A Few questions about real data in mod bikes

[TheCold0n3]

Hi guys,

I have been reading these forums for months and felt recently compelled to signup and discuss a few things. After X release
his fantastic bike which replicates real simulation physics with real geometry I was interest to see what had changed and how
he managed to put these changes into the game.

I found these changes in his files:

Tire:
p_Kx1 = 30.000000 / 25.000000
p_Cy2 = 0.600000 / 0.580000
p_Dy1 = 1.800000 / 1.821000
p_Ky1 = 20.000000 / 19.000000
SpringFactor = 500 / 520
Damper = 1520 / 1540

.cfg:
Steering:
KDamping1 = 0.2 / 0.3

From reading through these forums and the Italian forum there are a group of people who use real data and put it into the game.
I would like to know how X or anyone else converted real life tire or steering data into this simulator? or did X find/change these values himself?

Another note was that the Ducati GP10 bike shared the same clutch and shifting values of the original murasama bike. Can these not be changed? these are vital
strings that I would say defines a motorbike.

I did ask this on the Italian forum but my post was deleted (which I thought was quite strange). Maybe because in the bike release thread.

I did also notice that the 600cc bikes and the Aprilla 250 RS bike did not have such changes to the tire or .cfg steering.

I would like to learn modding and for me I would like to know where this "real data" is converted into simulation?

[HornetMaX]

[I have no idea why your post was deleted on the it forum, before anyone asks me]

Real tire data is extremely hard to find. Real race tire data is simply impossible to find.

Tire manufacturers sell this kind of data (road tires), and it's expensive: 1500-3000E for a single tire data set.

I have no idea how Piboso gathered the tire data he's using, but if you plan to experiment with the tire data, the first thing to do is to understand what each parameter does: this will allow you to avoid progressing blindly / at random. Of course, you should also have some technical knowledge of motorbike physics.

The bible of tires is the book "Tyre and Vehicle Dynamics" by H.B. Pacejka (2nd or 3rd edition recommended).
Chapter 11 details the model used for motorbikes (and used in GPB).

Preview available here (to see if it's something you can read and understand): http://books.google.fr/books?id=wHlkbBnu9FEC&printsec=frontcover&hl=fr&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false

MaX.

[TheCold0n3]

[I have no idea why your post was deleted on the it forum, before anyone asks me]

Real tire data is extremely hard to find. Real race tire data is simply impossible to find.

Tire manufacturers sell this kind of data (road tires), and it's expensive: 1500-3000E for a single tire data set.

I have no idea how Piboso gathered the tire data he's using, but if you plan to experiment with the tire data, the first thing to do is to understand what each parameter does: this will allow you to avoid progressing blindly / at random. Of course, you should also have some technical knowledge of motorbike physics.

The bible of tires is the book "Tyre and Vehicle Dynamics" by H.B. Pacejka (2nd or 3rd edition recommended).
Chapter 11 details the model used for motorbikes (and used in GPB).

Preview available here (to see if it's something you can read and understand): http://books.google.fr/books?id=wHlkbBnu9FEC&printsec=frontcover&hl=fr&source=gbs_ge_summary_r&cad=0#v=onepage&q&f=false

MaX.

Hi,

Being a racer I know quite a bit about tires. My question was not of what Piboso has done but what X changed from Piboso made already.

Because he changed something did he input real data or make changes his own based on his perceived version of "real" physics?

Some of the changes where so minute that I am surprised he could feel any change that why I ask does he convert real data.

Also, lets say I purchase data - straight from Motogp team for lets say $10000000 - does this give us real simulation conversion into the game?

[HornetMaX]

Hi,

Being a racer I know quite a bit about tires.

Most of the racers know quite a bit about how to use a tire and about how a tire behaves.
But 99.9% of the racers have zero idea about how a tire is simulated. You could be in the 0.01%, for all I know

My question was not of what Piboso has done but what _oDi_ changed from Piboso made already.

Because he changed something did he input real data or make changes his own based on his perceived version of "real" physics?

I can't answer for _oDi_, he'll has to answer by himself (if he wants to).

Some of the changes where so minute that I am surprised he could feel any change that why I ask does he convert real data.

That statement makes no sense. If a parameter is let's say 5.00, how much would be a tiny change and how much a big change ?
How big a change is cannot be inferred just looking at the numbers: you have to know what the parameter does on the end result.
Sometimes what looks like a tiny change makes a big difference. Unfortunately (or luckily, depending on the point of view), the world is not linear ...

Also, lets say I purchase data - straight from Motogp team for lets say $10000000 - does this give us real simulation conversion into the game?

Assuming the model you buy is a Pacejka 2002 model (or can be easily casted to it), then it can be slammed into GPB easily.
The overall tire model (Pacejka only gives the steady state behavior) needs some more data (for the dynamic behavior and other stuff like wear, temperature dependence etc).

In case you ever want to do this, you should get in touch with Piboso himsef, just to avoid wasting $10000000

MaX.

