Example of MoTeC data analysis

[Jimmy Howlett]

I downloaded the sowtware from Motec and loaded a quick run from SSRX. I loaded it into drag, but it doesnt give me the 0-60 or 0-100 /100-200 and all that.I get the read out but it doesnt give me a "run" I did a time trail run in Arcade I stopped at the start line, revved up and let go of the clutch & went all the way to 205mph then let off and exited the run.

How do I get the data you are pulling?

 

[MatskiMonk]

I downloaded the sowtware from Motec and loaded a quick run from SSRX. I loaded it into drag, but it doesnt give me the 0-60 or 0-100 /100-200 and all that.I get the read out but it doesnt give me a "run" I did a time trail run in Arcade I stopped at the start line, revved up and let go of the clutch & went all the way to 205mph then let off and exited the run.

How do I get the data you are pulling?

Basically you need to interpret the data in front of you, it doesn't just give you the numbers, it would be nice if it did though. I'm loading everything into Circuit, I think it gives you the most options (maybe). But, you need to get the time that you let off the clutch, and subtract it from the time you hit your desired target speed. If you got the run info on the right, and you are heavily zoomed in, you can be very precise.

 

[Jimmy Howlett]

Basically you need to interpret the data in front of you, it doesn't just give you the numbers, it would be nice if it did though. I'm loading everything into Circuit, I think it gives you the most options (maybe). But, you need to get the time that you let off the clutch, and subtract it from the time you hit your desired target speed. If you got the run info on the right, and you are heavily zoomed in, you can be very precise.

Ah okay. It looks like there may be a way for it to pull the data on is own, I will try some more.

 

[ALB123]

Wow. Those test times that @VZF posted are so depressing. I've always thought that cars seemed to be too fast in GT6, but I never realized just how unrealistic GT6 truly is. It's very interesting that the Lamborghini Murcielago LP 670-4 SV and Mclaren MP4-12C were actually slower 0-100kph than the real car. But the faster you go, the less realistic things become.

What confuses me is how there isn't an easy to spot pattern to this. The Mercedes SLS AMG ends up hitting 300kph 14.2 seconds sooner than it would in real life, but the Lamborghini is only 5.4 seconds quicker to 300kph than its real life time. Then we come to the Lexus LFA which hits 300kph 24.3 seconds quicker than it could in real life. So, the LFA is reaching 300kph in HALF the time it actually takes the real car to reach that speed. It's interesting that the SLS AMG, Murcie SV, ZR1 and LFA all hit 300kph within 4.2 seconds of each other in the game, but in real life the cars have an extremely wide disparity of times to reach 300kph. In fact, the difference between the four cars is actually 20.9 seconds!

 

[MatskiMonk]

Wow. Those test times that @VZF posted are so depressing. I've always thought that cars seemed to be too fast in GT6, but I never realized just how unrealistic GT6 truly is. It's very interesting that the Lamborghini Murcielago LP 670-4 SV and Mclaren MP4-12C were actually slower 0-100kph than the real car. But the faster you go, the less realistic things become.

What confuses me is how there isn't an easy to spot pattern to this. The Mercedes SLS AMG ends up hitting 300kph 14.2 seconds sooner than it would in real life, but the Lamborghini is only 5.4 seconds quicker to 300kph than its real life time. Then we come to the Lexus LFA which hits 300kph 24.3 seconds quicker than it could in real life. So, the LFA is reaching 300kph in HALF the time it actually takes the real car to reach that speed. It's interesting that the SLS AMG, Murcie SV, ZR1 and LFA all hit 300kph within 4.2 seconds of each other in the game, but in real life the cars have an extremely wide disparity of times to reach 300kph. In fact, the difference between the four cars is actually 20.9 seconds!

I don't take this data at face value. Yes, there is obviously a problem, but PD are trying to build an accurate simulation... what would you prefer.. a system that builds up a top speed and acceleration curve using variables and calculations that apply to real world physics (and maybe doesn't work at extremes).. or a system that just took a real world max and scaled it back. One is inaccurate and is trying to be accurate, one is inaccurate and is posing as accurate.

