GT6 Top Speeds Are Not Realistic, Please Fix PD!

[jlmcmillan1978]

I haven't seen another thread discussing any possible changes to drag coeficient and aero in general yet, and after some car testing I did today I wanted to put this up for discussion.

Just bought the F40 in GT6 and set it up just as I had it in GT5. 472hp, stock gearing, stage 3 weight reduction to get it down to that early F40 1100-1200kg weight range. A twin plate clutch fitted because real life cars came with one. Now the top speed of this car is supposed to be around 200mph and in GT5 that was certainly the case. It would almost achieve this speed in 4th before just being able to pull it's 5th gear, so tall was the gearing. And yet, in GT6 this car pulls right through all its gears and does over 370km/h!!

What gives? PD has not changed displayed ratios at all in the car settings section. They are the same as they were in GT5, and are the correct F40 ratios despite many wanting to argue in other threads that they are not. The car had an accurate top speed before and now it's just not accurate for a stock car, way too fast.

Has anyone else noticed that the cars in general are more slippery in GT6 or something? I'm going to do more testing with other cars but for the moment I can only assume that PD has not modeled drag very well in GT6 or something. Your thoughts everyone?

 

[Vegard]

The gear ratios may be correct on screen, as they are with the McLaren F1 and others, but the speed in gears is still off.
Look at these videos and do a comparison in GT6.

Can you empty 3rd gear in the F40 on the back straight?
Same with the McLaren F1, the numbers seems to match but the speed in gears is waay off (3rd in game is longer than 4th irl)


But to the main subject.
Do you test at SSRX? The air resistance seems to be absent on that track.
Try SSR7 or La Sarthe for more realistic results.

Edit: note that the F40 has a dog leg 1st gear, the McLaren has a traditional H pattern.

 

[YZF]

Yes, i have also noticed this problem and indeed top speed is flawed (wrong) for almost all cars in GT6. For example you can do ~410km/h with 700HP Ford GT40 which is total nonsence.

The stock AMG SL55 in GT6, with ~490hp, does ~370km/h ant hits rev limiter in top gear. This is also completely wrong and you can't achieve such numbers in real life. In reality it does about ~325 km/h !!

Aero physics must be fixed as well !

 

[jlmcmillan1978]

Well I did some more testing, and the results are really bad.

I loaded up GT5 and tested certain cars. I looked for their absolute maximum speeds on the downhill at SSRX, which is often faster than the cars will pull themselves on a level surface, but wanted to see just what the potential maximum difference was between titles.

F40 in GT5: 334km/h. GT6: 371km/h! NSX-R GT5: 290km/h GT6: 318km/h! Ford GT GT5: 355km/h GT6: 399km/h!

It's just ridiculous really, the top speeds in GT6 are completely unrealistic, and GT5 gave us a much better representation of top speed vs real life. The Ford GT is now modeled with the correct 5th gear ratio (vs the wrong one in GT5 ) meaning it pulls 6th gear quicker but the other two have absolutely no changes to ratios between the titles.

Also, I tested the LFA and found that it would pull itself to the rev limiter on the downhill in GT5 achieving 345km/h, but in GT6 it can pull itself to the same speed before you even hit the hump! So the cars are accelerating far harder than before which would again seem to point the finger at aero and drag being the cause.

I'll test at other circuits now, because I'm REALLY hoping that this is just an issue with SSRX. Because if it's the same at all circuits then it is 🤬!!!

I want a realistic driving simulator, not a game that seeks to satisfy 9 year olds with stupidly high top speed figures!

 

[challengerrt10]

In GT5 stock F40 has the maximum speed of about 320 km/h. In GT6 it reaches 350-360 km/h at SSX. At La Sathre I reached in F40 340 km/h as far as I remember. So top speeds are broken at all tracks.

 

[HBR-Roadhog]

The only cars I have paid attention to is the Nascars and that is only because I went back to GT5 to see how I had set up my Nascar back in the early days of GT5. In GT6 with the same set up I can hit 207 mph on Daytona without drafting. in GT5 I could easily hit 224 MPH in the same car with the same settings. Clearly this car is not slipping through the wind easier than before. Can't say about the others.

Other cars I have driven on Nurburgring still seem to hit pretty much the same speed as they did in GT5

 

[coolboy48]

There could be an incorrect tire size specified affecting the overall diameter (hidden in the physics) and thus causing the incorrect RPM to actual speeds despite the correct gear ratios. Ran into that a lot with another game I played and modded. I use the 'speed at max rpm in X gear' that some enthusiast publications reveal.

I'd agree SSRX's atmosphere is definitely 'off' too though I can't say anything about the other tracks. I wonder if temperature and altitude affects the top speeds too.

 

[YZF]

To me, it seems that the aero physics code is wrong at the end of the acceleration.

