Homologation & PI Balance/Accuracy

[MLX_Lunar]

If the new Motorsport keeps homologation, I wanted to start this thread to discuss how T10 could build upon PI accuracy and the PI balance for ‘Class’ and ‘Homologation’ criteria.
It’s evident to me that balancing cars in their respective PI and upgrading them to be perfectly balanced with other cars at the same PI maximum is a very hard task to complete. We all know there are “leaderboard” cars that simply dominate a multitude of tracks in their class and, in some case, dominate their homologation category altogether.
One present, clear example is the Alfa Romeo 33 Stradale competing as a “leaderboard” in multiple classes, filling up the Rivals top 5% by a large margin. This was worse on FM6, and has improved since the release of FM7.
This thread is to simply discuss ideas on how to measure weight, drivetrain, tire width, car length etc.

Does this feature need to be heavily focused on? Or is it balanced enough already that the new Motorsport need not focus on balance to a tee?
Do current players enjoy having a “top 5 leaderboard” car to pick? Or do players wish to have 20-30 options that have a chance to compete?

What do you suggest?

(In no way am I criticizing their methodology, however I am interested to see responses and ideas from others who have played many years like myself.)

[BookishFlunky90]

Things that comes to mind

• how does it work in real world races?
• there are not that many classes with the same car and model
• it’s about the driver to larger extent
• the more racing games become simulation the driver will be more important

When doing online races this will sort itself, but how should AI work offline?

What is the aim with 700 cars in a game?

• should they all be in garage and give collection tier points?

Common properties are

• speed
• handling
• acceleration
• braking

In real world I think classifications are not on those properties, more engine horse power, weight, measurements between tires etc. One team will optimize car better for that track and be on top leaderboard.

But think they did a reasonable good job that resulting properties classifies and driver makes the most difference.

More questions than solutions, just some thoughts.

As I got FM7 just a month ago, what you face are when looking at upgrades is what implications will it have if I move to another class for that car. It warns as you look at upgrades. But also confuse as it says limit in engine power, and well below max you still cannot upgrade.

So whatever they do in next FM maybe explain a bit more how it works.

• can you upgrade car but still adapt and downgrade later to participate in a race?

[Rainbow_Prisma]

It seems to me that the mechanics in the Motorsport series rely on the weight of the car, the chassis length, and tire width primarily for how cars handle.
Torque and HP numbers rely on the suspension and body roll from turn response during acceleration before the G’s cause tire slip.
With all of this being said, the PI numbers seem difficult to accurately balance in all 600+ cars (maybe 700 in new release).
Since this is the case, and I hear this topic come up a lot about “leaderboard cars”, I think the PI numbers should be focused most towards engine swaps and drivetrain swaps that can come from a good 50%+ of the cars.

Considering that some stock engines throughout history have an extra 200+ horsepower through aspiration and mods, yet still be reliable enough to race a few races without blowing up, I think the PI system should allow stock engines to have a slight bit more power than what was given to us in Forza Motorsport 7, as well as 6. Notibly for some reason, the 3.2L VVT engine swap was always a go-to swap for my builds because of the crazy horsepower increase gained with enough PI left over to do some more grip upgrades.
I’m hoping that we don’t see overpowered engines as often in the new release and I’m also hoping that PI can be heavily studied so that they can balance drivetrain grip for classes too.

For most drivetrain swaps that I’ve made, the All-Wheel-Drive setups usually held the best acceleration out of turns with the right tuning (no assists used).
For almost all of my Front-Wheel-Drive cars, tire wear and slippage came the easiest out of turns and tire wear came faster. Now this is realistic.

The problem isn’t exactly the realism of tire wear and how much we need to concentrate on throttling the pedal. To me, the problem comes in how the PI numbers added up once a drivetrain was swapped. My honest opinion is that Rear-Wheel-Drive and Front-Wheel-Drive setups seemed like slugs in comparison to All-Wheel-Drive when exiting a corner. Along with this, at most times would come less tire wear as well.
On most tracks, the straight are long enough to compensate for the time lost on turns, using an All-Wheel-Drive car. Only 3-5 tracks have enough turns in them that cause All-Wheel-Drive to be essentially useless.

To summarize:

• The PI seemed to not add enough points towards the All-Wheel-Drive (AWD) swaps for its designated class. (example: B-600) It left a slight bit too many points for other upgrades.

