Why is the Veyron undriveable?

  • Thread starter Topsu
  • 152 comments
  • 16,240 views
Deffo an issue.

The car cannot be driven in a straight line after certain speeds.

People arguing otherwise need to calm down with the science. It's a bug.

The mission in question was an easy, first time, pass when I first started playing in April/May.

Doing the Hypercar DLC races is when I started to encounter this issue.
 
Last edited:
I dont know how else to show you that the Veyron is stable. I've already shown you in one video that the car can hit its top speed without getting out of control, then in a second video doing the mission that people are complaining about showing the car being stable again. Is it always stable? Not in the mission. Like I said in the second video, the time I hit 250+ the car got unstable. So what if PD nerfed the car settings in that mission? The mission is the mission and you have to use what is given to you.

Have you ever driven a Veyron in real life? My guess is that you haven't. I haven't driven one in real life so I can not say how the car feels at its top speed. Is it 100% stable and would I ever take my hands off the wheel at top speed in real life? Probably not, but I can do that in the video game. So one could say that the driving physics of the car are better in the game than they are in real life.

The fact of the matter is, it's a video game. It's not real life. PD has no way of perfecting the real driving physics of the cars in the game. They can give us close approximations at best, but it's not real.
I'm lucky enough to have driven this car (not at high speed sadly), among other supercars, and I have to say that it's stable to drive it. Imagine Bugatti made a car that it's totally unstable when you drive it, the world will mock Bugatti for making a piece of crap.

Said that the car is not stable in the game. Once you reach 220 mph the car moves all over the track and it's impossible to control unless you release the gas and go back to max 210 mph. In your video, you show the car taking the turn at max 200 mph, which proves what people are saying in this post.
 
So this problem is still happening? I havent played this game in a a few weeks, I was grinding for the Chiron but then GTA caught my attention again.

If the Chiron also has this problem then im not going to bother with it.
 
I didn't. I just said that the same general physics that applies at 100 mph applies at 250 in the case of cars.

Yes, but it's still producing downforce. Like I said, I don't have the exact figures and I don't really feel like there is a need to look them up unless the Veyron produces lift, which isn't likely. The car at speed has more grip, I showed that. This is much better evidence than "I think X", which isn't evidence at all.

Also downforce is quadratic, not exponential.
Do you think this game actually models natural downforce at high speeds? Like if a car doesn't have any aero parts, does it have the downforce it SHOULD have at, say, 150mph? I know the ACR Viper was claimed to make its weight in downforce at like 170mph and I'm not sure that this engine is actually modeling "natural" downforce endemic to the car's shape, etc. Ex: Car x has 200lb (or ft lb?) of downforce at 100mph, and has 50lb of lift at the same speed, so the car has 150lb of downforce at 100mph after the effects of lift have been countered (assuming it's 1:1... I really don't know)

We can definitely see the effects of downforce at high speed with Gr.3/2/1 cars, but I'm unable to tell if the physics is modeling downforce and lift on cars that don't have wings, splitters, diffusers, etc. From what I can gather the number value given to the front/back is functional even at low speeds, like under 30mph through a chicane, and I don't see how that would be physically possible without some type of mechanical suction like with the 2J, the first F1 car Gordon Murray worked on before he went to McLaren (name of the car is escaping me...)

Genuine question by the way as you seem to know a bit about how downforce actually works. I took the required physics for my Chem degree but wasn't very good at it, and it was just the basic 100/200 level physics for STEM majors.
 
Some guy posts videos proving how stable the Veyron is and people still argue otherwise. People in internet try harder and harder everyday to try to have the last word in every discussion
 
This is turning in to the dumbest internet argument in history. Nothing but conjecture and non-evidence based arguments.

"You can make predictions based on physics". Okay, fine. Show your math. Prove your predictions.

"250mph isn't terribly different than 100mph"... I think Bugatti would beg to differ. If it were so easy, everyone could do it.