If you take (imagine) car's acceleration from 0 to top speed as one graph, from 0% to 90% everything seems more or less ok. But from 90% to 100% there should be more air resistance and thus the acceleration should be slower and top speed - lower.

That's why you need Top Speed track to test it, because on Nurburgring there is no place to reach 90%-100% range

It's either code error in the last part of the whole acceleration OR aero coefficient for cars is just wrong (number is too low).

 

[jlmcmillan1978]

I won't be able to test any more for a couple of days as I've gone away for the weekend, but I'm pretty keen to establish what it's like at Sarthe & Monza with no chicanes and the Daytona oval. If the speeds are too high there as well then it's safe to say that the aero is indeed broken, at which point I think I'll just pack up the game until they fix it.

However, I have thought of one thing that might be causing the problem. Like many people, I had to delete the game data from the original install in order to get the latest update to work without crashing my PS3 constantly. I did have my concerns that this might be deleting some important stuff that the update did not entirely replace. To anyone that knows what your game data is made of, is it possible I've deleted some important aero data or something? Might try a complete re-install seeing as I had not got all that far in the game yet anyway.

 

[jlmcmillan1978]

The gear ratios may be correct on screen, as they are with the McLaren F1 and others, but the speed in gears is still off.
Look at these videos and do a comparison in GT6.

Can you empty 3rd gear in the F40 on the back straight?
Same with the McLaren F1, the numbers seems to match but the speed in gears is waay off (3rd in game is longer than 4th irl)


But to the main subject.
Do you test at SSRX? The air resistance seems to be absent on that track.
Try SSR7 or La Sarthe for more realistic results.

Edit: note that the F40 has a dog leg 1st gear, the McLaren has a traditional H pattern.

In GT5 at least, I don't know that they were really that far off for the F40. You can see from that Best Motoring video that he never uses more than the first 3 gears at Tsukuba, so tall is the gearing, and I find that matches up well with my experience with the car in GT5. He short shifts to 3rd in the runs out the 1st and 3rd corners, in the game I hold 2nd, but other than there aren't too many differences.

What gets annoying is hearing people complain about tall gearing and stuff because in their minds the car shouldn't drive like that, never mind what's realistic. So I'm thinking, have PD put less drag in the game so that all cars accelerate faster, purely to appease the masses who wouldn't know a car from a corn cob? I sure hope not.

 

[Tornado]

I'm going to link these two posts together, because the information in them is directly related even though the threads are different:

I think I can explain this one.

The F40 as modeled in GT5 & 6 is a curious hybrid. There were two different types of F40's built, or three if you count pre-production prototypes & the first 30 or so production models. The two main specs however were Euro-Spec and the USA-Spec cars. Gran Turismo describes the F40 as a '92 model and by 1992 all the cars being built (I believe) were the USA-Spec cars. They had a lot more weight then the Euro cars due to to things like a heavier exhaust with cats, exterior bumpers, and aluminium fuel cells. Ferrari never quoted any changes to power, but it is believed that the cats made them less powerful too. All of this led to a slower F40. Now I've never read anything to confirm this bit, but it seems that Ferrari attempted to compensate for this with lower gearing for the US models. Internet searching reveals the same, tall, 2.9 final ratio for both specs, but the gearbox ratios are lower than those used in the the first cars.

So where does this leave the Gran Turismo F40? It is listed as a '92 car and has a US-Spec '92 car curb weight. Power is listed as the same as early models (478PS). However, visually, it is a very early model Euro F40 with plexiglass windows that were dumped very early in the production run because customers did not like them. By 1992 no F40 came with them. Along with other light weight measures, these sliding plastic windows helped to make the earliest F40's very light at around 1100kgs dry, but they got heavier and heavier from there. You can get your GT car down to this early car weight with a Stage 3 weight reduction and plastic windows mod. The gear ratios? Supercars.net and many other sources list them as 2.77, 1.71, 1.23, 0.96, & 0.77, 1st through 5th, and these are the exact ratios you'll find used in the game. PD did not get this wrong, the ratios are the correct ones as used by European F40's, however if it is correct to say that by 1992 all cars were US Spec, then the ratios should be 3.69, 2.30, 1.64, 1.28, & 0.95 for a '92 car.

In summary, give this car it's weight reduction as described above, and you have a perfectly correct Euro-spec '87 model F40 with perfectly correct gear ratios, that is incorrectly described as a '92 model.

What gets annoying is hearing people complain about tall gearing and stuff because in their minds the car shouldn't drive like that, never mind what's realistic.

Note the final drive. The F40's speed@gear numbers aren't even remotely close to that in GT5/GT6.

 

[jlmcmillan1978]

I'm going to link these two posts together, because the information in them is directly related even though the threads are different:

Note the final drive. The F40's speed@gear numbers aren't even remotely close to that in GT5/GT6.