 

[diptob79]

Just done mini endurance last night, wanted to see and play with the multi player data.
But left the usb at ps3 (at home)

sigh

 

[ALB123]

I don't take this data at face value. Yes, there is obviously a problem, but PD are trying to build an accurate simulation... what would you prefer.. a system that builds up a top speed and acceleration curve using variables and calculations that apply to real world physics (and maybe doesn't work at extremes).. or a system that just took a real world max and scaled it back. One is inaccurate and is trying to be accurate, one is inaccurate and is posing as accurate.

No, I hear what you're saying. I have never played any other "simulation" driving/racing game. I don't know how well the others do in this regard. I would have to imagine it's not very easy to accurately simulate these digital cars travelling through air resistance. The fact that a console video game is even as "accurate" as GT6 is, is pretty impressive to me and that's what keeps me playing day after day.

 

[Jimmy Howlett]

GTR Nismo Factory Stock 591hp Stock SS Tires

In Game

3.069 0-60mph & 11.002 1/4mile @ 124.7mph

Real Car

2.900 0-60mph & 11.100 1/4mile @ 125.3mph

Slower off the line, so I dont have too much grip. Ive heard the Nismo GTR will do 2.6 0-60. Its not uncommon for cars to hit numbers faster that what the factory says with good drivers & conditions, I don't take factory figures as set in stone.

 

[ALB123]

@Jimmy Howlett How did you get the video of your lap to appear in the software?

 

[Jimmy Howlett]

Record the video, then load it into the i2 software with the replay data from GT6 it auto synchronizes the video to the data perfectly. Gear shift, even throttle blips are matched up perfectly. This Motec software with GT6 replay data is quite impressive.

 

[ALB123]

@Jimmy Howlett So, you mean you recorded your lap by using a video capture device - like the way people put their lap replays on YouTube? And then just imported that video file into the MoTeC i2 software?

 

[Jimmy Howlett]

@Jimmy Howlett So, you mean you recorded your lap by using a video capture device - like the way people put their lap replays on YouTube? And then just imported that video file into the MoTeC i2 software?

Exactly

 

[Ridox2JZGTE]

Wow. Those test times that @VZF posted are so depressing. I've always thought that cars seemed to be too fast in GT6, but I never realized just how unrealistic GT6 truly is. It's very interesting that the Lamborghini Murcielago LP 670-4 SV and Mclaren MP4-12C were actually slower 0-100kph than the real car. But the faster you go, the less realistic things become.

What confuses me is how there isn't an easy to spot pattern to this. The Mercedes SLS AMG ends up hitting 300kph 14.2 seconds sooner than it would in real life, but the Lamborghini is only 5.4 seconds quicker to 300kph than its real life time. Then we come to the Lexus LFA which hits 300kph 24.3 seconds quicker than it could in real life. So, the LFA is reaching 300kph in HALF the time it actually takes the real car to reach that speed. It's interesting that the SLS AMG, Murcie SV, ZR1 and LFA all hit 300kph within 4.2 seconds of each other in the game, but in real life the cars have an extremely wide disparity of times to reach 300kph. In fact, the difference between the four cars is actually 20.9 seconds!

Bear in mind of the many aspect in stats and tire in GT6 are incorrect when stock / default The most important is fitting a tire that replicate the real car grip level ( factory ), the acceleration times reduced a lot with even 1 level higher grip tire. The real life weight of the car when tested will most often be different than in game stock weight, then actual power figure ( many cars in GT6 is incorrect ), then gearing ( make sure it's correct too ), then shifting ( use same method as in the real life test - for example, manual cars with a clutch and I doubt these real test would flat shift a expensive exotics ) Take for example, my Ferrari 512BB (CH)-Official Ferrari Curb Weight, Corvette C5 Z06 (CM)-full running weight, Gallardo SL LP570-4 (CS) - BM/C&D Tested curb weight, and F40 EU AutoMotorsport ver replica (CS) -AutoMotorsport Tested Curb Weight.

