GT6’s Game Breaking Pit Stop Flaw

  • Thread starter MGR
  • 33 comments
  • 8,717 views

MGR

335
MGR
Updated 15/5/2014

While not formally acknowledged by PD it seems patch 1.07 has resolved the issue finally!


OP Updated 17/4/2014

”In Gran Turismo 6, we have collaborated with leading companies in the racing scenery such as Yokohama Rubber and KW Automotive as technical partners. Insight and data for developing real cars in the partnership has been fed back into the physics model which governs the dynamics and behavior of the cars in the game.”

http://www.gran-turismo.com/au/products/gt6/technology/

All of which are completely broken the second you perform a pit stop.

Background
Ever noticed after a pit stop that your car:
- Starts handling like a turd and you can’t hit the same lap times as before?
- Rips through rubber at a rapid rate?
- Screeches tyres even when traveling in a straight line?
- Suddenly starts bouncing off the limiter down the straight?

Introducing the pit stop bug! (est. 6 December 2013)
Upon pitting it takes all that collaborative effort with Yokohama and KW and promptly flushes it down the toilet. Online and Offline.

Lap times
First set of tests I ran included using a few different types of cars and compounds around Motegi Speedway. I chose this track because you can load up the car through the long gentle turns to test grip and still reel off laps within a tenth or two easily. In these examples I strung three consecutive laps together and averaged the times.

l8smNM3.gif


For the second set of tests I only used two cars on a range of tyre compounds. Please note I de-tuned the 787B with the lower grip tyres. The purpose of the test is to check consistency, not outright speed.

XGM7ibf.jpg


Two very different cars obviously but the results were still surprising. Whatever physics bug is causing this discrepancy seems to manifest itself in different ways. It's possible that grip may be reduced, but in some cases may also improve balance keeping lap times similar. But under no circumstances does the post-pit tyre set improve laps times.

Tyre Wear
Most people will also notice excessive tyre wear after pitting, especially with race tyres. To illustrate this I ran some controlled tests around SSRX. This track produces very little tyre scrub as the turns are long and banked. The tyre degradation was set to ‘Fast’ for this test.

Toyota Supra (Tuned) – Racing Soft (RWD)
Toyota Supra (Tuned) – Full Wet (RWD)
Ford Focus ST (Tuned) – Full Wet (FWD)
Escudo Pikes Peak – Full Wet (4WD)


2wrqb09.png

As you can see a full lap on the pre-pit tyres barely shows any signs of wear. In fact the tyres actually seemed to cool and turn blue. Obviously PD has not implemented any heat generation from sidewall flex. The Supra sustained a constant 450km/h+ without the tyres warming up ;)

What is interesting though is it seems only the driving wheels suffer from excessive wear after the pit stop. Also, there is a constant tyre screeching noise produced after pitting – which increases with the speed of the car. You can easily see the tyre wear graphics change as the tyres are ripped apart and screech all the way down the long SSRX straights.

Another by-product of the pitstop it that the camera view shakes much more violently at speed making the horizon to become something of a blur. The cars (especially the Supra) also become much more nervous even around the gentle high banked turns of SSRX.

RPM Changes
Many have also noted cars hitting their rev limiter much earlier in top gear after pitting. While the data logger is now available it is not able to log data from races – therefore no direct pre and post pitstop comparison can be made available using this feature (as pitting & tyre changes are only enabled in races).

So first up I ran a series of controlled race tests on SSRX using the Escudo and saved the replay data. I then captured the replays onto a PC and reviewed the video by pausing the car at approximately 8000rpm (as per the tacho needle) and noted the digital speed readout.

The second test was performed using the same car around Daytona – this time in free run mode and saving the logged data. I then used the data logger to review the various speed values available while accelerating at 8000rpm.

33574nd.png


Based at viewing the replay video playback I noted a significant difference in the recorded speed between the original and pit set of race tyres. While I fell all tyres are affected by this bug the race tyres defiantly seem to be the most problematic. This speed discrepancy between pre and post pit stop is probably a significant factor.

Also the speed of the car seems to increase with the sticker tyres. Initially I thought there was something screwed up with the tyre circumference calculation but upon reviewing the logged data it would seem that PD have implemented some sort of tyre slip calculation depending on whether the car is accelerating or decelerating. To check this I ran a set of controlled test around Daytona using three types of car on Racing Soft tyres.

