Effect on Tyres,Horsepower, AND weight on A-Spec Points: a 'Scientific' Study Updated

[D2 Convection]

Wow. Super informative. It's too bad that now that I know how to get a lot of A-spec points, I don't know what they are used for in the game besides be a growing number...

 

[ZeratulSG]

The opposition lineup doesn't improve to compensate for changes that you make to your car. That would defeat the purpose of the point system. Now, it's VERY common to have lineups where the opposing cars aren't stock.

I beg to differ. In races where the opposition is NOT stock, they often seem to change to compensate for your car. I experienced this last night in a very frustrating way (maybe you guys can shed some light on this for me).

Okay, it was the 206 manufacturer's event, and I was running a 206 RC. All tests were done in race 1 at Opera. In all cases the grid was more or less as follows:

3x 206 S16
1x 206cc
1x 206 RC

Occassionally, one of the S16's would be replaced with another 206cc, but this didn't seem to make any difference, as I'll illustrate here.

First I tried with my car stock + oil change. 200 pts. and by the end of lap 1, I was almost 10 seconds down and losing ground every second (I was 20kms slower at the end of the back straight before the Firestone corner, the fast left-hander, with identical exit speeds from the previous corner).

Right, so I figured they were running modded 206's, no problem, I pushed the car up to around 230bhp and tried again. 200 pts., similar results.

So I threw some more money at the car, tuned it up to around 270bhp and tried again. 200 pts., similar results. At this point, I had almost completely tuned out any understeer in the car, and had adjusted the suspension for the bumpy Opera track, both of which I doubt the AI does (I would imagine they use the default setup for any parts upgrades like suspension). The car should have been cornering faster than the AI, so even if I was losing time on the straights I should have been able to keep up. No dice. The replays showed the AI corner speeds matched or exceeded mine, and their speed on the straights was still superior. My splits at the end of lap 1 remained more or less consistent throughtout this test, give or take a second for driver error or taking a faster/slower line etc.

Oh, and before someone suggests I drive better, let me just say that I may not be the Michael Schumacher of GT4, but I can hold my own, so please don't put this down to simple bad driving.

At that point I had to hit the sack, but it was very frustrating to experience this. Since you guys seem to have the measure of the whole A-Spec points system (or at least, a lot more than I do) maybe you can suggest why this is? Is the AI simply retuning to compensate for my mods, thus keeping the split (and the points) the same?

 

[Mt. Lynx]

I beg to differ. In races where the opposition is NOT stock, they often seem to change to compensate for your car. I experienced this last night in a very frustrating way (maybe you guys can shed some light on this for me).

Okay, it was the 206 manufacturer's event, and I was running a 206 RC. All tests were done in race 1 at Opera. In all cases the grid was more or less as follows:

3x 206 S16
1x 206cc
1x 206 RC

Occassionally, one of the S16's would be replaced with another 206cc, but this didn't seem to make any difference, as I'll illustrate here.

First I tried with my car stock + oil change. 200 pts. and by the end of lap 1, I was almost 10 seconds down and losing ground every second (I was 20kms slower at the end of the back straight before the Firestone corner, the fast left-hander, with identical exit speeds from the previous corner).

Right, so I figured they were running modded 206's, no problem, I pushed the car up to around 230bhp and tried again. 200 pts., similar results.

So I threw some more money at the car, tuned it up to around 270bhp and tried again. 200 pts., similar results. At this point, I had almost completely tuned out any understeer in the car, and had adjusted the suspension for the bumpy Opera track, both of which I doubt the AI does (I would imagine they use the default setup for any parts upgrades like suspension). The car should have been cornering faster than the AI, so even if I was losing time on the straights I should have been able to keep up. No dice. The replays showed the AI corner speeds matched or exceeded mine, and their speed on the straights was still superior. My splits at the end of lap 1 remained more or less consistent throughtout this test, give or take a second for driver error or taking a faster/slower line etc.

Oh, and before someone suggests I drive better, let me just say that I may not be the Michael Schumacher of GT4, but I can hold my own, so please don't put this down to simple bad driving.

