GT4 vs REAL - An investigation of times

Dave_George

Dave_George

812
United Kingdom
Leicestershire
davegeorge78
Hi everyone,

I have recently been working on an article which is the result of a question I have had for some time - Just how close to real life is GT4?

I wanted to know if I could put a time to it, to see how close actual lap times could get.

I have completed 2 tests at 2 circuits involving 20 cars in total, The Nurburgring Nordschleife and Tsukuba are the venues as there is so much real life lap time information available at both circuits for a massive variety of machines that are also available in gt4.

So I have created an article and also a small website as a means of hosting it, The file is in pdf format and you can link to the relevant page HERE

The file is GT4 vs REAL.pdf, left click to view only or right click and "save link as" to download it to your pc.

I hope you find the article interesting, whilst I hope it raises discussion, please understand it is not a definitive answer as to how real GT4 is, but I can assure you the laps are all genuine and it is a reflection of how real gt4 is for me, when i ran these 20 cars vs real lap times.

When time allows I will post some of my replay files to the same site under "gt4 replay files" in .max format so that anyone interested can download and view the ghost replays. I will put them here also.

Please do have a look and let me know your thoughts and comments:tup:

regards

David

*EDIT*

I have put 4 replay files below for anyone who is interested in downloading them to view, for some reason the file names have been cut off, so from top to bottom they are:

SLR @ Nurburgring N2 tyre
Civic Type R (EK) @ Tsukuba N2 tyre
Civic Typr R (EP) @ Tsukuba N2 tyre
Vanquish @ Nurburgring N2 tyre
 

Attachments

  • 41.139.max
    54.6 KB · Views: 29
  • 13.576.max
    21.3 KB · Views: 15
  • 10.839.max
    21.8 KB · Views: 17
  • 15.895.max
    54.7 KB · Views: 19
i just skimmed through the pdf real quick and the lap times you set were kinda slow...

I have to think of GT4 as much more of a game than a simulation... or else i start to analyse things and end up so dissapointed i dont want to even touch my ps2....

i didnt sleep at all so i cant bring myself to thuroughly read the pdf... later doay i will though:)

damny i cant even type anymore... i meant to say later today
 
This is the good stuff. A brilliantly made comparison which doubtless was a lot of work. Thumbs up, rep points are on the way! 👍

Regards
the Interceptor
 
Also.. I think a shorter popular track would have been a better choice for comparison laps. On shorter tracks like tsukuba, laguna seca, and on fuji... this "Fear Factor" people speak of wont have as much of an effect on lap times.

now that i think about it... it wouldnt make much of a difference what track you choose.
there are just tons of things to consider as to why lap times are off by so much... the effects of the environment (temp of track and air etc...) on the parts of the car... the many things not simulated by gt4... they all play a part in lap time differences.
 
Good work Dave 👍

I like the fact that you've mentioned the 'real world' car weights as mentioned in last months EVO - i found it really interesting that so many cars actually weighed so much more in reality then the official published manufacturer weights.
 
Well written Dave and good to see that I'm not the only one with an interest in GT4 comparisons.

Obviously a lot of work in this.

👍

Scaff
 
Hi, Thanks to all for the positive feedback so far:tup:

gabkicks
i just skimmed through the pdf real quick and the lap times you set were kinda slow...
Click to expand...

Like I said in the article, someone else might find my times slow, I do however feel they were driven with plenty of conviction, and the weight does make a difference too.

the interceptor
This is the good stuff. A brilliantly made comparison which doubtless was a lot of work. Thumbs up, rep points are on the way!

Regards
the Interceptor
Click to expand...

Thanks again, and yes it was quite a bit of work!

gabkicks
Also.. I think a shorter popular track would have been a better choice for comparison laps. On shorter tracks like tsukuba, laguna seca, and on fuji... this "Fear Factor" people speak of wont have as much of an effect on lap times.
Click to expand...

errm, Tsukuba??? (Have another read!)

TheCracker
Good work Dave

I like the fact that you've mentioned the 'real world' car weights as mentioned in last months EVO - i found it really interesting that so many cars actually weighed so much more in reality then the official published manufacturer weights.
Click to expand...

Yes, vehicle weight struck me as possibly the single most overlooked thing when making comparisons with real life, and now we know that even the manufacturers quoted figures are "optimistic" to say the least...(sometimes)

scaff
Well written Dave and good to see that I'm not the only one with an interest in GT4 comparisons.

Obviously a lot of work in this.

Scaff
Click to expand...

Thanks Scaff:tup: I think GT4 was made for comparisons! Doing this has actually given me a whole new outlook on gt4 too.

regards

David
 
I luv to push people around. I really don't see GT4 as a sim game.

