PP Standardisation

[srefre]

I was thinking about PP, and I thought how it would be more useful if there was a method of standardising PP for each track.

On an official level, it would be cool if a car had a base pp which was then modified based on the track. For example two cars of equal pp one light, nimble and low powered and one heavier, less agile but more powerful.

On a balanced track, they might remain equal. On a tight track like autumn ring the lighter cars track specific pp will increase and the heavier decrease to reflect the (dis)advantage. On a big open track, oval or La Sarthe, the reverse will happen. What do you guys think of this concept?

Unofficially has anyone come up with a spreadsheet or app to do this already, or does anyone have ideas for a formula to do it?

 

[sk8er913]

I was thinking about PP, and I thought how it would be more useful if there was a method of standardising PP for each track.

On an official level, it would be cool if a car had a base pp which was then modified based on the track. For example two cars of equal pp one light, nimble and low powered and one heavier, less agile but more powerful.

On a balanced track, they might remain equal. On a tight track like autumn ring the lighter cars track specific pp will increase and the heavier decrease to reflect the (dis)advantage. On a big open track, oval or La Sarthe, the reverse will happen. What do you guys think of this concept?

Unofficially has anyone come up with a spreadsheet or app to do this already, or does anyone have ideas for a formula to do it?

It's almost common sense really... A kart will be faster than the equivalent racecar at gran turismo arena or kart space, and at the Nurburgring, the racecar is faster than the kart. I don't think we need this... although you can post it in the suggestions thread if you would like.

 

[TenEightyOne]

It makes sense but the driver would always an arbitrary factor.

 

[srefre]

For sure it's common sense, and the driver will be a factor.

I was really getting at trying to have equal races, without being 1 make. It's easy to set a pp limit, but this doesn't show the full picture.

It would be nice to be able to set 2 cars to be equal for the same track. That might mean that the actual pp of each car is different, but the track pp is the same.

 

[t3kc0h]

For sure it's common sense, and the driver will be a factor.

I was really getting at trying to have equal races, without being 1 make. It's easy to set a pp limit, but this doesn't show the full picture.

It would be nice to be able to set 2 cars to be equal for the same track. That might mean that the actual pp of each car is different, but the track pp is the same.

It makes sense but the driver would always an arbitrary factor.

I agree with this fully.

As for the driver being a factor, that's the point of having an even playing field. It's to show who really has the advantage in skill. Not who has the better car.

 

[TenEightyOne]

I agree with this fully.

As for the driver being a factor, that's the point of having an even playing field. It's to show who really has the advantage in skill. Not who has the better car.

True, but what I'm trying to say is that arbitrarily saying "this car is worse at this track by n" isn't possible due to the differing nature of individual drivers. An arbitrary per-car-per-track performance bias could actually disadvantage some drivers depending on their own strengths and styles.

I don't know what the solution is, just what it isn't

 

[t3kc0h]

True, but what I'm trying to say is that arbitrarily saying "this car is worse at this track by n" isn't possible due to the differing nature of individual drivers. An arbitrary per-car-per-track performance bias could actually disadvantage some drivers depending on their own strengths and styles.

I don't know what the solution is, just what it isn't

Well I don't know how to respond to that and I'm not gonna act like I know what I'm talking about, but you do have a point

 

[danielc]

This is partly "our job" to discover cars better suited for each situation.
Said that i believe there should be more indicatives of performance, like acel, braking, handling, etc

 

[SyzygyDarklock]

As above post it really is "our job"... Moreover what about adjusting for driving styles, tuning, player's skills or use of aids chassis reinforcement, wings??? You'd be there 'till Doomsday making adjustments for everything... I do see your point though

 

[Morgoth_666]

I would prefer that PD focus on getting the PP system to function apporprately overall so that there are no longer obvious issues with cars always being much faster or slower than other reasonably similar cars at the same PP in virtually every situation.

Once they have it so that you can be fairly certain that equal PP cars will be at least similar on most tracks then they can start worring about adding another layer of incredibly complex and varied calculations to determine exactly how a given track will affect a given car's performance. If even then.

 

[eran0004]

It's an impossible ideal to dream of a performance rating system that would make every car competitive. It can't be done, because some cars are simply better than others and no matter how you tweak the rating system, it will always be unfair in one way or another.

It's like having two students in school, asking "which one is the best student" and measure it by taking their grades in a dozen different subjects and calculating an average. Even if they have the same average, one may be excellent in chemistry while the other is an ace in latin. If you give them both a test in mathematics, chances are that they'll score very different.

 

[Johnnypenso]

It's an impossible ideal to dream of a performance rating system that would make every car competitive. It can't be done, because some cars are simply better than others and no matter how you tweak the rating system, it will always be unfair in one way or another.

It's like having two students in school, asking "which one is the best student" and measure it by taking their grades in a dozen different subjects and calculating an average. Even if they have the same average, one may be excellent in chemistry while the other is an ace in latin. If you give them both a test in mathematics, chances are that they'll score very different.

My thoughts exactly. There are too many variations in grip, power and weight to equalize all cars at all tracks or even come close.

What's more realistic is logical car groupings, within which PP is more equalized. For example you might have one called Supercars (or whatever name works) that would have the 458 Italia, Aventador, Enzo etc. Another might be sports cars, sports sedans, tuner cars, wagons, FF's, Rally Cars, GT500 etc. all cars with fairly similar weight, power etc. That way you could tweak the PP within each group to allow for the small differences between the cars, rather than designing one huge all encompassing calculation that applies to all cars.

 

[srefre]

Or how about breaking the PP down into components?

So if you have 2 500pp cars, you can see the components that contribute to that 500pp - e.g. 100pp acceleration, 200pp top speed, 150pp mechanical grip, 50pp aero grip. Ok, so I just made those components up without much thought, it's more about the concept.

 

[snowgt]

Or how about breaking the PP down into components?

So if you have 2 500pp cars, you can see the components that contribute to that 500pp - e.g. 100pp acceleration, 200pp top speed, 150pp mechanical grip, 50pp aero grip. Ok, so I just made those components up without much thought, it's more about the concept.

You could do that, but it won't help much to equalize cars easily on any given track.

• One component is power and torque. The integral of these curves are clearly one parameter of the PP calculation. You can try the following: Take a car where power and torque curves are very similar, something like a normally aspirated engine with peak power and torque being at or close to red line. Then note down the PP figures when you restrict the engine's power in certain intervals, so you can draw a graph. You will see how low the PP's drop when you restrict the engine to 50% and if the PPs drop linearly or exponentially. (The influence of power restriction will be different, if you do this with cars where torque peaks down low and power on high revs - the influence of power restriction will be much more unpredictable)
  
• Now do the same with adding ballast. Note how much PP you lose by adding weight in 5kg or 10kg steps. You can extrapolate the curve beyond the 0kg-200kg range then. (by a reasonable amount - at "-500kg" you surely will not get reasonable results)

These are the 2 known parameters of the PP calculation that are known. From these 2 you can deduce how much of the PP figure other factors make up. The other parts you cannot test that easily or alter them though. Aerodynamics do not factor in anymore in GT6, and that's reasonable, because downforce and drag have been modeled much better now, and one benefit gets outweighed by the other when you fiddle with wing settings. That's not to say that it won't make you faster on certain tracks, but having one PP figure, it's the better approach. Traction also seems to be one component, but it's hard to pin down exactly how.

Of course, not knowing how PPs exactly are calculated, there will always be a mysterious part that you will not be able to quantify.

Note: above tests are simplifications and will only produce crude approximations. I'm by no means claiming this can be used to deduct anything more than that.