Theory on the Audi R8 LMS Ultra Handling

[KAWIGREEN23]

After much testing and trial and error, I was finally able to get this car to behave without having to resort to radical tuning. Basically, I left the front end alone and concentrated on the rear of the car. The end result is a bit of differential tuning, softening the rear end response, and more travel out back.

So happy as I am with the car now, I started wondering.... I saw the video on YouTube of Chris Harris testing the car.
Based on what he said, the car was very forgiving and manageable - a gentlemen's racecar. So how in the world did the car go from a gentleman in real life to a horribly oversteering, spinning mess in GT6?

One comment got me thinking... Audi is known for their quattro 4WD system, but it is not allowed in GT3 racing. Therefore, Audi put a traditional RWD setup in the LMS ultra. Relating that tidbit to my tuning to get the rear end under control with GT6's version of the LMS ultra lead me to this question...

What if PD modeled the LMS ultra as a 4WD car, but coded it as a RWD car? PD got the road car with the quattro system right and have had the model for years. Could there be a mistake/shortcut at work here?

Since 4WD naturally understeer due to the front wheels braking, turning, and accelerating at the same time, you need a stiff(er) spring in the back to get the rear to rotate more in a corner. There is also the matter of people claiming that the front end has too much grip and overpowers the rear... That could be a valid observation - in a 4WD, power to the front wheels can counter slipping in the rear and lessen total grip up front at the same time...

Hmm.... That may explain why the car oversteers so bad. To test my theory, I bought a R8 5.2 FSI quattro '09 and built it up with suspension, racing hards, bumped power up to match the LMS, did the chassis stiffening. Then I took the stock suspension settings from the LMS ultra and put the values on the 5.2 FSI then headed to the track. Now, I know that I won't get exact times due to the lower downforce, but I should get within 2 or 3 seconds if my theory is right....

On the track using a built up FSI with the LMS ultra suspension values, guess what??? I have a very manageable, gentlemanly near racecar!!! The car slides predictably, tracks well over bumps and curbs, oversteer can be controlled and saved, there is no hint of twitchiness on braking, no ridiculous snap lift-off oversteer (there is still some, it is an MR after all) AND I was within my 2 second test time comparing to the LMS ultra.

I've noticed many on this forum saying that dialing in a bunch of understeer, reducing grip at the front, or using harder tire compounds on the front than the rear got the LMS ultra to handle right for them. I tried it too and using the different compounds front to rear worked especially well. Using a harder compound up front would, in effect, reduce grip up front enough to reproduce the natural understeer a 4WD has when the fronts are doing most of the cornering work. That may explain why reducing grip up front made the LMS ultra handle better.

So, did PD make a mistake in the model and/or coding of the car? I will stand up and say that I believe the car is NOT broken, but rather, there seems to be a disconnect between the modeling and the coding...

 

[Janduin68]

Great post! Will try this tonight.

 

[PletdeKoe]

Hahahahah how do you guys come up with this? It is a great bit of imagination, but let's not forget: GT6 is a game.

Yes it has a lot of "real-life features" but it's all fake. You cannot apply real-life logic into a videogame.

 

[Hayden]

@PletdeKoe It's a game that goes to a great deal of trouble to recreate the real world. Using real world techniques makes absoloute sense. Hell, maybe i should try taking corners in game with an in-out-in technique. It's just a game, it won't make a difference to lap times or how the car behaves. 💡

If you're going to troll the least you could do is make it funny.

 

[TenEightyOne]

Hahahahah how do you guys come up with this? It is a great bit of imagination, but let's not forget: GT6 is a game.

Yes it has a lot of "real-life features" but it's all fake. You cannot apply real-life logic into a videogame.

You apply your concept of the presented reality to the mathematical control/display framework provided by the software.

That software is designed by people who considered the real concept and coded it into the representation that is GT6.

For those reasons you can apply real-life logic into a videogame when you're making decisions or judgements based on the counterpart concept. It was created that way.

