Calling all mathematicians

  • Thread starter Thread starter acjy1985
  • 56 comments
  • 3,369 views
I think for any solution theese are the main problems to be solved.

I belive there is a non-linear functional dependency of laptime to price. So what I would do, is to plot your measurings into an diagram with x-axis=time and y-axis=price. With linear regression you might be able to draw a line.
Cars sitting below your line would be the sleepers; cars sitting above would be the lemons.

picture.php


Another question is how to sum up the lap-times from various races, because the time-gaps from my observation are quite track-dependant. Longer tracks of course would create larger gaps, offroad-tacks seem to produce a wider variance of lap-times, technical tracks would favour well handling cars, high-speed tracks would favor ... well high speed cars, etc.

So in the very end you might find out that with a lot of effort you gathered no valid information ;)

This is a quite excellent suggestion. I love graphs as well and this would make it easier to see which works out best(furthest from the line). I have 5 tracks, all real world which incorporate relatively slow tracks(Top Gear, Indy Road Course) fast tracks (Monza, Circuit de la Sarthe) and the Nurburgring which is a lot of everything.
 
I am wondering the fairest way to work out the 'best value for money' car. I am testing cars with strict guidelines which keeps it fair. What would be the best way to calculate the result? Say I've tested 50 cars, my idea was to assign the cars points. 1-50 (1 for the cheapest car). Then assign 1-50 points per track (1 for the fastest car). Then the car with the lowest score after all tracks and cost are taken into account is the winner. I have 5 tracks I've tested on and realised that this probably wouldn't be a fair way to compare the cars as the value would only carry the weight of 1 circuit.
I could always multiply the cost score by 5 to give a range of 5-250 points but can anybody more mathematically gifted than myself come up with a better way to score the cars. Please assume my total of 50 cars just for ease of maths.

Cost per second.

a Golf GTI does the lap in 1:15, and the Pagani Zonda does it in 55 seconds (pretending), the Golf costs 40,000cr and the Zonda 500,000cr.

040,000/75 = 0533.333
500,000/55 = 9090.91

So the Golf is 17 times more cost effective. Don't know if this will work, but it's an idea, real figures with very fast and very slow cars (as well as very cheap and very expensive cars, que the Veyron), will be the real test. If you want higher to be better, then make the cr the denominator instead.

Good Luck
 
I'm glad there have been so many suggestions. This theoretically should be an easy thing to achieve, but as there are so many theories it makes it much more complicated and interesting.
 
Cost per second.

a Golf GTI does the lap in 1:15, and the Pagani Zonda does it in 55 seconds (pretending), the Golf costs 40,000cr and the Zonda 500,000cr.

040,000/75 = 0533.333
500,000/55 = 9090.91

So the Golf is 17 times more cost effective. Don't know if this will work, but it's an idea, real figures with very fast and very slow cars (as well as very cheap and very expensive cars, que the Veyron), will be the real test. If you want higher to be better, then make the cr the denominator instead.

Good Luck

The problem is, that this result of the Golf being "more cost effective" will have almost no application in the game whatsoever. It will not give you a win against supercars, no matter how many times you try, so you will just have to buy a car which will let you win.

Somebody suggested to set the car-price in relation to the money that you can win in a certain time-frame. This idea has some merit, but there you have changing seasonal races, so you would have to do it in offline a-spec or b-spec races.
 
Ferrari SP1[/B] - Why you would buy something slower than the 280,000cr. F430 it's based on for 1.5 MILLION cr. is beyond me. Waste of bloody money

Because it is beautiful and unique. Like the Miura or the Alfa Carrazota
 
I've been running through stock cars to see what I can win with... just won a 500PP seasonal fairly easily for 400k+ in a stock Holden Commodore. I think so far that's the cheapest car I've won that race in, but I'm still working down my list.
(yep, stock... oil change, nothing else, not even tires)
 
A few fast cheap cars, Tommykaira ZZII for Handling, Supra MKIV RZ for Top Speed, Viper SRT-10 ACR for Acceleration/Top Speed, Camaro RM for the BA Factor.
 
I belive there is a non-linear functional dependency of laptime to price. So what I would do, is to plot your measurings into an diagram with x-axis=time and y-axis=price. With linear regression you might be able to draw a line.
Cars sitting below your line would be the sleepers; cars sitting above would be the lemons.

picture.php


So in the very end you might find out that with a lot of effort you gathered no valid information ;)

That`s the German way of engineering. It´s a gift and it is a curse!
:lol:
 
[QUOTE="sprplay"

So in the very end you might find out that with a lot of effort you gathered no valid information ;)[/QUOTE]

This. Lol.
 
How about you undergo a series of tests (presented to you in a golden envelope by a man in a white coat) where you write the results on a large cardboard table. You can have a benchmark lap set by the ethnic cousin of a racing professional and award a nominal number of points based on how close you get. Then the winner is the one that isn't James May...
 