Honestly, this is turning in to trolling and I'm done dealing with anyone who can't put up fact based evidence to prove their point. The fact that this discussion has over 100 replies is ridiculous. It's 1 stupid car out of 450 in the game that has been proven to 1) be stable at top speed and 2) complete the mission on the first try with a gold medal result AND the car was stable the whole way.
It seems however that it's perfectly acceptable for you to claim it must be difficult to drive because over 250mph without any evidence based arguments. And no, nothing you have posted counters that in any evidential way.

These claims work both ways you know and some simple maths has since been provided, unless you believe the Veyron produces lift now.

I remind you, the discussion at present is that the game isn't replicating real world driving phyiscs for that car, likely a bug in the cars stock suspension physics. Something evidenced by @BobDx123 just 3 posts before yours which you've conveniently ignored.

In the meantime I do intend to try to record a video at some point showing the cars behaviour in question, however I don't turn my PS4 on most days, sometimes weeks apart, so it'll happen when it happens. And the "it's one stupid car out of 450" argument doesn't fly, this is an ongoing discussion about that one car, if you don't like the discussion why are you taking part in it? Don't tell everyone else to stop, just leave if you want to.
Some guy posts videos proving how stable the Veyron is and people still argue otherwise. People in internet try harder and harder everyday to try to have the last word in every discussion
Because the videos, whilst informative to a point, don't disprove the argument at all. It's ironic that here you are, attempting to have the last word whilst offering nothing to the discussion yet accusing others of wanting the last world.

Plenty of people have explained the behaviour being discussed and it isn't what is covered by those videos beyond the claim in the OP about the mission being impossible which it clearly isn't and was probably hyperbole, yet keeps being referred back to despite the fact the discussion has long moved on from that claim.

To quote Wolfgang Schreiber who knows a damn sight more about the Veyrons stability at top speed “The Veyron covers a wider speed spectrum than any other sports car. Therefore, it must ride reasonably well at 40km/h, and at the same time it must be perfectly stable and confidence-inspiring at 400km/h. To meet these conflicting requirements, there are three suspension settings to choose from: standard, handling and top speed. Instead of installing an adjustable suspension package, we opted for a quick-acting central hydraulic system which permits, among other things, instantaneous ride height adjustments.”

Of course when you look at the downforce figures and coefficient drag of the Veyron you can see it clearly generates downforce at speed not lift. Therefore the more speed = the more downforce, basic maths. The more downforce, the harder the car is pushed towards the ground. The harder it's pushed towards the ground the more it should stick to the ground. Also basic maths. Thorw in engineering designed to facilitate the car traveling at that speed in a stable fashion including

But in GT7 the car gets very slippy at speed, it behaves unrealistically, but let's just go "nah, it handles right in GT7 regardless of what real cars behave like or engineers say".
 
Do you think this game actually models natural downforce at high speeds? Like if a car doesn't have any aero parts, does it have the downforce it SHOULD have at, say, 150mph? I know the ACR Viper was claimed to make its weight in downforce at like 170mph and I'm not sure that this engine is actually modeling "natural" downforce endemic to the car's shape, etc. Ex: Car x has 200lb (or ft lb?) of downforce at 100mph, and has 50lb of lift at the same speed, so the car has 150lb of downforce at 100mph after the effects of lift have been countered (assuming it's 1:1... I really don't know)
Computing real time aero forces from shape is possible, but I don't think Gran Turismo or any racing game does it. The simulator X-Plane uses Blade Element Theory to compute real time aerodynamics, but this is a flight simulator and planes are actually easier to do the math on than cars because they are more streamlined. BET is also not as accurate as more costly methods like CFD which can't be performed in real time, even when it comes to the types of shapes BET was designed to model well like wings and airfoils.