2w4g2sl.jpg


As you can see the rear driving tyres are slipping slightly under power as they attempt to push the car through the increasing air resistance resulting in a slightly higher wheel speed than the actual speed of the car. But when decelerating the natural engine braking has the opposite effect. In both cases the non-powered wheels match the cars actual speed.


2hx0vp2.jpg


This time it’s the front driving wheels that are trying to drag the car forward faster than its actual speed. The non-powered rears match the actual speed of the car.

Four Wheel Drive

2rqllvq.jpg


Now all four wheels are attempting to push the car through the air resulting in a slightly higher wheel speed than actual speed. Upon deceleration this is reversed with engine braking.

This animated GIF is taken from gameplay and shows the effect in action.
- Car coasting down to 7000RPM under engine braking
- Hit throttle open and the car jumps 500RPM but only 3km/h
- Car then adds another 500RPM but this time an extra 10km/h

Wlhevmq.gif


Theory

This tyre slip calculation seems to be at the heart of the problem. Something or some variable is calculated or reset incorrectly during the pit stop routine. This would explain why the tyre screech becomes clearly audible after pitting even in a straight line – especially using racing tyres. The massive difference in speeds recorded pre and post pit stop for the racing tyres is quite likely evidence of what is contributing to the flaw.

Every time a player enters the track the bug is not present. As soon as a pit stop is performed the game goes into a ‘bugged state’ but this is reset by exiting the track (or returning to the lobby) and re-entering the track.

This really should be a simple bug for PD to resolve.
 
Last edited:
Great analysis. But, I have a noob (actually, I never race online and only hotlap, so I never have to pit) question, is this offline and online?
 
I would definitely classify that as a game breaking bug. That is a HUGE flaw in the tire model IMO. It was bad enough this was in the game PD/Sony charged full price for on launch day, but this still has not been paid any attention to by PD while they throw new VGT and Toyota concept cars at us and waste time creating videos to tease tracks that don't release when a Sony representative says it will.

It's now been over 4 months since the game launched with 6 patches and nothing has been done to address this OR the camber issue, which just means the suspension model is not at a stable/market ready state either.

It's been 7 years since the PS3 released, and PD still can't develop like a AAA studio should, especially with the time frames and funding gifted to them.
 
Last edited:
Wow... Now I am definitely going to use the Delta Wing and LCC Rocket in my S Class events on this second run through the game so I can avoid this dumb bug. I'm glad I only do short sprint races online where we don't have to do the first pit stop on normal tire wear.
 
MGR
All of which are completely broken the second you perform a pit stop.


Background
Ever noticed after a pit stop that your car starts handling like a turd? Rips through rubber at a rapid rate? Screeches tyres even when traveling in a straight line? Suddenly starts bouncing off the limiter down the straight?

Introducing the pit stop bug! (est. 6 December 2013)

Upon pitting it takes all that collaborative effort with Yokohama and KW and promptly flushes it down the toilet.

I started doing some testing a few days ago to try and measure the effect of the bug. Few different cars, few different tyre compounds on Motegi Speedway and here were the results:

l8smNM3.gif


I decided to run more testing this evening to back up the data. First some more laps tests;

Motegi Speedway lap times.
Pick a car, punch in three consecutive laps and take the average times. Please note I de-tuned the 787B with the lower grip tyres. The purpose of the test is to check consistency, not outright speed.

Bs2SFuM.jpg


Two very different cars obviously but the results were still surprising. Whatever physics bug is causing this discrepancy seems to manifest itself in different ways. It's possible that grip may be reduced, but in some cases improves balance which keeps lap times equal. But under no circumstances does the post-pit tyre set improve laps times.

Next up - Accelerated tyre wear test
Standard Escudo Pikes Peak - SSRX
2 laps with pit stop.
Full wet weather tyres
Fast tyre/fuel depletion

DPOvaEh.jpg


At the end of the first lap the tyres on all three cars had barley done any work. Only the highly tuned Supra showed slight wear on the left rear while the fronts cooled over the run. Which is slightly bemusing given it maintained an average speed of 450km/h. Obviously the tyre physics are only heating up due to corner scrub.