An idea: Blind testing from an average. It gives you a value, a meaningless number, but a value.

50 Cars, 10-15 tracks. Assuming use of a spreadsheet.
Steps:
1- Test each car for 5-10 laps. Take the average lap for each car.
2- Assign them percentage off the median. Transfer it to the decimal equivalent.
3- Multiply that by the car's value/price.
Repeat for each track, tallying the total. It's basically a deviation per credit. Lowest total amount wins.

Sample:
Car 1: 100.0secs = 1.00
Car 2: 101.0secs = 0.99
Car 3: 99.0 secs = 1.01

Car 1: 300 credits x 1.00 = 300
Car 2: 100 credits x 0.99 = 99
Car 3: 200 credits x 1.01 = 202

If a 100 credit car can be that close to a 300 credit car, it's the best value for credit. This isn't perfect, but over 10+ tracks, it will give you some result. You could use the plotting too, with this if you wish. It will help with a basic, easy computation. If you want a preset excel sheet, let me know.
 
Anyway the best value car is hands down camarro ss

It was after testing this car that got me thinking. A 35,000cr car that is so quick. If I had known about it I would have bought one early in my GT5 career.
 
Ive got 3 of the beauties 1 basically stock 500pp
1fully done up (not rm)
And a rm version fully done up and it's one of my favourite cars

The toyota 86gt is good for versitality
 
I am wondering the fairest way to work out the 'best value for money' car. I am testing cars with strict guidelines which keeps it fair. What would be the best way to calculate the result? Say I've tested 50 cars, my idea was to assign the cars points. 1-50 (1 for the cheapest car). Then assign 1-50 points per track (1 for the fastest car). Then the car with the lowest score after all tracks and cost are taken into account is the winner. I have 5 tracks I've tested on and realised that this probably wouldn't be a fair way to compare the cars as the value would only carry the weight of 1 circuit.
I could always multiply the cost score by 5 to give a range of 5-250 points but can anybody more mathematically gifted than myself come up with a better way to score the cars. Please assume my total of 50 cars just for ease of maths.

Now, for a Mathematician's answer:

You are attempting to determine 'value for money' for your 50 cars. Therefore in order to do this you must first determine what you mean by 'Value' and 'for Money'.

Clearly the 'for Money' would relate to how much the car costs. For simplicity it is probably best to use the 'new' value for each car as stock. This ensures you are comparing like-for-like. The advent of the OCD with 0 miles cars also means that this is a realistic measure to use. To make the test more equal, it might be sensible to purchase RS tyres for all cars (I don't believe any come with this tyre as stock, so this would cost the same for all cars) - the value of this purchase you may wish to not take into account as it presents a greater % of a cars value for each car.

The difficulty arises when you try to determine 'Value', as this is a subjective measure. Suggestions have included PP, lap-times and the ability to earn credits. As this is subjective, the characteristics should be determined by the OP (acjy1985). The OP should also determine what weighting should be applied to each characteristic. To ensure fairness, ideally this should all be determined from a more abstract position, and should ideally not play to any particular pre-conceptions.

Once all the characteristics and weightings to be taken into account are decided, a mathematical formula can be determined to calculate the 'value for money' which the OP has requested, based on the OP's assessment of value.

As a note, all cars should start equal, in that they all have the same driver, weather, upgrades, etc...

Finally, on the choice of track times, this is one way of determining value, but could be miss-leading. Imagine you picked Le Mans, Route X, SSR7, Indianapolis & Daytona for the comparison. This would clearly favour speed over cornering and a Veyron would probably come out better than the ZZII. However is you were to choose Eiger Short, Tsukuba, Kart Space & Monaco instead, the result would most likely be completely reversed in favour of cornering over speed. Also, lap-times can be taken into account, however if finishing position can also be taken into account instead without detracting from the timing gaps with the right equation and tests.
 
Right, firstly changing the tyre type is a big no no from my point of view. The stock cars have their stock tyres to help replicate(badly) the grip levels of that particular car. Adding RS nullifies the difference in grip. I only use the new value of premium cars and the only modification is the oil change(I do a couple of changes at set intervals for every car). The cars have the same driver in the form of me. The tracks are run on non-weather change layouts so weather is not a factor. I have picked 5 real world circuits which I believe cover most aspects to give both nimble and fast cars their chance to shine.
Also I would suggest your answer is more that of a politician rather than a mathematician as you have given the groundwork/policies but not the answer :sly: It has been helpful though.
 
That's fine, as the person defining 'Value' the parameters are your's to determine.

I assure you that I am a mathematician, I have a piece of paper which says so.

In no way did I refuse to determine the formula, but there's no use me trying to second guess how you determine 'value', or what you would like to take into account.