OP, I'll try that, the LMS keeps trying to spit me out.

 

[phil_75]

Hahahahah how do you guys come up with this? It is a great bit of imagination, but let's not forget: GT6 is a game.

Yes it has a lot of "real-life features" but it's all fake. You cannot apply real-life logic into a videogame.

You'd be surprised how deluded some people on here are with regards to GT but the OP isn't one of them.
Everything he said makes perfect sense and gave evidence by putting his theory into practice.

Good post OP!

Btw, The Audi LMS Ultra and MR in general are more planted now imo, they still have the snap oversteer but it doesn't seem unnatural anymore.
Reset the suspension settings to default and try now.
The Delta wing is also more drivable but to make it better drop rear dampers and ARB to 1 and camber to 0.0 with full rear downforce.

 

[stb155]

What if PD modeled the LMS ultra as a 4WD car, but coded it as a RWD car? PD got the road car with the quattro system right and have had the model for years. Could there be a mistake/shortcut at work here?

With the Nissan GTR it is similar IRL, road car 4WD, GT3 car RWD and PD got that one right it seems (or at least no one is clomplaining about it).

Compare the R8 5.2 FSI quattro (stock) to the Lamborghini Gallardo (stock).
IRL they are quite similar (share lots of parts also) and they get similar results on test. (good front turn in, nice to drift around, easy to correct if you go in a corner to fast, easy/safe in the wet)
In the game the Gallardo fits that description, whil the R8 tries to kill you pretty often.
(that was before the patch, did not test them again after)

So i don't say your theory is not possible but i still think that it's just some feature/value of the R8 car model that causes the GT6 physic engine to act funny with those cars.

 

[teth75]

So what is a good setup for this car? I'm struggling on dream car championship, Cape Ring. The car just won't handle on the curbs, bumps and jumps!!
Help! this car would have been one of my favs but the handling is just awful !!

 

[stb155]

So what is a good setup for this car? I'm struggling on dream car championship, Cape Ring. The car just won't handle on the curbs, bumps and jumps!!
Help! this car would have been one of my favs but the handling is just awful !!

https://www.gtplanet.net/forum/conversations/audi-r8-lms-ultra-team-phoenix-simple-tune-fix.1538064/

Not sure if i remember correctly but are there not some stronger cars in those races ? (LMP, Group-C)
Of course its possible to win with the R8 but it takes a lot of "help" from the AI or using RS tires and tuning.
I would use a stock LMP on RH witout aids...

 

[teth75]

https://www.gtplanet.net/forum/conversations/audi-r8-lms-ultra-team-phoenix-simple-tune-fix.1538064/

Not sure if i remember correctly but are there not some stronger cars in those races ? (LMP, Group-C)
Of course its possible to win with the R8 but it takes a lot of "help" from the AI or using RS tires and tuning.
I would use a stock LMP on RH witout aids...

I know but i'm saved up to this point and it was a challenge in the other two races but possible. But this cape ring really has me until i can get the car working better. Doesn't help that the AI hit you all the time defending like madmen! lol
the link isn't there anymore but i've found a few tunes to try thank all the same

 

[Ridox2JZGTE]

I know but i'm saved up to this point and it was a challenge in the other two races but possible. But this cape ring really has me until i can get the car working better. Doesn't help that the AI hit you all the time defending like madmen! lol
the link isn't there anymore but i've found a few tunes to try thank all the same

The link is for a conversation, only those invited can be view it if I'm not mistaken - sent an invitation

 

[LVracerGT]

That video was posted quite recently. When PD modeled the R8 Racecar it had serious handling issues that Audi didn't work out until last race season, hence the poor handling in GT. The car needs to be updated to match the current R8.

 

[OdeFinn]

What if you just lower brakes to 1/1 or 2/2 and brake gently(under 40% braking), down shift later than "normally" to keep revs low enough, don't let revs stay up when on corner. During cornering be gentle on gas and short shift gears. Plus remember MR behavior of car

Car is very powerful and none of drivers in real life have such big balls to try drive it like most are trying to do in GT6.