After the pit stop the Supra had worn both rear tyres down to 7 before the first bank. It didn't even make it to the second banked turn as the tyres wore to 1 and 2 respectively, sending it into a massive spin on the straight. The front tyres didn't seem to be affected at all but the rear (driving) wheels were getting torched.

So I decided to switch to the 4WD Escudo. Once again lap 1 showed virtually no tyre wear and the rubber remained cold. After the pitstop all four tyres started to wear noticeably, although the rears were a little worse.

Then I tuned up (power wise anyway) a Focus ST. Again, all fours tyres not worn after lap 1 and remained cold. After the pitstop the rears didn't suffer, but the front driving wheels showed significant wear.

In all cases above I noticed the following:

Constant tyre screeching in a straight line
I know many already are aware of this, but if you want to replicate it for yourself then throw a set of full wets on during a pitstop. It happens with all the tyre types, but the full wets give the loudest screech.

Car shakes violently after pitstop
The game tries to replicate cockpit movement somewhat anyway, but after a pitstop this shake becomes much more violent. The horizon becomes something of a blur because the camera is shaking so much.

Finally - The ratio test
Tests were run on SSRX before and after pit stop.
Measurements were done by capturing the recorded replays and frame advancing the footage until the tacho needle hit 8000rpm.

3FKLIqi.jpg


The race and wet weather tyres are broken. Also seems each tyre increases in diameter as the compounds get softer. No idea why that is, but the purpose of this exercise is to check consistency between the first set of rubber and the set fitted during the pit stop.

-----------

Seems like the game is referencing one set of default values when you enter the track (values for tyre compound, diameter etc) and another (screwed up set of erroneous data) when performing a pit stop. Whenever you go to track the tyre values seem correct lobby (but leave your car entered or 'on track'), when you go back to the car the tyre values are correct again and the bug is gone.

Whatever values are referenced during a pit stop scenario must be screwed. We know for a fact that the tyre circumferences are wrong for the racing & wet tyres and something is 100% getting screwed in the physics department at least with the driving tyres. There’s no questioning the results of the speed and tyre wears tests. All can be easily replicated by anybody.

4 months on and this bug is still ruining GT6 racing.
Great post/work!
Just to nitpick your tires are in the wrong spot (left is on the right and vice versa).

What's that thing about tire circonference? I don't understand the "we know for a fact" part...
 
@MGR
Also seems each tyre increases in diameter as the compounds get softer.

Check data logger with some 2wd car, you'll see at drivetrain wheels are going faster(skidding), not rolling wheels.

So skidding decreases when putting better compound tyres under car. Same behavior was on GT5 also, thinking at this might be some result of combined physics from GT5 and GT6.
 
Thanx for posting this, it's been really annoying me trying to do the le mans 24 min race. The speed 8 just wheel spins along the straight and constantly screeches, even at low revs.
 
MGR
Next up - Accelerated tyre wear test
Standard Escudo Pikes Peak - SSRX
2 laps with pit stop.
Full wet weather tyres
Fast tyre/fuel depletion
-----------

You didn't say which kind of tyre you used in this part. Could you specify, please?
 
MGR
Look again, it's in there ;)

Well, I just made a fool of myself. I just read your post quickly and was looking for the info below the figure. I didn't realized it even when I quoted it.

Sometimes I forget that this game have weather tires as I never use them. Maybe I'll try that test with Comfort tires, as those are the ones I use the most.

Thank you for the answer.
 
MGR
While I agree with your sentiments in general, this bug and the physics model accuracy are two separate issues. Yes camber is broken and the tyre model still needs lots of work, but at least it’s consistent for everybody and doesn’t affect gameplay.

At it’s core the game is about competition against other drivers and the clock. Any bugs that effect the core gameplay are inexcusable.

I disagree, it absolutely does affect game play. Not everyone who plays GT views these forums or does research on the game like we do. My recent ex co-worker (and good friend) plays GT and he had no idea about the camber issue. After I told him he realized why all his cars were not driving like he thought they should.

Same thing with the tires / pit stop glitch. That affects game play both online and offline or "career mode."

As you stated, it's inexcusable. I stand by my original post.
 
@MGR,
Still, it would be interesting to see what data was being logged by the game post pit stop. But you can only capture ‘Best Lap Data’ in free run mode, which does not let you do pit stops.