That's well and good, but it happens to be the ONLY reference I've ever seen to the F40 having anything other than a 2.90 final drive ratio, look at these:

http://www.supercars.net/cars/638.html

http://www.maximum-cars.com/Cars/Car.php?carnumber=336

http://www.top-supercars.com/ferrari_f40.php

http://www.carsession.com/car-specs/1987-ferrari-f40.html

and the US cars:

http://www.supercars.net/cars/6021.html

so who is right? All I know is that just about every resource out there lists a 2.90 final ratio.

 

[Vegard]

In GT5 at least, I don't know that they were really that far off for the F40. You can see from that Best Motoring video that he never uses more than the first 3 gears at Tsukuba, so tall is the gearing, and I find that matches up well with my experience with the car in GT5. He short shifts to 3rd in the runs out the 1st and 3rd corners, in the game I hold 2nd, but other than there aren't too many differences.

What gets annoying is hearing people complain about tall gearing and stuff because in their minds the car shouldn't drive like that, never mind what's realistic. So I'm thinking, have PD put less drag in the game so that all cars accelerate faster, purely to appease the masses who wouldn't know a car from a corn cob? I sure hope not.

The rev limit used in GT is wrong.
If you look at the video you can clearly see there's a big difference in the speed in gears.
The McLaren emptied 3rd in that video, at Tsukuba. It is much the same story as with the F40.

 

[Cosbuster]

Anything regarding top speeds is a total joke currently. Is this the advertised "improved" aero model? How exactly did they improve it? By removing air from the equation?

 

[omgsean1]

Delete.

 

[Tornado]

auto und motor sport actually tested that car and got those speeds in those gears. Those speeds were roughly backed up by a Car and Driver test of the European model in 1991, which shows the axle ratio as being 2.90 with a 1.33 transfer gear. Compare those speeds in gear to this:

Even allowing for the redline on the car in GT5, the car is still pulling the same speeds in 3rd and 4th as the real car achieved in 4th and 5th, causing 5th gear to practically be an overdrive gear when in real life it was not. Plugging in the 3.62 number and all of the other specs into this calculator nets a speed in gear number very similar to the real car, whereas the 2.90 number GT5/GT6 uses nets pretty much the same speed numbers that GT5/GT6 do.

 

[Lawndart]

My guess is though there is aero it's not stacking the calculation through the load of the suspension and into the tire model, let alone the need to factor for drag based on physical shape vs just spading and bogging the engine.

If everything is in fact systemic with the physics model then it would do a lot more than it does now. It feels as though it still just a base grip value adjusted to speed and load value of the wing working in the background.

One visual indication is all the LMP cars bouncing on rough tracks like LeMans. If they used a proper CFD and balanced it to the collateral suspension of the cars then you wouldn't see the suspension move or body roll. What your seeing is rebound, but with max aero your suspension is heavily bound. Just imagine a 1500 lbs car being 3500+ lbs at full speed with downforce...

Visual indicators can be very miss leading in a digital medium but PD pride themselves on their systemic approach to modeling physics... And this is likely how they "balance" 1200 cars...

 

[jlmcmillan1978]

Well there sure seems to be a massive difference between real life top speed figures & those that you can achieve in GT6. If gear speeds are way out too, then regardless of which in game error is most to blame, none of it is pleasing to know about. I'm so disappointed that PD has failed to get these things right, it really ruins the game for me to know that Im not driving a realistic simulation.

I apologise too, because with regards to the F40 it really is off, I took a look at every in car F40 vid I could find, & there isn't a single one that can possible defend that gearing in Gran Turismo. However in GT5, at least the top speed was accurate. In GT6 we have the same gearing issue teamed up with awful changes to the drag and aero modeling that creates a way out there top speed. Sheesh, it's so sad to see PD going backwards..................................

Meanwhile, I will take care to make sure I don't bite into a cob of 'car' next time I'm having dinner!

 

[jlmcmillan1978]

Went to Sarthe 2009 no chicanes in GT5 with 3 cars. Then, after a complete re-install to make sure there would be no missing aero data in my game or anything else wrong, tested the same cars in GT6. Thankfully, it took no time at all to tell that there was a difference at Sarthe compared to SSRX, definitely a more realistic drag level, yet still much less drag than GT5:

Acura NSX 269km/h (GT5) to 281km/h (GT6)

Honda NSX-R 274km/h (GT5) to 285km/h (GT6)

Nissan GT-R R32 253km/h (GT5) to 262km/h (GT6)

The manufacturer claimed top speed of the Acura in real life is 270km/h, the NSX-R 280km/h, so these particular readings are no that far off thankfully. 5-10km/h either way doesn't seem too bad to me, but the problem is that the faster you go the bigger the difference becomes, so very high powered cars in GT6 are going to do a lot more than what they can realistically do, not just a little. Power is a much greater advantage in GT6 compared to GT5 and the only thing I can put this down to is less drag at high speeds. I have no idea if this has any effect on downforce. Unless there is less tire grip but, everyone is going to be much faster in GT6 than they were in GT5.