Typically a car sim like Gran Turismo gets aero data from an outside source (manufacturer, wind tunnel data, CFD results) and just hard codes the data as part of a car's stats. I imagine that in the case of GT they take a single CD value for drag and then have front and rear values for simulating downforce and that these values correspond to the car driving straight and flat, with best guess modifiers for when the car pitches and yaws. This is simple but effective and it basically totals the effects of all the parts of the car like you did with the addition of lift and drag without having to individually model all the positive lift and negative lift surfaces of the car. As long as the numbers used are correct or close to correct, the result should be convincing as long as the car is doing something unusual like sitting at an odd angle with respect to the ground or the direction of travel. Unfortunately, I know GT has been very wrong in the past. In GT5 the Viper ACR was horrendously undermodeled and was closer to the regular Viper SRT-10 than anything else for example. I have not played GT since GT5, so I'm less certain about how the newer games work.
We can definitely see the effects of downforce at high speed with Gr.3/2/1 cars, but I'm unable to tell if the physics is modeling downforce and lift on cars that don't have wings, splitters, diffusers, etc. From what I can gather the number value given to the front/back is functional even at low speeds, like under 30mph through a chicane, and I don't see how that would be physically possible without some type of mechanical suction like with the 2J, the first F1 car Gordon Murray worked on before he went to McLaren (name of the car is escaping me...)
Aerodynamics are functional at all speeds other than zero, it's just the effect can be small at low speeds. Racecars might have large enough wings, etc that you can feel the effect even at low speed, but it's going to be a lot less than at 100 or 150 mph. You can work out what the ratio is by squaring the ratio of speeds ie (30 mph/150 mph)^2 tells you that at 30 mph a car will produce 4% of the downforce that it does at 150 mph.

Back to GT, whether a car has wings or not shouldn't matter, the same method I outlined above, plugging in data from external sources, will work as long as the car is driving under expected conditions.
Genuine question by the way as you seem to know a bit about how downforce actually works. I took the required physics for my Chem degree but wasn't very good at it, and it was just the basic 100/200 level physics for STEM majors.
I've always had an interest in aero and aviation so I pursued a degree in the field and ended up being an engineer. I don't mind engaging in discussions like this at all.
 
So this problem is still happening? I havent played this game in a a few weeks, I was grinding for the Chiron but then GTA caught my attention again.

If the Chiron also has this problem then im not going to bother with it.
There's nothing wrong with either Bugatti if you change the set-up.
I run the Chiron at 470+kph(without drafting) at Route X and its very stable and Daytona tri-oval i did a 32.4 second lap.
The Veyron is just as stable but a little slower, 440+ at X, 33.9 at Daytona.
 
Computing real time aero forces from shape is possible, but I don't think Gran Turismo or any racing game does it. The simulator X-Plane uses Blade Element Theory to compute real time aerodynamics, but this is a flight simulator and planes are actually easier to do the math on than cars because they are more streamlined. BET is also not as accurate as more costly methods like CFD which can't be performed in real time, even when it comes to the types of shapes BET was designed to model well like wings and airfoils.

Typically a car sim like Gran Turismo gets aero data from an outside source (manufacturer, wind tunnel data, CFD results) and just hard codes the data as part of a car's stats. I imagine that in the case of GT they take a single CD value for drag and then have front and rear values for simulating downforce and that these values correspond to the car driving straight and flat, with best guess modifiers for when the car pitches and yaws. This is simple but effective and it basically totals the effects of all the parts of the car like you did with the addition of lift and drag without having to individually model all the positive lift and negative lift surfaces of the car. As long as the numbers used are correct or close to correct, the result should be convincing as long as the car is doing something unusual like sitting at an odd angle with respect to the ground or the direction of travel. Unfortunately, I know GT has been very wrong in the past. In GT5 the Viper ACR was horrendously undermodeled and was closer to the regular Viper SRT-10 than anything else for example. I have not played GT since GT5, so I'm less certain about how the newer games work.

Aerodynamics are functional at all speeds other than zero, it's just the effect can be small at low speeds. Racecars might have large enough wings, etc that you can feel the effect even at low speed, but it's going to be a lot less than at 100 or 150 mph. You can work out what the ratio is by squaring the ratio of speeds ie (30 mph/150 mph)^2 tells you that at 30 mph a car will produce 4% of the downforce that it does at 150 mph.

Back to GT, whether a car has wings or not shouldn't matter, the same method I outlined above, plugging in data from external sources, will work as long as the car is driving under expected conditions.