Actually some tracks what allows you drive thru pit on "time trial" will change tyre behavior even just doing drive thru, might be worth of testing/checking.
 
I came to the conclusion a while ago that the bug is in fact with the starting set of tyres, not the following sets.

RWD cars in the first stint have a lower degradation on the rear than on the front, the stints after pitting although slower, wear down as you would expect.
 
I disagree, it absolutely does affect game play. Not everyone who plays GT views these forums or does research on the game like we do. My recent ex co-worker (and good friend) plays GT and he had no idea about the camber issue. After I told him he realized why all his cars were not driving like he thought they should.

Same thing with the tires / pit stop glitch. That affects game play both online and offline or "career mode."

As you stated, it's inexcusable. I stand by my original post.

I disagree, it absolutely does affect game play. Not everyone who plays GT views these forums or does research on the game like we do. My recent ex co-worker (and good friend) plays GT and he had no idea about the camber issue. After I told him he realized why all his cars were not driving like he thought they should.
Same thing with the tires / pit stop glitch. That affects game play both online and offline or "career mode."
As you stated, it's inexcusable. I stand by my original post.

There are many, many anomalies with the GT6 physics engine (camber adjustment being one of the worst offenders) and it will never be perfect. But the debate on the accuracy of the implemented physics in GT6 in comparison to real world physics is a separate issue.

Take a first-person-shooter for instance:

People will argue endlessly about how much recoil a particular gun should have based on it's real life counterpart. But ultimately the competitive game play aspect is not affected because the gun's recoil is the consistent and the same for everybody, all the time.

Sticking with the FPS analogies, imagine reloading at an ammunition pickup in the middle of a game - which in turn made all your guns have terrible recoil and accuracy for the rest of the match. That's the equivalent of what we're discussing in this thread.

The pit stop bug alters the laws of physics mid game.
The camber issue does not, but it still sucks.
 
It seems to me that it's a very easy fix for the tyres' diameters and wear rate. Brilliant work on figuring out where the problems lie. Could it just be the case that GT purposefully did that for the career mode? I haven't completed it yet, but I can't see how it's taking them to fix this.
 
I started this game in February with no knowledge of any of the bugs (shoulda checked my:gtpflag:). When I did my first "extended" race at the Spa, everything was going fine (though I did notice my usually s:censored:t-n-git ZR1 Race Car was not as settled as it was in GT5....I'm looking at YOU, camber:idea:) and I was leading when the pit stop came due. My wife had to intervene approximately two minutes later because of the fuss I was raising at how the car was handing. I actually assumed at that point it was PD either making any mistakes count much more because I was now leading, or that some "boost" mode had been assisting me while I wasn't leading and now had been removed. All this ^^^ makes much more sense now. Good thread:tup:. :cheers:
 
MGR
There are many, many anomalies with the GT6 physics engine (camber adjustment being one of the worst offenders) and it will never be perfect. But the debate on the accuracy of the implemented physics in GT6 in comparison to real world physics is a separate issue.

Take a first-person-shooter for instance:

People will argue endlessly about how much recoil a particular gun should have based on it's real life counterpart. But ultimately the competitive game play aspect is not affected because the gun's recoil is the consistent and the same for everybody, all the time.

Sticking with the FPS analogies, imagine reloading at an ammunition pickup in the middle of a game - which in turn made all your guns have terrible recoil and accuracy for the rest of the match. That's the equivalent of what we're discussing in this thread.

The pit stop bug alters the laws of physics mid game.
The camber issue does not, but it still sucks.

I get what you're trying to say, but it's apples to oranges. You can make adjustments to your gun in FPS that would adjust the recoil, if there were glitches in that coding that provided an advantage or it was a broken feature, it affects gameplay.

Also, the camber adjustments do break the laws of physics, by working the opposite as it should. Also as I stated before, not everyone who plays knows the camber is broken, so someone who dials in camber to their cars and goes and races someone online who knows 0 camber ATM is more grip has the advantage, thus affecting game play.
 
I find it really interesting that the rear tires are spinning faster than the front ones. It may explain why the rear tires squeal more and break grip sooner.