At that point I had to hit the sack, but it was very frustrating to experience this. Since you guys seem to have the measure of the whole A-Spec points system (or at least, a lot more than I do) maybe you can suggest why this is? Is the AI simply retuning to compensate for my mods, thus keeping the split (and the points) the same?

AI does not retune to compensate. All cars in that particular race are tuned to match the 206 rallycar (300+ bhp), wich is also eligable to enter. Ad a wing to your S16, and you will see a difference in cornerspeeds. I can´t remember my A-spec points for the race, but it was not down to 1. And I think you´re unable to get just 1 point unless you enter a tuned rallycar...

This thread is really interesting, but can someone try to sum things up, as far as it goes right now? I´d say it´s easier to see the whole picture in a summary, than spread out like this. I don´t want to do it myself, since I´m not the expert here...

 

[neilX]

I beg to differ... The replays showed the AI corner speeds matched or exceeded mine, and their speed on the straights was still superior. My splits at the end of lap 1 remained more or less consistent throughtout this test, give or take a second for driver error or taking a faster/slower line etc.

Oh, and before someone suggests I drive better, let me just say that I may not be the Michael Schumacher of GT4, but I can hold my own, so please don't put this down to simple bad driving.

It is my assumption here that they had better tyres fitted than you did.

I'm currently operating under the assumption that each car has a *base* ASpec value, something that was pre-calculated and stored on disk. I'm assuming this for two reasons - 1) it makes sense for a programmer 2) it would explain some inconsistencies. If power/weight was calculated on the fly, the inconsistencies are much harder to explain. So... Occam's Razor and all...

I'm not so sure about that Base Value.
I can't see how that would be of any use to calculate the A-Spec points when they must be calculated using your own power:weight and the power:weight of your opponents. You can change your amount of power drastically and enter it into a wide variety of races against far lower or higher power cars. Every time you start the race, the system must calc it there and then, as you could have made many changes to your power/weight/tyres/downforce.
A base value would indicate that a given car is, by default, better or worse than another given car, regardless of mods/tyres/weight... and that doesn't sound right to me.

 

[quati]

great graphics, great datas!!

 

[STINGER05]

This MIGHT be a really stupid question, but has anyone thought to try and get ahold of PD for an interview of some sort and have THEM explain how all this stuff works? :

I think this would be the right way forward, as the not even the GT4 Bible gives clear explanations and calculations.

I also hope that PD is aware of this forum as a whole, as it will give them a lot of ideas for the development for GT5 and what people like/dislike about the game, its set up, cars, tracks, etc....

💡 May be Der Alta or another moderator will be able to contact PD and make them aware of the existence of GTP? Or may be we should all contact PD and let them know of GTP and/or what we think....? or am I too optimistic?

 

[jdw]

At that point I had to hit the sack, but it was very frustrating to experience this. Since you guys seem to have the measure of the whole A-Spec points system (or at least, a lot more than I do) maybe you can suggest why this is? Is the AI simply retuning to compensate for my mods, thus keeping the split (and the points) the same?

No, the computer isn't modifying their car to match yours. They start with very souped up cars (with Racing Tries) to begin with. How do I know this? Preview the Race. It will tell you the HP and tires of the opposing car.

My guess is that you forgot to change your tires. Slap on some R3s and you'll be good to go. Just looked at my notes... I used an S16 w/NA3 chip and racing muffler to bring me to 247hp. Added R3s and had 200p per race.

The Speedster Cup is the same way.

 

[jdw]

This thread is really interesting, but can someone try to sum things up, as far as it goes right now? I´d say it´s easier to see the whole picture in a summary, than spread out like this. I don´t want to do it myself, since I´m not the expert here...

Well... unfortunately, there is no summary yet. But that's the goal!

We've got a bunch of data and some ideas and the picture is beginning to come together but it's not quite there. More studies need to take place.

The one summary I can make is that... just about every ASpec point difference seems to be calculated linearly... which is very strange but seems to be the case.