I've always had an interest in aero and aviation so I pursued a degree in the field and ended up being an engineer. I don't mind engaging in discussions like this at all.
I completely agree, the figures in the game will be fudge jobs and that's very, very reasonable. It's a case of how well they fudge it. But because games fudge a number of these complex calcuations you can often end up with strange anomolies here and there, things like the old wheelie gitches where you went thousands of MPH on 2 wheels for example. Some cars becoming more slippery and wallowy at speed when that shouldn't be the case. Sure, a break in traction at 250mph is quite a serious issue, but you wouldn't ever see it expressed with wallowing and tyre slippage as we get with the Veyron in GT7.

The whole issue wth the Veyron in GT7 is that it's an outlier, there are other cars that exhibit similar characteristics, but in GT7 they aren't the norm. The issue also goes away when the car is fitted with adjustable suspension yet given settings to mimic stock, which is why I beleive it's an issue connected with the stock suspension physics. Sure other factors will contribute, but the issue goes away with adjustable suspension, so there's the comparable data point.

At speed you'll see the weight of the car increase, the pressure on the tyres will increase and suspension will compress. The whole car will become stiffer not looser. Unless it's a car with poor aero and it produces lift, then, well you might have problems above certain speeds (depending on the car).
 
Last edited:
Computing real time aero forces from shape is possible, but I don't think Gran Turismo or any racing game does it. The simulator X-Plane uses Blade Element Theory to compute real time aerodynamics, but this is a flight simulator and planes are actually easier to do the math on than cars because they are more streamlined. BET is also not as accurate as more costly methods like CFD which can't be performed in real time, even when it comes to the types of shapes BET was designed to model well like wings and airfoils.

Typically a car sim like Gran Turismo gets aero data from an outside source (manufacturer, wind tunnel data, CFD results) and just hard codes the data as part of a car's stats. I imagine that in the case of GT they take a single CD value for drag and then have front and rear values for simulating downforce and that these values correspond to the car driving straight and flat, with best guess modifiers for when the car pitches and yaws. This is simple but effective and it basically totals the effects of all the parts of the car like you did with the addition of lift and drag without having to individually model all the positive lift and negative lift surfaces of the car. As long as the numbers used are correct or close to correct, the result should be convincing as long as the car is doing something unusual like sitting at an odd angle with respect to the ground or the direction of travel. Unfortunately, I know GT has been very wrong in the past. In GT5 the Viper ACR was horrendously undermodeled and was closer to the regular Viper SRT-10 than anything else for example. I have not played GT since GT5, so I'm less certain about how the newer games work.

Aerodynamics are functional at all speeds other than zero, it's just the effect can be small at low speeds. Racecars might have large enough wings, etc that you can feel the effect even at low speed, but it's going to be a lot less than at 100 or 150 mph. You can work out what the ratio is by squaring the ratio of speeds ie (30 mph/150 mph)^2 tells you that at 30 mph a car will produce 4% of the downforce that it does at 150 mph.

Back to GT, whether a car has wings or not shouldn't matter, the same method I outlined above, plugging in data from external sources, will work as long as the car is driving under expected conditions.

I've always had an interest in aero and aviation so I pursued a degree in the field and ended up being an engineer. I don't mind engaging in discussions like this at all.
Damn, that was a fantastic explanation. Thank you.
 
Some guy posts videos proving how stable the Veyron is and people still argue otherwise. People in internet try harder and harder everyday to try to have the last word in every discussion
Since when does a video equal telling the truth?

I've tried this car at Le Mans on controller and it is not stable. Add sports suspension and the problem goes away.
 
Clarkson is another self-proclaimed "expert" TV personality.
It has nothing to do with stock or customized suspensions. I had the Genesis G70 doing the same thing at high speeds. The issue was not enough rear downforce. Slapped a wing on it and she's planted.
 
C1158D9A-9895-4AB6-A2FC-3FD4C18ABEA5.png

This Poor Horse Passed Away 4 Months Ago
RIP Unknown Steed, the beatings have left him unrecognizable

This thread caused so many unnecessary tempers to be flared through the entire winter
 
Last edited:
Wow, lot's of discussion here.

I'm pleased to report that the recent physics update has fixed my issue. The Slipstream challenge with the Veyron now behaves as expected. The car won't start sliding like on ice like it did before. On SSX I can use any suspension, any tyres, any aero, and the Veyron will be stable.