 

[teth75]

The link is for a conversation, only those invited can be view it if I'm not mistaken - sent an invitation

OK got it now thanks. i'll try them out later

 

[Fuzzi0n]

That video was posted quite recently. When PD modeled the R8 Racecar it had serious handling issues that Audi didn't work out until last race season, hence the poor handling in GT. The car needs to be updated to match the current R8.

Going to call BS on this. The '12 version that we have that handles really stupidly ingame won the Nurburgring 24H and was only 1 lap of the distance record set the year before. It also won the Spa 24H race in 2012 and 2011, Zolder 24H in 2012 and 2011 and Bathurst 12H in 2012 and 2011. The 2012 wins were with the Phoenix racing car that we have. The car also won the Blancpain endurance series in 2012 and 2011 with a different team. If PD actually tried to make the car realistic, rather than making it up, it should handle properly, much like Chris Harris found with the version he drove. I'm doubting it had so much success while handling like it does in GT6 with huge lift off oversteer, wheelspin in most gears and huge oversteer when you touch the throttle. Considering the other GT3 cars don't have a problem, and road cars on road tyres seem to have more grip than the R8 I'm pretty sure something is wrong with it. If I compare to to another game, the McLaren in iRacing is about 300x easier to drive quickly, and people say the cars in that game have no grip. Chris Harris thought the Audi was easier to drive than the McLaren.

Not sure if they didn't get enough info on the cars or what happened, but the R8, R10 and R18 all have their own issues.

 

[OdeFinn]

Good car doesn't make you a good driver, fast car doesn't make you a fast driver.
Practice makes you handle those and more practice makes you faster driver.
Chris Harris have practiced a lot.

 

[Ridox2JZGTE]

Going to call BS on this. The '12 version that we have that handles really stupidly ingame won the Nurburgring 24H and was only 1 lap of the distance record set the year before. It also won the Spa 24H race in 2012 and 2011, Zolder 24H in 2012 and 2011 and Bathurst 12H in 2012 and 2011. The 2012 wins were with the Phoenix racing car that we have. The car also won the Blancpain endurance series in 2012 and 2011 with a different team. If PD actually tried to make the car realistic, rather than making it up, it should handle properly, much like Chris Harris found with the version he drove. I'm doubting it had so much success while handling like it does in GT6 with huge lift off oversteer, wheelspin in most gears and huge oversteer when you touch the throttle. Considering the other GT3 cars don't have a problem, and road cars on road tyres seem to have more grip than the R8 I'm pretty sure something is wrong with it. If I compare to to another game, the McLaren in iRacing is about 300x easier to drive quickly, and people say the cars in that game have no grip. Chris Harris thought the Audi was easier to drive than the McLaren.

Not sure if they didn't get enough info on the cars or what happened, but the R8, R10 and R18 all have their own issues.

A video of the car when tested during tuning session :



Simple lsd, suspension changes and brake balance, I didn't have much wheelspin, oversteer ? Only when I made it to, and never lift off the throttle in mid corner, when the rear break away, jam the throttle to regain traction ( shift the weight rearward ).

Good car doesn't make you a good driver, fast car doesn't make you a fast driver.
Practice makes you handle those and more practice makes you faster driver.
Chris Harris have practiced a lot.

Chris Harris's car might have adjusted power curve ( ECU ), suspension and brake tweaks to keep it safe in the rain, no racing team would allow someone like him drive the car on the wet surface using racing spec setup, unless they trusted him and have plenty of cash for insurance if mishaps happen.

 

[Vegard]

...no racing team would allow someone like him drive the car on the wet surface using racing spec setup...

The car was most likely on a wet setup.

And Chris Harris finished 13th overall in the 2010 Nurburgring 24H with a stock GT3 RS, so he's not your average journalist when it comes to driving fast.

