don’t have my numbers right in front of me, but I believe that the curb weight for the 330P4 was in the neighborhood of 1800lbs, while when building to spec in game it’s weighs over 2000lbs. And both the 330P4 and GT40, were able to reach over 210 mph with 450 and 485hp, respectively, but in game neither barely reaches 200 with the same output.

Those are only two examples, but they are the two that come to mind immediately.

I share your fanaticism for vintage racing, but my biggest gripe is that we don’t have the GT40 Mk I in the game. The first car (same vin#) to win back-to-back 24hr Le Mans in 1968-69. It was a five-year old race car at that point which had little chance of beating the new, more technically advanced and scary-fast Porsche 917. Driver Jacky Ickx was the only driver to WALK to his car in protest of the dangerous running start, and thus started at the back of the field, and still managed to win in that glorious old beast.

Wait, I’m breaking our own forum rules…This sounds like a car wishlist item!

Even we at Turn 10 don’t always get the cars we want :wink:

1 Like

Aha, so you’re the guy who insists that a different quarter panel makes it an entirely different model that needs to be in the game!


I don’t think the problem is with the PI ratings but with the performance some of the cars. Like the GTE/3 cars being slower than the Hypercars, or the WTCC Civic, STCC Volvo and GT500 Nissan being slower than they should be, or both Mazda P cars being faster than they should be.

Hey guys,

Does anyone know if Forza / Turn 10 is going to correct the PI on the TS040? Because my game says the Rebellion R One is better, which I find odd?

Apart from that I love the car so much, it’s perfect, I’ve had no issues against others in other cars, because that’s the beauty of LMP1. But when the R One isn’t a hybrid or works car the stats don’t add up.

Anyway thanks if anyone has the answer!