Impossible with LFS as it depends on your set/tyre temps/fuel etc. Why would you want to know that anyway? The WR's can all be beaten, thats all you need to know

He does have a point. He's asking whats the theoretical best possible laptime if the setup was spot on and everything was done correctly.

There are soo many variables for racing, I don't think there could ever be a best possible lap, other than the WR's currently available =)

it wouldn't be too hard if the track was smooth as glass and consisted of simple corners separated by long straights, but as soon as you introduce complex corners, where how you exit one corner affects your entry to the next, or where there are multiple possible lines, the calculations would start to be very intensive. if there are any bumps it would make the calculations even almost impossible.

the reason racing is so fun is because it takes a certain amount of instinct, experimentation and feel to find a fast line, and computers aren't very good at such things.

If anything it isn't impossible and I would think for the developers it would be easy to simulate and come up with the ultimate fastest lap -
Just think about it - All what the SIM is a bunch of variables well many of them - setup, tires, temp, fuel etc... and also your inputs accelerate, break, turn so based on the variables set for setup, tires, temp, fuel etc... and how fast you going when you break and how much you break how much you turn will determine if and how much you slip. Therefore if you know the boundaries of there variables which the developers do then coming up with a model that calculates the ultimate lap time is simple math’s

The complexity of the track is already known in the game and based of all your variables setup, fuel, temp tires etc.. and current speed, position on track, track conditions (bumps) the program knows exactly how to respond when you break, turn even to the point of putting down tyre marks - and it does this for every bit of the track - stating that the calculations would be very intensive implies that playing the game there would be very intensive calculations and almost impossible to play - yet we all play quite easily with not so high end machines and it runs fine.

And why would I want to know - Perfection - how close can we get to perfection - isn't that what we strive for in most things in life and it nice to know how far off the mark we are.

Kinda takes the human out of the whole thing! Don't really see the point

Actually there should be a best possible lap on lfsworld according to the best splits uploaded, wr lap doesn't always have the best splits. I hope you get what i mean.

If it was easy then we would already have perfect AI. But we don't.

I don't think any human would ever achieve it - so I don't think it takes the human out of it. Just the game it self takes the human out of real racing because it's just that - In real life when I lock up or slide around a corner I get a lot more feedback than just a force feedback wheel vibrating in my hand and noise in my ear's and a 2d screen displaying a 3d image of me slipping. When that happens in a sim I might be picking my nose, sipping coffee or scratching my balls what ever, - in real life I'll be swearing if it was a mistake, soil my pants if I hit the pavement or even if it seems like I'm going to hit the pavement or another car.

Quote from Count Duckula :

... stating that the calculations would be very intensive implies that playing the game there would be very intensive calculations and almost impossible to play - yet we all play quite easily with not so high end machines and it runs fine.

sorry man, but you don't know what you are talking about. when we play the game, our computers calculate our position on the track based on driver inputs we give, and the physics of the game. it's one set of calculations made my the computer, and some very different processing going on in our brains to generate those inputs.

but to find the fastest possible lap, without a human brain involved, you would have to calculate every possible combination of driver inputs, and simulate all of the resulting car motions, and even for a single setup this would be an almost infinite number of calculations. (almost infinite, and not actually infinite because computers are digital, not analog). taking into account all possible setups, the number of calculations required would be truly huge.

Quote from Taavi(EST) :

Actually there should be a best possible lap on lfsworld according to the best splits uploaded, wr lap doesn't always have the best splits. I hope you get what i mean.

Yea I do much like what the do on F1 now days show the fastest sector time and what a possible lap time would be should a drive hook up all 3 sectors with the fastest achieved time - that would be also very interesting because it will show what is humanly possible - But I also like to know what is the absolute fastest time - its a simulation game - so I still can't see why it would be so difficult to simulate a fastest time. As someone else said the AI not perfect - so what then all you going to do is determine the fastest time for this programs not so perfect AI

Quote from evilgeek :

sorry man, but you don't know what you are talking about.