[TheCold0n3]

Most of the racers know quite a bit about how to use a tire and about how a tire behaves.
But 99.9% of the racers have zero idea about how a tire is simulated. You could be in the 0.01%, for all I know

You are correct, though am interested to see how X has done this.

I can't answer for X, he'll has to answer by himself (if he wants to).

I hope we do.

That statement makes no sense. If a parameter is let's say 5.00, how much would be a tiny change and how much a big change ?
How big a change is cannot be inferred just looking at the numbers: you have to know what the parameter does on the end result.
Sometimes what looks like a tiny change makes a big difference. Unfortunately (or luckily, depending on the point of view), the world is not linear ...

This is true, therefore we need an explanation.

Assuming the model you buy is a Pacejka 2002 model (or can be easily casted to it), then it can be slammed into GPB easily.
The overall tire model (Pacejka only gives the steady state behavior) needs some more data (for the dynamic behavior and other stuff like wear, temperature dependence etc).

In case you ever want to do this, you should get in touch with Piboso himsef, just to avoid wasting $10000000

MaX.

So your saying that providing that the game is given "real data" a computer game simulation will simulate real life handling period? I can understand
implementation but the result being a faultless replication of real life physics (which is yet to be done by anything - only thing close is a F1 simulation)

No answer for shifting or clutch values being the same as Murasama? the reason I ask is on one hand you have a "real bike" on another it shares the same strings
as that of a bike with totally different handling characteristics?

Please note, that as Piboso is the game developer, my questions are not to Piboso but to X or anyone else who uses "real data"

[girlracerTracey]

I have read through this thread with much interest.

I too would be interested in some further explanation on this theme from Odi. It is indeed an intriguing subject in which I hold a genuine interest. Against the background of recent experimentation performed by Rodney.

grT

[TheCold0n3]

Rodney... i recognize you among a thousand people...

shame on you

or maybe Camillo?

LOl..

Freddy, I have been watching your behavior on this forum for sometime now, I wished to say something but decided not to. I would like to carry
through with this ideology.

Can we please return to a civil conversation in relation to the merits or real data implementation into the game?

[HornetMaX]

So your saying that providing that the game is given "real data" a computer game simulation will simulate real life handling period? I can understand
implementation but the result being a faultless replication of real life physics (which is yet to be done by anything - only thing close is a F1 simulation)

Geez, some guys seem to have little understanding of what "simulator" means. I can feel Piboso's pain.

Yes, I'm saying that as the physics of GPB is pretty accurate (for our needs, at least), providing real data (as much as possible) is the way to go.

Now let me ask: you're saying what ? That since the physics is not accurate enough we should provide fake data ?
If yes, then you have never been working on simulating something, because otherwise you'd know the following:

• Garbage model + garbage data = garbage
• Garbage model + good data = garbage
• Good model + garbage data = garbage

If you do think the physics is not accurate enough, please make suggestions on how to improve it. I'd be seriously interested and Piboso too, likely.

But as discussing the physics is something way harder than spotting problems (or even just complain about them), in the last 5 years I've only read two suggestions related to physics made by people here (other than me). And both have been implemented by Piboso.

No answer for shifting or clutch values being the same as Murasama? the reason I ask is on one hand you have a "real bike" on another it shares the same strings as that of a bike with totally different handling characteristics?

I haven't yet had the time to have a close look at oDi's work on the gp10, but don't worry, if I find something wrong I will report, as always.

I actually have already reported him something (which was obviously a typo or a file swap) in the 1st version of the gp10 he delivered.
And guess what  ? He fixed the bug in no time, as opposed to ignore the critic and whine, as others have been doing in the recent past. Big difference.

In particular, I had a quick look at the changes he has made to the tires: they are nowhere as suspect compared to what I've seen on the rc8 (if that is what you're referring too).

I guess the original thread questions have been distracted from. Enjoy your typical forum banter young ones!

If your original question was this one:

I would like to know how X or anyone else converted real life tire or steering data into this simulator? or did X find/change these values himself?

Then I've already answered it: no real tire data available, so my best guess would be that oDi has modified the data provided by Piboso. But unlike others, he's done it in a physically sound manner, with insight.

MaX.

P.S.
Beer for me too, thanks !

[TheCold0n3]

So your saying that providing that the game is given "real data" a computer game simulation will simulate real life handling period? I can understand
implementation but the result being a faultless replication of real life physics (which is yet to be done by anything - only thing close is a F1 simulation)

Geez, some guys seem to have little understanding of what "simulator" means. I can feel Piboso's pain.

Yes, I'm saying that as the physics of GPB is pretty accurate (for our needs, at least), providing real data (as much as possible) is the way to go.

Now let me ask: you're saying what ? That since the physics is not accurate enough we should provide fake data ?
If yes, then you have never been working on simulating something, because otherwise you'd know the following:

• Garbage model + garbage data = garbage
• Garbage model + good data = garbage
• Good model + garbage data = garbage

If you do think the physics is not accurate enough, please make suggestions on how to improve it. I'd be seriously interested and Piboso too, likely.