If you feel that the way you want to determine value is to run laps around a handful of circuits, and the best value is the one which performs best, my suggested answer is as follows:

(works for any number of tracks and/or cars)
1) Add the lap times together, from each track, for each car.
[e.g. Car 1 - 1.13.389 + 59.043 + 1.54.345 = 4.06.777]
2) Convert this into an decimal value for number of seconds
[e.g. 4.06.777 => 246.777]
3) Divide the 'new' cost of the car by this number

This number will then give you an accurate method of comparison between all the cars.

Of course, if there are more things you would like to take into account other than these lap-times, a more complex solution is required.
 
I like the idea of adding all the times together. I'll give it a bash and see how it turns out. According to a piece of paper I am a Geologist but I will trust your maths over my geology! Cheers for the suggestions.
 
Fastest lap don't say nothing. Some cars are more dificult to drive than others. Sometimes you will get a nice lap with the dificult car, with a gap of X seconds comparing to the "easy to drive" car. But if you take in acount the average time of 10 laps, it's posible that the easy car has a better average time. Maybe the dificult car has the best "hot lap"... But this means the car is better? Try the Vertigo Race Car for example. Is fast as a devil with a chilli in her ass, but is really dificult (at least for me) to have consistent times with it. When you take it to the limit, this cars don't forgives you nothing.
So, if you want to set a "value" for a car, you must take in acount this factor. So in my opinion the "best lap" means nothing. I think is better to take the time of 10 consecutive laps.
Another factor that can afect your "value" of a car is the setup. Default setup works nice in some cars. and is totally broken on others. There's some dificult cars that with a nice setup change a lot his "character" and became easy and sweet and you can get better times... This doesn't mean the easy to drive setup is the fastest setup...
Also you have to test the car in a few diferent circuits, some slow like autumn ring, some fast... The choice of representative tracks is dificult. And diferent tracks requires diferent setups... For example, you can't drive Nürburgring and Autumn with the same setup...
Also you have some clasic cars with huge prices wich perform similar to others with a lowest price... So what's the point?
After doing a huge work (if you don't do it well you will get strange useless results) you will say, this car has "151 points" at price/value ratio... So what? Who really needs this ratio? What it count is wich car wins the race. :)
A PP/average-time ratio, for example, would be interesting to me... But not price/average-time ratio.. Just my opinion. Don't want to discourage you...
 
Figured I'd weigh in with another random thought on this topic.

While the performance point system isn't perfect it does give us a general idea about a cars expected performance VS cars of a similar PP.

An easy way to get a rough general idea of cars values would be to simply determine cost per PP. Some cars will obviously have a much higher cost per PP and others will be real bargains.

Came up with this due to my recent involvement with the Budget Car Challenge. For the March season we were given Cr.30,000 to buy and tune a car with. Oil changes/engine rebuilds were free and all cars were to be equiped with comfort hard tires which also didn't come out of the budget. Everything else had to be paid for with whatever you had left after buying your car. After looking in the UCD for a bit I found my car which after an engine rebuild had around 470PP and left me a little over Cr.9,000 to spend on upgrades making it a real bargain.
 
Figured I'd weigh in with another random thought on this topic.

While the performance point system isn't perfect it does give us a general idea about a cars expected performance VS cars of a similar PP.

An easy way to get a rough general idea of cars values would be to simply determine cost per PP. Some cars will obviously have a much higher cost per PP and others will be real bargains.

This way you can easily get an aprox blueprint of your idea... You can make a data base with this and tell us if you found something interesting...
 
2J will earn 1 million plus in B-Spec Seasonal now @200% = Most valued car currently.
Anyway the best value car is hands down camarro ss
Its an expensive car but if the Seasonal rules dont change then its well worth it!!
My car for this months Budget Car Challenge series is a Mitsubishi Lancer Evolution VI RS '99.
I paid Cr.20,773 for it in the UCD with 25,000miles on it.
Added ECU
Racing Air Filter
Catalytic Converter Sports
Sports Exhaust
Height Adjustable Suspension
and a Wing to it.
Total cost was Cr.29,723
The car has anywhere from 502-513PP.

Remove the wing and add a set of Sports Soft tires and for less than Cr.37,000 you have a car that can easily win the Seasonal A-Spec 500PP challenge races.

At over 600,000 credits a race (assuming 200% bonus) you tell me what's a better value my car or the 2J?:crazy::D

Don't get me wrong the 2J is a wonderful car that can win races mine can't even hope to compete in but from a strictly Cost VS Rewards point of view the Lan Evo blows it away.

Not saying my Lan Evo is the best bargain either. There are probably other cars you can find in the UCD for less credits that with only minimul modification can also win the 500PP races.

In the end the higher the level of performance you wish to achieve the more you'll have to pay just like in real life.
 
Back