Jerome
 
Do you guys think that GT6 physics engine might be broken beyond repair?
Roll-overs that surpass belief, aero that doesn't affect PP, epic camber and tire effects failures, a game where online participation is obviously overstressed this time and yet by all accounts online play is massively flawed, and consoles are dying/malfunctioning across the land in epic numbers while PD fiddles with GPS cars on three race tracks. I think more than the physics is flawed here.
 
Good effort on this!

I think it's not just tire slip after pit stops that causes the problems though, I think tire slip in general is problematic (and this could be entirely separate from the pit stop issue...).

The reason I say this, is that I have experienced that there are some cars which will have audible tire slip even at very high speeds in a straight line. This is best noticed when using headphones. One example which is very obvious is the Mercury Cougar XR-7. On this car the tire slip is not only audible, but the tires are clearly experiencing a severe load - you can tell because it's dramatically unstable, snapping out extremely easily.
To clarify - this is in any offline mode without tire wear.
 
Thanks @MGR for the excellent analysis and I only hope that PD is reacting at some point in time....

So far there seems to be no need from PD for corrections, as this had been discussed in various posts since the beginning of the year (or better when GT6 players reached the endurance races in their careers and wondered why the cars behaved so weirdly after the first pit-stop). I have lost my hopes here as the only things that the GT6 players are getting from PD are hot lap time trials and for these "wonderful races" there is for sure no need for pit-stops.
 
Excellent post MGR. One thing that no one has mentioned yet about the pit stop bug is that sometimes you can pit for tyres only, enter the pits 2 seconds behind the car in front and then exit the pits in front of him. I have noticed this a couple of times and won a few races because of it.

I always thought it was due to the amount of fuel that was added BUT when there is only a stop for tyres it still occurs.

The only thing is it is not consistent and I don't know how to reproduce it. I have a suspicion it something to do with your pit entry speed. Somewhere like Bathurst or Brands hatch where there is a tight chicane before the pit entry it won't happen but at Madrid Reverse (where I first noticed it) you can enter the pits at full speed and exit very quickly and get the jump on the car in front.

The actual times were, I entered the pits 2.0secs behind and exited 1.8 secs in front. On a last lap pit stop for a mandatory tyre change you can't get that time back. Why a last lap pit stop you ask? Well because of this tyre bug I would rather race on half worn out original tyres than change to new slippery tyres ;-)

I did tease the other guy and told him I had bribed his pit crew to fumble his wheel nuts :-) He was not amused.

Anyone else noticed this?

BTW MGR you know me as TT__777
 
Excellent post MGR. One thing that no one has mentioned yet about the pit stop bug is that sometimes you can pit for tyres only, enter the pits 2 seconds behind the car in front and then exit the pits in front of him. I have noticed this a couple of times and won a few races because of it.

I always thought it was due to the amount of fuel that was added BUT when there is only a stop for tyres it still occurs.

The only thing is it is not consistent and I don't know how to reproduce it. I have a suspicion it something to do with your pit entry speed. Somewhere like Bathurst or Brands hatch where there is a tight chicane before the pit entry it won't happen but at Madrid Reverse (where I first noticed it) you can enter the pits at full speed and exit very quickly and get the jump on the car in front.

The actual times were, I entered the pits 2.0secs behind and exited 1.8 secs in front. On a last lap pit stop for a mandatory tyre change you can't get that time back. Why a last lap pit stop you ask? Well because of this tyre bug I would rather race on half worn out original tyres than change to new slippery tyres ;-)

I did tease the other guy and told him I had bribed his pit crew to fumble his wheel nuts :-) He was not amused.

Anyone else noticed this?

BTW MGR you know me as TT__777

Haven't really tested this at all in GT6 because of the pit stop bug, but I know GT5 stop times were definitely affected by entry speed. If you blasted full speed into the pits at Daytona, the game would immediately 'reset' your car at the start of pit lane and hold you in your pit bay for what seemed like an eternity.

I had always assumed that if your car was 'reset' to the beginning of pit lane you were going to get a penalty, but if you made it in cleanly then you were OK. Thinking back though, I don't know that for certain and its possible if you did exceed the entry speed a bit then you may have been held an extra few seconds.

The biggest problem is this 'feature' is (as far as I know) completely undocumented and the game does not explain what the hell is happening or why. The speed limited areas are not clearly marked and there's no feedback from the game to say when you've received (or are receiving) a penalty.
 
Back