I tested the Ferrari 512BB, F40 EU AutoMotorsport and Gallardo SL LP570 on their replica tire at SSR7 ( rolling acceleration ) and SSRX back when I built them, and they were close up to 400m / quarter mile both in trap speed and time ( no flat shifting for the 512BB/F40 and launch like you would IRL ) as well as 1 KM standing start for the 512BB at SSRX - quite close to the official Ferrari Test Report in time and trap speed.

I also felt that on some tracks, the aero drag is different, take SSR7 and SSRX, the SSRX has almost no aero drag in high speed. Tsukuba is also quite unique, most of my replicas built to replicate a real life car that sets lap record there have almost the same top speed at the longest straight ( before final corner ) - some of the high powered cars can reach over 200kmh here. Take this for example :

​

Almost identical even at 200+kmh on the long straight. Have you downloaded the replay run on the video above @ALB123 and run it in the Motec software

 

[Jimmy Howlett]

Motoc i2 automatically will synchronize the video to the replay data perfectly.

In Game

3.069 0-60mph & 11.002 1/4mile @ 124.7mph

Real Car

2.900 0-60mph & 11.100 1/4mile @ 125.3mph

I also felt that on some tracks, the aero drag is different, take SSR7 and SSRX, the SSRX has almost no aero drag in high speed. Tsukuba is also quite unique

Im facepalming so hard right now. I had not considered that SSRX might have no high speed drag but this not being carried over to all other tracks. Ive always done my top speed testing at SSRX because its got so much room to play.

This opens the doors to interesting things, SSRX vs other tracks and Original tracks vs Real Tracks

 

[ALB123]

@Jimmy Howlett That's what I was afraid of. I don't have a video capture device and as much as I want to get one, mainly for making YouTube videos of my GT6 experiences, I could also use it for this now... I just think I'll wait because I don't want to buy something now that might not be compatible with the Playstation 4. That certainly IS super cool how it synchronizes to your lap perfectly. I was hoping that somehow we could export video files, somehow from the PS3 directly to be used in the MoTeC i2 software, but I had a feeling that was wishful thinking. I had a strong suspicion that a video capture device would be required. Ah well... I can still have fun with the software.

I even noticed that MoTeC has an Official Forum for asking questions related to their software and Gran Turismo 6 support. That's a pretty serious partnership, if you ask me. Major props go out to MoTeC, whether you find this feature useful or not, many people will find it useful and it shows you that PD hasn't just been sitting on their butts doing nothing these past several months.

@Ridox2JZGTE Oh, I know exactly what you mean. It might be this thread, or another, where someone was talking about 0-60mph times and I mentioned that if you really want to get closer to reality you'll need to add some ballast to simulate the additional weight of a driver. It has been my experience, and perhaps I am wrong, that when trying to find the "right tire" for a particular car, we always seem to fall in-between what GT6 offers us. For instance, on a particular car the CM tire seems to offer a fairly realistic lateral grip, but the longitudinal grip is too low. So you move to CS tires and that seems to offer a proper longitudinal grip but now the lateral grip is too high.

It's also driver dependent, in my opinion. You could let me take a sporty 550PP car around Nurburgring on SS tires and I guarantee you that someone like @mike_gt3 could take that same car and whoop my time running on SH, for sure, and probably even CS tires. He's just that much better of a driver in GT6 than I.

But back to the main issue...Aero drag is not calculated even remotely properly in GT6. I don't know if its supposed to be increasing logarithmically as speed increases or what. That's why I don't develop video games/driving simulators. I wish someone would chime in and let me know if other Sims/Driving Games are more accurate in this respect. Does Assetto Corsa or iRacing accurately represent a car's 0-300kph time? how about 200-0kph?

 

[Jimmy Howlett]

Many examples on youtube of just how close GT6 can come to the real thing, its a lot about how you ply the game.