• The PI for Rear-Wheel-Drive (RWD) and Front-Wheel-Drive (FWD) seemed to be sluggish on corner exiting in comparison to AWD in its designated class. (example B-600) It left too few points for other upgrades.

• Some engine swaps left too many PI points for their racing class in comparison to the stock engine that came with the car. (swap example - 3.2L VVT) This engine and a few others gave too much HP in comparison to what is realistically capable from stock engines in the same class.

• And lastly, not enough engine swap options were available in FM7, however I assume it’s difficult enough to balance the cars PI as it is already. It still would be nice.

[FRA_FancyFace]

I completely agree, the current PI system as is favors especially old, light cars too much. Some of these cars have a stock tyre compound that are in fact slicks. These cars can be found in the ‘Vintage GT Racing’ division (Alfa 33 Stradale, Nissan R380, …). And I’m fine with that, but with how the current system is set up, these ‘stock’ tires are worth less PI than an ‘upgraded’ compound, and that’s where things are starting to get messy.

Now, a lot of these cars (and a lot of muscle cars as well) have the option to change the tranmission to upgrade it to a street transmission (mostly because a lot of them are 4 speeds), also a really good thing, except for the fact that when you upgrade said transmission you get PI back.

All this is without even swapping the engine for a ‘3.0L V8 - Racing’ engine, which in the R380’s case only costs 119 PI for 258 HP, not only is there a massive spike in HP, it also makes the car 198 lbs lighter.. So you’d end up coming from stock B570 PI → A689 PI.

Now, you can also upgrade all of the engine parts for 1 PI, gaining another 72 HP.

Well, the sport tranmission used to cost 1 PI before engine upgrades, but since you’ve upgraded them, you’re getting it for free.

Oh you want lighter rims? Sure, let’s take off another 17 lbs free of charge.

Strangely enough this is not even the most broken car in the game, there are certain cars where you’re getting PI back for upgrading engine parts.

Experienced Forza players will find every glitch (for lack of a better word) and will utilize it to it’s fullest potential, that’s part of this game.

The Lotus 340R for example completely dominates handling tracks from B through S class.

Now let’s talk about the opposite side of this issue, heavy cars. Most cars above 2500 lbs in A/S class (I’m not too experienced in other classes so can’t speak for those) are completely useless against a regular lobby car. Even though some have some serious speed in them (> 200 mph for 6.4H and more), it takes way too long for them to get up to speed. Not only will the lighter car handle better in corners, it will also have the upper hand in accelerating out of them. There are a few exceptions to this rule, mainly the Vipers (Dodge '08, '16 and SRT '13), but also the Ford Mustang GT350R (and its FE counterpart) and a couple of others.

So yes, there’s a lot of stuff to improve on the PI system. And yes, some cars are going to be faster than others, that’s normal, and in the new game we can make use of the ballast to balance it out (I’d prefer incrementing by 50 lbs instead of 100, but that’s a different topic). But that ballast is only going to be of use in custom, private lobbies, this won’t help for dominated leaderboards or public lobby racing.

Apologies for the long post, and there’s a good chance it will never see the light of day but it needed to be said.

[Swergenglergen]

The PI of any car in the game, with any combination of parts, is determined by a simulated AI run around a test track, by my understanding at least.

So the problems likely sit with how that AI is able to control certain car characteristics, and hence why some upgrades give lots of power but at little PI cost; because the AI can’t use the extra power.

HokiHoshi just mentioned this in a YouTube, looking at the AI improvements of the new FM, so seems plausible that just the AI update will sort out a lot of the past odd PI results.

What worries me is that a better AI can probably control a RWD even better now (new AI having controlled throttle vs old AI having only binary throttle), so could widen the gap between AWD and other types of car.

Another major issue from my experience with the FM series PI calculation is that the AI seems to use the DEFAULT tune, hence if the upgrades selected unbalance the car, the PI goes down, but the player can then tune the car back to fast without affecting the PI, so in effect cheat the PI system by selecting deliberately unbalanced and bad default tuned parts.

A few solutions that I can see:

1. Races could be both PI limited and a certain drivetrain requirement.
2. For open races (ie any car within a certain PI); the PI could be recalculated with updated tuning (making the racing more about skill than messing with the PI system flaws/tuning).
3. To retain the fun of tuning cars, have single-make and controlled parts/spec races, where players compete through tuning.