But as discussing the physics is something way harder than spotting problems (or even just complain about them), in the last 5 years I've only read two suggestions related to physics made by people here (other than me). And both have been implemented by Piboso.

No answer for shifting or clutch values being the same as Murasama? the reason I ask is on one hand you have a "real bike" on another it shares the same strings as that of a bike with totally different handling characteristics?

I haven't yet had the time to have a close look at oDi's work on the gp10, but don't worry, if I find something wrong I will report, as always.

I actually have already reported him something (which was obviously a typo or a file swap) in the 1st version of the gp10 he delivered.
And guess what  ? He fixed the bug in no time, as opposed to ignore the critic and whine, as others have been doing in the recent past. Big difference.

In particular, I had a quick look at the changes he has made to the tires: they are nowhere as suspect compared to what I've seen on the rc8 (if that is what you're referring too).

MaX.

P.S.
Beer for me too, thanks !

My interest in implementation is based around the fact I have a competitive level racing bike and being a "tech" guy would be quite interested to work
with someone in order to put input into the game if it resulted in accurate handling referencing.

I am not comparing this with anything other than the standard motogp racing machine that comes standard with this game.

[PiBoSo]

So your saying that providing that the game is given "real data" a computer game simulation will simulate real life handling period? I can understand
implementation but the result being a faultless replication of real life physics (which is yet to be done by anything - only thing close is a F1 simulation)

No answer for shifting or clutch values being the same as Murasama? the reason I ask is on one hand you have a "real bike" on another it shares the same strings
as that of a bike with totally different handling characteristics?

Please note, that as Piboso is the game developer, my questions are not to Piboso but to _oDi_ or anyone else who uses "real data"

GP Bikes is not a "computer game simulation". It is a simulation, period, even if it has many helpers to make it usable with game controllers.
And believe it or not, it is a very accurate simulation, too: very close to F1 ones. So yes, using real data would make the output very close to real life physics.

Your idea that the shifting parameters "are vital strings that I would say defines a motorbike" is utter nonsense. A motorbike is defined mostly by engine, geometry ( including tyres radiuses ) and suspensions.

Last but not least, your use of "real data" with quotes and your calling into question a user that is rarely on this forum, about a mod that wasn't even released on this forum, and your complaining about another forum is extremely annoying.

Please change your attitude immediately or leave these premises. Thank you.

[TheCold0n3]

So your saying that providing that the game is given "real data" a computer game simulation will simulate real life handling period? I can understand
implementation but the result being a faultless replication of real life physics (which is yet to be done by anything - only thing close is a F1 simulation)

No answer for shifting or clutch values being the same as Murasama? the reason I ask is on one hand you have a "real bike" on another it shares the same strings
as that of a bike with totally different handling characteristics?

Please note, that as Piboso is the game developer, my questions are not to Piboso but to _oDi_ or anyone else who uses "real data"

GP Bikes is not a "computer game simulation". It is a simulation, period, even if it has many helpers to make it usable with game controllers.
And believe it or not, it is a very accurate simulation, too: very close to F1 ones. So yes, using real data would make the output very close to real life physics.

Your idea that the shifting parameters "are vital strings that I would say defines a motorbike" is utter nonsense. A motorbike is defined mostly by engine, geometry ( including tyres radiuses ) and suspensions.

Last but not least, your use of "real data" with quotes and your calling into question a user that is rarely on this forum, about a mod that wasn't even released on this forum, and your complaining about another forum is extremely annoying.

Please change your attitude immediately or leave these premises. Thank you.

I did not involve myself with forum conversation here due to the nature in which I have observed. I wish for no disrupt here only mature conversation in-regards to how real-data is implemented and its accuracy once done. However I feel this is a sensitive topic with both the game developer and mod creators.

[HornetMaX]

My interest in implementation is based around the fact I have a competitive level racing bike and being a "tech" guy would be quite interested to work
with someone in order to put input into the game if it resulted in accurate handling referencing.

"tech" guy = ? Today it may mean plenty of different things.

If you own a bike (which ?), then you could provide some tech info like dyno charts, weights and some measures etc.
Depending on the bike, some info is hard to come by for modders.

But even owning such a bike, it's very unlikely you'll be able to get the tire parameters (for example).

MaX.

[janaucarre]

Hello,

- After many off topics posts I decided to clean them. I deleted all the posts that seemed off-topic for me.

- I think many here have read TheCold0n3 posts in an agressive mod due to the tensions happened weeks before.
  Add fuel to the fire is not a good thing from everyone.

- Personnally i didn't read something hard in the first post, the only two things should not be here are to ask something about a mod not here publied, and mainly to speak about a problem wich happened on another forum. We have not read what happened there.

- If the thread continues with off-topics posts,
by anyone-that, I will close it.

Thank you, and have a good night.