 

[ALB123]

It's funny...There are so many different cars in my GT6 Garage that I have just randomly taken out for a rip around Nurburgring and wouldn't you know that my time ends up being almost right on the money with the real life car's official lap time! But then I remember that I'm not even that good and there are many people who can destroy the "real life" lap time. Like Ridox pointed out, that does have a lot to do with the stock tires on our cars being too grippy compared to the real world tires, but I still think that even with CS tires the really good GT6 drivers will beat those real world times quite easily. I could be wrong.

 

[Ridox2JZGTE]

It's funny...There are so many different cars in my GT6 Garage that I have just randomly taken out for a rip around Nurburgring and wouldn't you know that my time ends up being almost right on the money with the real life car's official lap time! But then I remember that I'm not even that good and there are many people who can destroy the "real life" lap time. Like Ridox pointed out, that does have a lot to do with the stock tires on our cars being too grippy compared to the real world tires, but I still think that even with CS tires the really good GT6 drivers will beat those real world times quite easily. I could be wrong.

Pace is important, and in virtual world, you don't have to worry about crashing and losing your life When I run the HKS CZ200S, I intentionally drive in the pace like I would IRL, pushing it, but not all for nothing pace ( like aliens would do, and they can always restart when there's one little mistake of a tenth in a section of the lap ).

 

[Jimmy Howlett]

Pace is important, and in virtual world, you don't have to worry about crashing and losing your life When I run the HKS CZ200S, I intentionally drive in the pace like I would IRL, pushing it, but not all for nothing pace ( like aliens would do, and they can always restart when there's one little mistake of a tenth in a section of the lap ).

Also In real life you don't have a blue dotted racing line to follow that tells you when and where to brake, coast and get on throttle with an indicator telling you what gear to use for the upcoming corner that also flashes when you need to get on the brakes, or a Ghost to chase down and a magic ABS system stabilizing the car with countless hours to run laps to chase down a lap time on a track with perfect conditions and the exact same conditions every time you get on the track with a car that never fails no matter how hard you push it for unlimited hours..

 

[Brend]

This is amazing stuff, thank you for posting this example.

 

[YZF]

Wow. Those test times that @VZF posted are so depressing. I've always thought that cars seemed to be too fast in GT6, but I never realized just how unrealistic GT6 truly is. It's very interesting that the Lamborghini Murcielago LP 670-4 SV and Mclaren MP4-12C were actually slower 0-100kph than the real car. But the faster you go, the less realistic things become.

What confuses me is how there isn't an easy to spot pattern to this. The Mercedes SLS AMG ends up hitting 300kph 14.2 seconds sooner than it would in real life, but the Lamborghini is only 5.4 seconds quicker to 300kph than its real life time. Then we come to the Lexus LFA which hits 300kph 24.3 seconds quicker than it could in real life. So, the LFA is reaching 300kph in HALF the time it actually takes the real car to reach that speed. It's interesting that the SLS AMG, Murcie SV, ZR1 and LFA all hit 300kph within 4.2 seconds of each other in the game, but in real life the cars have an extremely wide disparity of times to reach 300kph. In fact, the difference between the four cars is actually 20.9 seconds!

I noticed this unrealistic car performance back in december, when i tried to test top speed of some high powered cars and instead of 320+ they were doing 360+ ... wtf?? And it is really strange why PD can't fix it. Do they know about it, but it's hard (I can understand and accept that perfectly) or they didn't notice it yet or do not take it seriously?

Anyway, back to tests: I wouldn't take 0-100 measurement very seriously as I tested each car only once and used Sports Hard tyres and it's hard to launch such powerful cars ideally every time. So if there's difference of some tenths here and there - I would say this is totaly acceptable. And in fact 0-200 most of the time is very close to real life, which also means that overall physics calculation is quite accurate except air resistance calculation.