 

[Ridox2JZGTE]

The car was most likely on a wet setup.

And Chris Harris finished 13th overall in the 2010 Nurburgring 24H with a stock GT3 RS, so he's not your average journalist when it comes to driving fast.

What I meant was less aggressive engine mapping etc to make it less edgy in the wet

Oh, I see, didn't know he is that good in driving fast.

 

[sparkytooth]

Chris Harris's car might have adjusted power curve ( ECU ), suspension and brake tweaks to keep it safe in the rain, no racing team would allow someone like him drive the car on the wet surface using racing spec setup, unless they trusted him and have plenty of cash for insurance if mishaps happen.

It is Chris Harris you know.

I'm pretty sure racing teams change their machine's setups for the varying track conditions.

 

[Ridox2JZGTE]

It is Chris Harris you know.

I'm pretty sure racing teams change their machine's setups for the varying track conditions.

My bad, I only know very little of journalist who can really drive race cars on the track My favorite are still Gan-san ( one of the best was NSX Type S Zero at Fuji ) and Nakaya-san ( his driving in racing spec Evos were jaw dropping ) from Best Motoring, they are truly legendary.

 

[LVracerGT]

Going to call BS on this. The '12 version that we have that handles really stupidly ingame won the Nurburgring 24H and was only 1 lap of the distance record set the year before. It also won the Spa 24H race in 2012 and 2011, Zolder 24H in 2012 and 2011 and Bathurst 12H in 2012 and 2011. The 2012 wins were with the Phoenix racing car that we have. The car also won the Blancpain endurance series in 2012 and 2011 with a different team. If PD actually tried to make the car realistic, rather than making it up, it should handle properly, much like Chris Harris found with the version he drove. I'm doubting it had so much success while handling like it does in GT6 with huge lift off oversteer, wheelspin in most gears and huge oversteer when you touch the throttle. Considering the other GT3 cars don't have a problem, and road cars on road tyres seem to have more grip than the R8 I'm pretty sure something is wrong with it. If I compare to to another game, the McLaren in iRacing is about 300x easier to drive quickly, and people say the cars in that game have no grip. Chris Harris thought the Audi was easier to drive than the McLaren.

Not sure if they didn't get enough info on the cars or what happened, but the R8, R10 and R18 all have their own issues.

Well considering I got the information from my gearhead boss from an Audi dealership, and he was given the information from meetings he had to go to with Audi by Audi engineers, I'm going to trust the information. Think about it, Audi designed that car chassis to work with AWD. Making it RWD plays havoc on the handling. Also take a look at the R8 LMS results by year.

Just because it's a 2012 in the game doesn't mean it's updated to 2012 specs. They could have easily just made a premium car model and used the physics from the original model PD did. And PD being PD I wouldn't put it past them.

 

[GranTurismo guy]

After much testing and trial and error, I was finally able to get this car to behave without having to resort to radical tuning. Basically, I left the front end alone and concentrated on the rear of the car. The end result is a bit of differential tuning, softening the rear end response, and more travel out back.

So happy as I am with the car now, I started wondering.... I saw the video on YouTube of Chris Harris testing the car.
Based on what he said, the car was very forgiving and manageable - a gentlemen's racecar. So how in the world did the car go from a gentleman in real life to a horribly oversteering, spinning mess in GT6?

One comment got me thinking... Audi is known for their quattro 4WD system, but it is not allowed in GT3 racing. Therefore, Audi put a traditional RWD setup in the LMS ultra. Relating that tidbit to my tuning to get the rear end under control with GT6's version of the LMS ultra lead me to this question...

What if PD modeled the LMS ultra as a 4WD car, but coded it as a RWD car? PD got the road car with the quattro system right and have had the model for years. Could there be a mistake/shortcut at work here?