I noticed that a lot of people have been very keen to "disprove" my claim that there was any unstability with the Veyron physics model at all. I can only say that I am not a liar. There was absolutely something wrong with it. Now there isn't.
 
Last edited:
Wow, lot's of discussion here.

I'm pleased to report that the recent physics update has fixed my issue. The Slipstream challenge with the Veyron now behaves as expected. The car won't start sliding like on ice like it did before. On SSX I can use any suspension, any tyres, any aero, and the Veyron will be stable.

I noticed that a lot of people have been very keen to "disprove" my claim that there was any unstability with the Veyron physics model at all. I can only say that I am not a liar. There was absolutely something wrong with it. Now there isn't.
I think it probably comes from the fact that this physics update has focussed predominantly on MR cars. I appreciate that Veyron is 4WD, but still mid engined. As such the stability of all cars with mid engined layouts have increased - it's just there's no others (save for the Chiron) that naturally hit the same kinds of speeds hence why it seemed an issue specific to the Veyron.

Glad it appears to be resolved, though.
 
Still waiting for the Bugatti Invitation Ticket to give an opinion :(
The best way to check the issue is the slipstream challenge, so no need to buy it.

Edit: I managed to gold the challenge months ago in spite of it being undriveable. I've just tried again and it's easy because you can actually control it now.
 
Last edited:
Computing real time aero forces from shape is possible, but I don't think Gran Turismo or any racing game does it. The simulator X-Plane uses Blade Element Theory to compute real time aerodynamics, but this is a flight simulator and planes are actually easier to do the math on than cars because they are more streamlined. BET is also not as accurate as more costly methods like CFD which can't be performed in real time, even when it comes to the types of shapes BET was designed to model well like wings and airfoils.

Typically a car sim like Gran Turismo gets aero data from an outside source (manufacturer, wind tunnel data, CFD results) and just hard codes the data as part of a car's stats. I imagine that in the case of GT they take a single CD value for drag and then have front and rear values for simulating downforce and that these values correspond to the car driving straight and flat, with best guess modifiers for when the car pitches and yaws. This is simple but effective and it basically totals the effects of all the parts of the car like you did with the addition of lift and drag without having to individually model all the positive lift and negative lift surfaces of the car. As long as the numbers used are correct or close to correct, the result should be convincing as long as the car is doing something unusual like sitting at an odd angle with respect to the ground or the direction of travel. Unfortunately, I know GT has been very wrong in the past. In GT5 the Viper ACR was horrendously undermodeled and was closer to the regular Viper SRT-10 than anything else for example. I have not played GT since GT5, so I'm less certain about how the newer games work.

Aerodynamics are functional at all speeds other than zero, it's just the effect can be small at low speeds. Racecars might have large enough wings, etc that you can feel the effect even at low speed, but it's going to be a lot less than at 100 or 150 mph. You can work out what the ratio is by squaring the ratio of speeds ie (30 mph/150 mph)^2 tells you that at 30 mph a car will produce 4% of the downforce that it does at 150 mph.

Back to GT, whether a car has wings or not shouldn't matter, the same method I outlined above, plugging in data from external sources, will work as long as the car is driving under expected conditions.

I've always had an interest in aero and aviation so I pursued a degree in the field and ended up being an engineer. I don't mind engaging in discussions like this at all.
Granturismo 5 did calculate wind data real time , The few cars with opening doors would pull hard when the doors would pop open in a crash and so sould standard cars that deformed in odd ways. I doubt PD simplified the calculations from then to now.
 
Granturismo 5 did calculate wind data real time , The few cars with opening doors would pull hard when the doors would pop open in a crash and so sould standard cars that deformed in odd ways. I doubt PD simplified the calculations from then to now.
Changing drag forces or moments doesn't require any real time calculation. What real time calculation means in this case is using fluid dynamics equations on the shape of the object to determine forces. Gran Turismo doesn't do this, instead it's likely using precomputed coefficients and modifiers to determine forces.
 