Regarding the time differences between SLS/LFA and Lambo, I would say this is all because of the way air resistance impacts acceleration. The more powerful car you have, the less it is being slowed down by air and vice versa. It's like driving uphill - if you have strong engine, the difference between flat acceleration and acceleration up the hill will be minimal. But if you have weak engine, going up the hill will be much slower than accelerating on flat surface.

Same with aerodynamics: lambo and zr1 and especially veyron have more powerful engines so in real life they aren't slowed down that much, while SLS and especially LFA 'hit the wall' as M. Brundle likes to say: weaker engines have trouble fighting with this air force. That's in real life.

Now if you don't have air resistance, then you don't 'hit the wall' because there is no wall (or at least not that much), so the difference between engine power doesn't affect times that much. This is what we have in GT6. The code calculation is incorrect - too low. I'd say some number, responsible for 'air resistance' calculation, should have higher value.

Also I would guess that air drag coef. of the car body isn't calculated very precisely, as such Lambo and LFA might have very similar calculation formulas in the game but these cars may have quite different drag coefficients in real life. This adds too.

 

[Rotorist]

Tsukuba is ok in speed same as in RL just because most cars do not pass over 200 kph there. Aero drag problem in GT6 begins just before 200kph. It certainly is a fault in math formula game uses. They should have fixed it it long ago and the most weird with this bug is that it didn't exist in GT5...

 

[YZF]

I don't take this data at face value. Yes, there is obviously a problem, but PD are trying to build an accurate simulation... what would you prefer.. a system that builds up a top speed and acceleration curve using variables and calculations that apply to real world physics (and maybe doesn't work at extremes).. or a system that just took a real world max and scaled it back. One is inaccurate and is trying to be accurate, one is inaccurate and is posing as accurate.

As much as I love whole GT franchise, if I see evident inaccuracy, I accept it. I don't think there is point in looking for excuses, 'how they are doing', etc. We don't know how they are doing it, and frankly, we don't care. What we care about is the final result.

You could say that 'it is immposible' to calculate correctly (although other developers were doing this years ago for much more simple racing games and they managed to make it very accurately), but the fact that GT5 has implemented acceleration and top speed accuractely, says that this is just a coding flaw, bug, issue, whatever. Nothing more.

I would put this all down to the fact that Sony wanted (and pushed) PD to release GT6 before last Christmas and they didn't want to wait another year. Especially with PS4 coming out, etc. So PD released the game, but it was made in a hurry and the game wasn't finished. And it's still not yet.

 

[YZF]

GTR Nismo Factory Stock 591hp Stock SS Tires

In Game

3.069 0-60mph & 11.002 1/4mile @ 124.7mph

Real Car

2.900 0-60mph & 11.100 1/4mile @ 125.3mph

View attachment 259819

Slower off the line, so I dont have too much grip. Ive heard the Nismo GTR will do 2.6 0-60. Its not uncommon for cars to hit numbers faster that what the factory says with good drivers & conditions, I don't take factory figures as set in stone.

Jimmy, take a closer look at my tests and conclusions I wrote. The difference between real life and GT6 is really noticable above 125 mph. Up to 125 mph (200 km/h) everything is more or less OK. It's when you go for higher speeds and top speed - thats' were the issue is.

 

[Jimmy Howlett]

Jimmy, take a closer look at my tests and conclusions I wrote. The difference between real life and GT6 is really noticable above 125 mph. Up to 125 mph (200 km/h) everything is more or less OK. It's when you go for higher speeds and top speed - thats' were the issue is.

Thats nothing new, my post is not directed at you it is showing the 1/4mile run details, I also posted videos of laps in different cars at different tracks even Nurb where the GT6 & Real lap are very tight. I get that you have been tripping about the top speed issue, & I agree its an issue, but to me its not a big deal when I strip all driving assist like driving line and hud with gear shift indicator even the ghost & my laps are so close to real life laps. I dont care if others can beat real life laps by 5 or more seconds when they follow a driving line, use a gear shift and brake point indicator and not to forget ABS braking assist while chasing a ghost over and over. Lets be real, not many drive in GT6 in a Real way.