1) Changing tires are a linear relationship (+/-)
2) If i'm reading scraps data correctly, power and weight mods are also linear changes
3) Lineup changes are... linear (the strangest of all )

I'm also guessing that each car has some *base* value that is changed (+/-) according to the opposing lineup, the tires, downforce and power/weight mods.

Why do I think this? Well...

Suspension and Drivetrain mods do NOT affect points. What does this mean? Start with cars that have the same tires (SMs), the same downforce (sports cars == 0) and the same lineup. That leaves exactly TWO variables for the game to play with - power and weight.

IF the computer calculated power and weight on the fly (think STOCK cars now) and used it as its base comparison, then it's correct to assume that a car with a lower kg/hp value will generate fewer A-Spec points than a car of higher kg/hp value.

I have literally dozens of examples where that assumption doesn't hold true. (While searching for 200p races, I'll typically take at least a half dozen cars and compare them against the same lineup).

It would also easily explain why the RAM is a bug... because it has a bad value.

This isn't the only explanation of course, but it's a plausible one, yes?

I'm not so sure about that Base Value.

I can't see how that would be of any use to calculate the A-Spec points when they must be calculated using your own power:weight and the power:weight of your opponents.

That's an assumption. It sure *sounds* sensible but the more I look at the data... the less convinced I am that's what's happening.

Try it out. The more people we get involved in the experiments, the faster we can get an answer!

You can change your amount of power drastically and enter it into a wide variety of races against far lower or higher power cars. Every time you start the race, the system must calc it there and then, as you could have made many changes to your power/weight/tyres/downforce.

Yes. And, so far as we can tell, it just add/subtracts points for any mods that you make. From programming standpoint, starting with a pre-calculated value and add/subtracting a mod from that value and then comparing it to the lineup is probably easier than doing all the calculations on the fly.

A base value would indicate that a given car is, by default, better or worse than another given car, regardless of mods/tyres/weight... and that doesn't sound right to me.

Yes.

Not regardless... we're just talking about comparing cars in their STOCK configuration.

Why doesn't it sound right? You don't think that the designers could drive a car around during testing, evaluate it against other cars and give it a point value... perhaps based on a best lap time?

 

[yeti]

Why doesn't it sound right? You don't think that the designers could drive a car around during testing, evaluate it against other cars and give it a point value... perhaps based on a best lap time?

Hmmm.... you might have something with that... I've also seen A-spec points change slightly per course on a couple of occasions with what I'm sure is the same field and cars etc... so this might imply that the base A-spec value is for car per course?

Or I might have that wrong... and I would expect that the point comparator would be based on a track that has everything in it... i.e. Nur.

Also I notice that the Special Conditions halls always seem to have quite low points. Could it be the case that the base value varies for these tracks too?

C.

 

[neilX]

Why doesn't it sound right? You don't think that the designers could drive a car around during testing, evaluate it against other cars and give it a point value... perhaps based on a best lap time?

Hmm, maybe...

I'm only just getting into the racing now (after going for Gold and Missions) and I'm after high points, so I'll be looking into all this myself more. I'll keep an eye on what comes up on these threads.

 

[Orion_SR]

Excerpt from: B-spec Point Theory and Possible Application to A-spec
https://www.gtplanet.net/forum/showthread.php?t=61761&page=14&pp=20

Testing on Family Cups might provide clues to A-spec car mysteries. The competition of Family Cup races is selected based on the car and tuning entered in the race. By adjusting the difficulty setting, it may be possible to force the game to identify in advance which cars qualify for certain classes. My experience during B-spec testing was that the Dodge RAM 1500 Laramie was an annoying opponent and worth avoiding by maintaining a slightly higher difficulty. If there is a relationship between this car and other big point cars, then the Toyota PRIUS G Touring Selection may be a good choice for A-spec points. (This may provide clues to the “base value” of cars.)

It is my understanding that the online play engine is intact for GT4 and can be accessed by LAN play and internet trickery. If the mechanism intended for scoring online races is related to A-spec points, then it may be possible to control the competition and make more precise measurements of the influence of different upgrades and handicaps. Lag should not be an important factor for decoding points. (To my knowledge, this method has never been tried.)