He does know what HES talking about, just not what YOUR talking about

Quote from Count Duckula :

I don't think any human would ever achieve it - so I don't think it takes the human out of it. Just the game it self takes the human out of real racing because it's just that - In real life when I lock up or slide around a corner I get a lot more feedback than just a force feedback wheel vibrating in my hand and noise in my ear's and a 2d screen displaying a 3d image of me slipping. When that happens in a sim I might be picking my nose, sipping coffee or scratching my balls what ever, - in real life I'll be swearing if it was a mistake, soil my pants if I hit the pavement or even if it seems like I'm going to hit the pavement or another car.

Errr, yeah. We all know it's a game you don't really need to play the "oh it's just a game anyway" card.

I'd still rather try to beat another person and not the computer, that's kinda the whole point of the LFS!

Quote from Taavi(EST) :

Actually there should be a best possible lap on lfsworld according to the best splits uploaded, wr lap doesn't always have the best splits. I hope you get what i mean.

BUT you have to consider "how" those splits were achieved.

Example:

Maybe the guy got a better split (than the WR one) on a really twisty sector of the track cause he's using a high downforce setup but the track has a lot of really long straights and a low downforce is a better/faster setup overall. And the same could be done with a low downforce setup on a track which starts with long straights but has it's most important sectors full of twisty turns.

Or maybe someone got into a turn (which has a split "marker" right before or after it) really hot and then, because of it, got a better split than the WR but since he went wide he won't get OUT (which is the most important, as you all know) as fast as the WR and so on.

Would be a really neat info, that of combined "best splits", but you should be really careful about that "best humanly possible" stuff.

Hell, maybe even better splits than those can be all achieved by the same guy on a single lap. But splits "mean nothing" if you know what i mean, given my examples. Would be neat tho.

Guys just step back and think about this.
Take a track, would it be easy for the computer to determine the shortest route around the track which will give you a distance? I think we all agree yes it would! So now we have the shortest distance around the track lets determine for a specific car, what would be the fastest that car could cover that distance if it was a straight line. That means you need the best possible setup for acceleration and top speed – Would you say the computer could handle that? Yes I do.
So what do we have - distance and fastest time to cover that distance if it were a straight line. WOW we have a starting point!
The next step would be to take every corner and find out the best line to get through there the quickest and also the best setup to get through there the fastest.
Would you agree that just for one corner it shouldn’t be too difficult for a computer to determine the best line and setup?
Now we do that for every corner. And what we are left with is maybe a different setup for every corner – a setup if the distance was entirely straight and we also have the best possible line to take for each corner and when you stitch the best possible line for every corner together you have the best route around the track.
So all what really left to do is simulate each scenario you have for the entire track and with the combinations of setups you have – and it may lie between two setup’s but as it narrows down which setup is best for the entire track it can then try combinations that are in between two setups and see if it’s better or worse. The net result is after a few seconds processing this data on a 386 you will come up with a lap time that is theoretically the fastest around that track. – Not so difficult now was it. - just need a programmer to code it now

Quote from garph :

Errr, yeah. We all know it's a game you don't really need to play the "oh it's just a game anyway" card.

I'd still rather try to beat another person and not the computer, that's kinda the whole point of the LFS!

A theoretically fastest speed would never be achieved by a human - its theoretical and it should never be beat because it is theoretically the fastest.

But you know what - I guess why they don't do it - just by the responses it would seem most people would loose or have no interest in playing the game should they have such information. And I'm sure the people who develop these games know this and that is why they don't make it known.

Quote from BlueFlame :

He does know what HES talking about, just not what YOUR talking about

true, true. and he's still on about it.


mr. duckicluck. please go back to my original post and compare to your latest posts. you will see that the scenario you are describing is the one that i said wouldn't be that hard to calculate. ie, simple turns, separated by long straights, where each turn has exactly one good line, and the straights are long enough to allow you to get into position for the next corner no matter how you exit the current corner.

but there aren't any tracks like that. you can't calculate the best line through each corner without putting them in the context of the corners before and after. the perfect line through one corner might spoil your line going into the next, so the fastest line through the section will almost certainly not take the perfect line through either corner.

add to this the accumulated affect of earlier corners on tire temps and you have yet another almost impossible calculation on your hands. how you take the corners in sector 1 will affect your tire temps, and therefore affect how you take corners in sector 2 and 3. for example, you might be able to save time on a corner in sector 1 by late braking, only to lose that time in a sweeping corner in sector 2 because your tires are too hot.