Changing drag forces or moments doesn't require any real time calculation. What real time calculation means in this case is using fluid dynamics equations on the shape of the object to determine forces. Gran Turismo doesn't do this, instead it's likely using precomputed coefficients and modifiers to determine forces.
How can you use pre calculated drag forces dynamically change with deformed objects to the degree the game knows if the deformation is on the left right or rear.. Any one who played the game knows standard cars were slower than premium cars when damaged because they deformed as one whole piece. The evidence supporting my hypothesis exist , there is no evidence supporting yours.
 
How can you use pre calculated drag forces dynamically change with deformed objects to the degree the game knows if the deformation is on the left right or rear.. Any one who played the game knows standard cars were slower than premium cars when damaged because they deformed as one whole piece. The evidence supporting my hypothesis exist , there is no evidence supporting yours.
You can have precalculated values for damaged states. If the right side of the car takes damage, add some drag value to that side of the car to make it pull to the right, etc.

I'll repeat that I don't know for sure how GT works under the hood, I'm just extrapolating from what I know and what I've experienced from the games. If you have some knowledge of how it's actually computing forces, I'd be interested to know how it's actually being done. What you described so far though doesn't require real time fluid dynamics.
 
Im just here to confirm that the issue is fixed at least on tuned cars, I don't need to run full wets on the front tires and racing softs on the rear to make my engine swapped cars be stable on a straight-line, my Veyron with a ultra-high Turbo is also fixed.

The only one that is still a bit unstable is my LS swapped Silvia Q, but that one always had stability issues at high speeds even on Route X.
 
Because it's a toy for rappers not a race car. Not enough downforce for the horsepower it has.
It literally drives like a pig.

Then again this car was only made to be for top speed only.

It broke speed records but thats about it as the Mclaren backed up its credentials by winning Lemans in 1995.

Comparsion between both the Mclaren is much more superior.
 
Then again this car was only made to be for top speed only.
First off, the post you're replying to is wrong, there was a very blatant bug in the game with some cars and thier stock suspensions causing the Veyron to be extremely unstable at high speed and to behave in a very unrealistic manner.

Secondly, this statement is also wrong. Whilet the Veyron was not built to be a race car (neither was the McLaren F1 by the way), it was most certainly designed with far more in mind than top speed only.

Is the McLaren F1 superior? Well, that'll depend on who you ask and which aspects of each car they consider to be more important. In a racing game the F1 certainly makes more sense, but in real life? That's a far more open question.
 
I do not understand this thread, I personally have no issues with the Bugatti Veyron Gr.4 (the only one I drive regularly) The other versions are not for daily races....
 
I do not understand this thread, I personally have no issues with the Bugatti Veyron Gr.4 (the only one I drive regularly) The other versions are not for daily races....
It's the road car that had issues not the Gr.4 version. At high speeds it would very easilly begin to wallow and slide massively from side to side with minimal inputs, behaviour that was about as realistic as the wheelie glitch in past GT games. I've not tried it for a while, but I've seen a few different people claim it's been fixed with the latest update.
 
Last edited:
What you described so far though doesn't require real time fluid dynamics.
Does drafting and changing wind directions (not during a running event) not support it?

I was also under the impression that retracteable headlights would have an impact, but actually it seems the long tunnel in SSRX is a part of the track with a slight change of elevation upwards (in forward direction) where the cars are all losing a bit of speed, but thats a different story.
 
Does drafting and changing wind directions (not during a running event) not support it?

I was also under the impression that retracteable headlights would have an impact, but actually it seems the long tunnel in SSRX is a part of the track with a slight change of elevation upwards (in forward direction) where the cars are all losing a bit of speed, but thats a different story.
No, and while this can most certainly be done, a lot of this stuff is often deemed to be too complex and resource heavy to be worthwhile taking valuable computing power to calculate in real time.

Ultimately, all you need to simulate drafting is a X position v Y position calculation and sick a formula in that that reduces the wind resistance value to allow the car to speed up/slow down.

You can fudge vehicle deformation affecting the aerodynamics quite easilly too with the right formula and criteria to look at.
 
Aero is a complex thing. So many people think airplane wings push them up rather than the physics that says they are pulled up so it's always gone be contentious.
 
Back