Look at all the data

Even you must admit this data is quite promising even given they are having hiccups implementing computational fluid dynamics, the fact they are implementing CFD at all is also quite promising and the only logical step in the direction of an authentic driving simulator.

 

[azidahaka]

Thats nothing new, my post is not directed at you it is showing the 1/4mile run details, I also posted videos of laps in different cars at different tracks even Nurb where the GT6 & Real lap are very tight. I get that you have been tripping about the top speed issue, & I agree its an issue, but to me its not a big deal when I strip all driving assist like driving line and hud with gear shift indicator even the ghost & my laps are so close to real life laps. I dont care if others can beat real life laps by 5 or more seconds when they follow a driving line, use a gear shift and brake point indicator and not to forget ABS braking assist while chasing a ghost over and over. Lets be real, not many drive in GT6 in a Real way.

Look at all the data

View attachment 259888

Even you must admit this data is quite promising even given they are having hiccups implementing computational fluid dynamics, the fact they are implementing CFD at all is also quite promising and the only logical step in the direction of an authentic driving simulator.

I would like to have a Tutorial on this Motec data analysis... You seem knowledgeable enough lol

 

[YZF]

Im facepalming so hard right now. I had not considered that SSRX might have no high speed drag but this not being carried over to all other tracks. Ive always done my top speed testing at SSRX because its got so much room to play.

This opens the doors to interesting things, SSRX vs other tracks and Original tracks vs Real Tracks

Shall we test other tracks then?

Mercedes SLS AMG
0-280 km/h: 23.8 sec Fuji
0-280 km/h: 23.1 sec SSRX
0-280 km/h: 24.5 sec Monza


Lexus LFA
0-280 km/h: 22.8 sec Fuji
0-280 km/h: 22.1 sec SSRX
0-280 km/h: 23.7 sec Monza

At Fuji, top speed at the straight is 280km/h. Also at Monza, I had to start in a corner and make some steering corrections (I am testing with DS3). This could affect by a second or so.


Against real life and other tracks too:

Mercedes SLS AMG
0-300 km/h: 30.2 sec Monza (steering corrections)
0-300 km/h: 28.0 sec SSRX
0-300 km/h: 42.2 sec Real Life


Lexus LFA
0-300 km/h: 28.6 sec Monza (steering corrections)
0-300 km/h: 26.3 sec SSRX
0-300 km/h: 50.6 sec Real Life

Basically it's not because of a specific track. It's everywhere.

 

[NandoRock]

Shall we test other tracks then?

Mercedes SLS AMG
0-280 km/h: 23.8 sec Fuji
0-280 km/h: 23.1 sec SSRX
0-280 km/h: 24.5 sec Monza


Lexus LFA
0-280 km/h: 22.8 sec Fuji
0-280 km/h: 22.1 sec SSRX
0-280 km/h: 23.7 sec Monza

At Fuji, top speed at the straight is 280km/h. Also at Monza, I had to start in a corner and make some steering corrections (I am testing with DS3). This could affect by a second or so.


Against real life and other tracks too:

Mercedes SLS AMG
0-300 km/h: 30.2 sec Monza (steering corrections)
0-300 km/h: 28.0 sec SSRX
0-300 km/h: 42.2 sec Real Life


Lexus LFA
0-300 km/h: 28.6 sec Monza (steering corrections)
0-300 km/h: 26.3 sec SSRX
0-300 km/h: 50.6 sec Real Life

Basically it's not because of a specific track. It's everywhere.

The difference is much big to be negligible. Did you use automatic transmission? TC or other aids?

 

[Jimmy Howlett]

Shall we test other tracks then?

Are you using motec i2 to analyze the data?

& what is your source for your top speed data from the real cars?

 

[NandoRock]

Suzuki MR Wagon Sport '04 stock at Circuito de La Sierra. Speed rainbow track.
I'm testing some tracks at Motec. Suzuka, for example, has not put correct times. It counts 2 laps when it should count 1. This error is both to the old Suzuka and the new one.