what you need to understand is that when it comes to racing, the sum is greater than the collection of it's parts. this is true on multiple levels. there is more to a fast lap than fast corners, and there is more to a fast race than fast laps. why do you think quali times are always faster than the fastest lap of a race?

for the nerds in the house, here is a real big number.

assuming a controller with 8 bit resolution (which is probably a low estimate):

256 possible steering wheel positions
x 256 possible throttle positions
x 256 possible brake positions
x 3 possible shifter positions (up/down/no change)

= 50,331,648 possible driver inputs per sample


100 hz lfs clock rate
x 100 second estimated lap time

= 10,000 time slices per lap


so to test all possible lines for ONE set, the LFS physics engine would have to calculate the car position 50,331,648^10,000 times, a number that is too big for my computer to even calculate.

even running such a simulation for just 0.1 seconds would hit the physics engine 1.04E+77 times, and a longer simulation would increase the number of calculations exponentially.

if someone where to attempt it, they could save a lot of computational steps by aborting laps that are ruined by going off track, but even chopping the work down to 1/1000th or 1/1,000,000th would still leave a rediculously vast number of laps to calculate. and remember, this is just for one setup. there are millions of possible setups. to test them all would simply be impossible, even using the worlds biggest super computers.

i think that if, like evilgeek suggested, you get rid of all inputs that just wouldn't work (i.e.:control inputs that steer/gas/brake/shift too early/late/much/little and setup variables {outlying values such as final drive too low or high, camber too much, etc..} that just dont make sense) then this could be done if someone were to write a program to do so and have the time to let it run.

almost too many variables to list though.

car, tire temps, gas, different lines, setup, wind. you get the idea with the two largest and most variable components being set and line.

the formula 15gr = (mph)^2 can be used to determine the highest speed at which a corner can be taken. The g is the traction limit for a particular tire on a particular road surface (which can usually be very easy to find given REAL LIFE tires and such). R is the radius of the circle that your car takes the path of through the turn.

What you can do is, find the radius of the BIGGEST circle you can make in a turn, figure out the grip of your car (might be difficult since LFS uses imaginary cars, not manufactured, factory produced ones). Plug those 2 numbers in and you have the FASTEST speed at which you can take the corner without losing traction. Do this for every corner and you should have a rough estimate of how long it would take to lap a circuit.

Calculating straights is self explanatory.


you would never be able to figure out THE fastest lap time but you can absolutely come up with an estimate, plus or minus maybe 1 or 2 seconds of the fast lap.

F1 teams have already been doing this for years. They use some form of simulation software (not driving for this purpose) to calculate braking points, optimum corner entry speed etc. Last season the UK F1 coverage had a short item showing Button's team reviewing the results.

IIRC the computer calculated best possible lap at Monaco for the Mclaren in 91 or 92 was actually beaten by Senna in his qualifying lap.

Quote from dontsimon :

F1 teams have already been doing this for years. They use some form of simulation software (not driving for this purpose) to calculate braking points, optimum corner entry speed etc. Last season the UK F1 coverage had a short item showing Button's team reviewing the results.

IIRC the computer calculated best possible lap at Monaco for the Mclaren in 91 or 92 was actually beaten by Senna in his qualifying lap.

Yea that would be expected - in real life humans always manage to exceed what is theoretically possible - BUT if you were to apply the same calculations to a SIM you would arrive at what I originally asked a theoretically fastest lap for the game which I doubt a human could beat because it a SIM and not real.

evilgeek you have just demonstrated that the only nerd is you throwing up such unrealistic possibilities - Now naughtygeekboy if they can use computers to land planes and put space ships in space calculating the fastest lap in a sim is like farting in thunder.

Another thing bad geekie boy just read ans7812 post and you hopefully should realise that it isn't a difficult thing to do and as ans7812 says it may not be the absolute fastest but it will the clostest we could possibly calculate. Which is then in theory the fastest because we can't calculate anything faster. And even though the SIM may not be a representation of real life it also has values for "15gr = (mph)^2"

Nice going there, starting to call people names...

It can just be a Approximation, a accurate fastest lap is impossible to calculate. LFS is not a game which know results, LFS just calculate physics. No lap will be the same like another lap, it need eternally calculations.