Since 4WD naturally understeer due to the front wheels braking, turning, and accelerating at the same time, you need a stiff(er) spring in the back to get the rear to rotate more in a corner. There is also the matter of people claiming that the front end has too much grip and overpowers the rear... That could be a valid observation - in a 4WD, power to the front wheels can counter slipping in the rear and lessen total grip up front at the same time...

Hmm.... That may explain why the car oversteers so bad. To test my theory, I bought a R8 5.2 FSI quattro '09 and built it up with suspension, racing hards, bumped power up to match the LMS, did the chassis stiffening. Then I took the stock suspension settings from the LMS ultra and put the values on the 5.2 FSI then headed to the track. Now, I know that I won't get exact times due to the lower downforce, but I should get within 2 or 3 seconds if my theory is right....

On the track using a built up FSI with the LMS ultra suspension values, guess what??? I have a very manageablet
, gentlemanly near racecar!!! The car slides predictably, tracks well over bumps and curbs, oversteer can be controlled and saved, there is no hint of twitchiness on braking, no ridiculous snap lift-off oversteer (there is still some, it is an MR after all) AND I was within my 2 second test time comparing to the LMS ultra.

I've noticed many on this forum saying that dialing in a bunch of understeer, reducing grip at the front, or using harder tire compounds on the front than the rear got the LMS ultra to handle right for them. I tried it too and using the different compounds front to rear worked especially well. Using a harder compound up front would, in effect, reduce grip up front enough to reproduce the natural understeer a 4WD has when the fronts are doing most of the cornering work. That may explain why reducing grip up front made the LMS ultra handle better.

So, did PD make a mistake in the model and/or coding of the car? I will stand up and say that I believe the car is NOT broken, but rather, there seems to be a disconnect between the modeling and the coding...

I'm pretty sure that the update fixed mr cars. Every mr I have driven since 1.03 handles perfect

 

[Sixflags48]

Chris did say that he had the TCS set to 7/10, I could test this when I get home

 

[LVracerGT]

I'm pretty sure that the update nerfed mr cars. Every mr I have driven since 1.03 handles perfect

Fixed that for you. Mr cars got "fixed" for all the people with controllers or don't know how to properly drive MR cars.

 

[GranTurismo guy]

Fixed that for you. Mr cars got "fixed" for all the people with controllers or don't know how to properly drive MR cars.

Lucky I do know how to drive mr cars then. I do use a controller and the majority of players do anyway. And the mr cars do handle better, and better as in easier and more realistic. Difficult handling doesn't always mean better realism

 

[LVracerGT]

Lucky I do know how to drive mr cars then. I do use a controller and the majority of players do anyway. And the mr cars do handle better, and better as in easier and more realistic. Difficult handling doesn't always mean better realism

All the MR cars weren't hard to drive to begin with. Some were, sure, but they should have been. Everyone complained they were broken, now they're "fixed" and much, much more forgiving.

Edit: I haven't tested all MR cars yet so disregard that last part.

 

[GranTurismo guy]

Some cars were correctly a handful, like the lancia stratos. But the r8 race car, r8 road car (4wd mid engined I know) and the delta wing and others were just crazy and shouldn't have been. There is still braking oversteer for example but it's not mad anymore.

 

[LVracerGT]

You mean it's easily controllable now for the people that couldn't handle it before. There should be a lot of braking oversteer in MR and RR cars. That's where the majority of the weight is and it wants to swap ends. That's what MR 's do. If people don't know how to handle it properly, they shouldn't complain and instead learn the proper technique it takes to drive MR. Now MR 's seem to have been dumbed down.

 

[4everdelayed]

I'm pretty sure that the update fixed mr cars. Every mr I have driven since 1.03 handles perfect

The handling of the MR was literally the only thing I liked about GT6. I guess I'm gong back to GT5.

Oh btw I think this is highly relevant in this thread: http://en.wikipedia.org/wiki/Dunning–Kruger_effect

You are not Chris Harris and high powered are not supposed to be easy to drive.