And I have no idea why SSRX has less 'air' than other circuits...............

As for the F40, some good news for those who would like to be driving a relatively realistic version. The 5-speed close ratio box when fitted gives the car more accurate gearing (not perfect, but better) and with it a top speed on the Mulsanne with no chicanes of 326km/h. Very close to real life top speed records.

I hope that PD fixes the very high speed drag issue so that cars with big power and tall gearing are not able to do Veyron-smashing speeds where it's not realistic for them to. A 550hp Ford GT should not be doing almost 400km/h. Makes me wonder actually, just what a stock Veyron in GT6 is capable of.

 

[YZF]

Stock Veyron, surprisingly, does ~410km/h on RouteX, at the end of 8km straight. Exactly like in real life...

 

[jlmcmillan1978]

Stock Veyron, surprisingly, does ~410km/h on RouteX, at the end of 8km straight. Exactly like in real life...

It's getting difficult to find some consistency in this! Does the Veyron hit it's rev limit in it's top gear at that speed, or will it just not pull itself against it's own drag any faster than that?

 

[YZF]

I have to test this later, but I think it doesn't hit the limiter...although i might be wrong.

 

[xSNAKEx]

It's getting difficult to find some consistency in this! Does the Veyron hit it's rev limit in it's top gear at that speed, or will it just not pull itself against it's own drag any faster than that?

Probably because the Veyron is known for its top speed record holding. They managed to get that right by matching the data. They appear to have gone lazy on everything else.

 

[research]

can I just say that I love you gies, for doing all of this testing?

👍

 

[xSNAKEx]

One visual indication is all the LMP cars bouncing on rough tracks like LeMans. If they used a proper CFD and balanced it to the collateral suspension of the cars then you wouldn't see the suspension move or body roll. What your seeing is rebound, but with max aero your suspension is heavily bound. Just imagine a 1500 lbs car being 3500+ lbs at full speed with downforce...

cars still bounce when pressed down from down-force. There is still a little suspension travel left and if the suspension cant compress then the whole car will move up and down. Actually both happen.

Its not like the car goes through the floor.

 

[Lawndart]

cars still bounce when pressed down from down-force. There is still a little suspension travel left and if the suspension cant compress then the whole car will move up and down. Actually both happen.

Its not like the car goes through the floor.

You feel DF because your sitting in the shell being pushed down so the suspension sensation you feel is more of the rebound over bumps because the suspension is heavily bound (not bottomed out). Bumps make the car feel more like its sinking than bouncing at speed, we all know what happened to Webber in the CLK when a car bounces at speed... The air needs to go somewhere... So GT 6 still isn't simulating air.

If one of my real racecars "bounced and rolled" at 160+ mph (without flipping) as much as we see in replays I would be like a piece of meat being shaken in an empty can violently.

 

[YZF]

As of patch 1.04 the wrong (broken) top speed is still not fixed. Cars still have way too high max speed!

 

[tuj]

My fully modified Supra will hit 310mph in the draft with nitrous. It only has 800-something horsepower, no way a car hits 300mph, not a regular car like the Supra. Maybe that Auto Union monstrosity could do it for real (298mph?).

Its almost like they got the exponent wrong on air resistance.

 

[BlueShift]

The manufacturer claimed top speed of the Acura in real life is 270km/h, the NSX-R 280km/h, so these particular readings are no that far off thankfully. 5-10km/h either way doesn't seem too bad to me, but the problem is that the faster you go the bigger the difference becomes

Air resistance is R * v² where v is the speedof the air flow.

Assuming no wind if you got "1" air resistance @ 1 km/h, you will have :
- 4 @ 2 km/h
- 9 @ 3km/h
...
- 72900 @270km/h

That's a normal situation to me. If they got it wrong by a tiny portion at low speed, then at high speed this became a big fraction.

e.g instead of 1, they used 0.99 => the 72900 becomes 72171 (diff = 729)
at 271, diff = 734.41
at 272, diff grows again to 739.84, etc

And I have no idea why SSRX has less 'air' than other circuits...............

Again, perfectly normal there. Sea level air.

The altitude of a track, temperature (so, day and nigth) and % humidity (day & nigth again) play an important part in the "R" factor of R * v². Not to mention the wind, there's some transversal wind on SSX aswell.

 

[Boone]

If all the forces acting on the car should equal zero, are we seeing negative numbers translating to higher top speed?