Somone have owned the time of the real car by a whole 3 second on tsukuba.
https://www.gtplanet.net/forum/showthread.php?t=79303

Here it show how much you can take the corner badly & still follow the professional driver on the nurb.
http://www.youtube.com/watch?v=lYgTJysPnLE

Here EPR & PGR3 as I know have the track layout accuratly.
http://www.youtube.com/watch?v=-BbD9bTVQ6A
Download only for PGR3(PGR3 have all the Nurburgring track layout unlike GT4 & EPR :sly: )
http://files.filefront.com/Project_N_wmvwmv/;5049084;;/fileinfo.html
http://files.filefront.com/PGR3_Nuburgringwmv/;5031088;;/fileinfo.html

I still think that EPR is the most accurate with real life driving. One mistake & you don't follow.
 
Fantastic report, great work.

I particularly like the detail - factoring in the driver weight especially. Is there an extra big red tick I can click on? :)
 
I read it all, great article. It did take time, months. The variety of the cars you tested was good, from FF to FR to 4WD to MR. Congratulations on this! You must be really proud.
 
I think overall the game laptimes should be a tad quicker for a variety of reasons, the main one of which would be the lack of DEATH...in the event of a crash. Driving at 240mph on the Nordshliefe strait is a lot scarier in a car than it is in a video game, and people are more likely to push a vehicle as hard as they can when they're not in it. I tested an S2000 to the same exact specs as one in real life that set a 58.9s Tsukuba lap time, and i eventually got a 56. Although I bet the room for improvement would be even greater if it were a large high speed track though
 
i cant remember ever hearing a pro driver using fear as an excuse for not setting a fast lap time (only "normal" people and journalists etc)... If they feel that they can take a corner a certain way and go faster they're gonna try it... its just there are lots of things not simulated correctly or not simulated/modeled at all in gt4.
 
Gabkicks
i cant remember ever hearing a pro driver using fear as an excuse for not setting a fast lap time (only "normal" people and journalists etc)... If they feel that they can take a corner a certain way and go faster they're gonna try it... its just there are lots of things not simulated correctly or not simulated/modeled at all in gt4.
Click to expand...
It might not be present in the same amount as with a "normal" person, but it's there ... not even a race driver will come into the last corner of Laguna Seca at 150 ... know why? Because if he comes off the track he's going straight for the wall and he might die. You can't die on GT4 ...
 
slo-car
It might not be present in the same amount as with a "normal" person, but it's there ... not even a race driver will come into the last corner of Laguna Seca at 150 ... know why? Because if he comes off the track he's going straight for the wall and he might die. You can't die on GT4 ...
Click to expand...

Nah, The only thing is accurate in GT4 is the speed of the car. Even if you mess the corner, you keep the speed in GT4. In real life, you mess the turn, you mess the whole lap.
 
its just i find it funny that fear is brought up as more of an explanation for lap times being off rather than physics/suspension differences in gt4 when compared to real life.

a real life racer doesnt let fear hold him back...
 
Excellent work, Dave_George! I can tell you put a lot of effort into this. +Rep! :)

I'm impressed that the times matched up so well, but unfortunately, laptimes don't tell the whole story...not only can laptimes be skewed by an inaccurate track model, they also don't represent how a car handles and behaves during cornering, and can only tell us how fast a car corners. :indiff:
 
True enough... but they are indicative of... something. :lol:

As for FF cars and laptimes, I'd like to point you to some work I did last year but never put into my sig... too burned out... :indiff:
Sport Compacts on the Ring
(this is the last of three articles, covering nearly 30 cars... sheer torture... :lol: )

Of course, the laptimes are indicative of the zillions of hours spent doing all that testing, the experience with FFs, and the complete disregard I have of the "fear factor". I have no doubt I'd get similar times to yours if I drove in a way I'd consider sane in real-life. :lol:

And @Gabkicks: If you don't like the word "fear factor", common sense might be a better term. I've driven at the edge in real-life, and it's definitely nowhere near how fast I can drive in GT4. Even taking into account inaccuracies in the game, it still plays a big part in how a person drives.

@Dave: I tend to agree with you, and I've personally seen that FR cars are as quick or much quicker in real life than FF cars, both in pure acceleration tests and in track tests (though track tests include other factors).

Others, like Wolfe, have postulated that the "realistic" understeer (meaning some people like Scaff (and me!) think it is, but others, like Wolfe, don't) is the result of PD adding understeer into the mix as an afterthought. Basically, it's a patch.

What this means for me, personally, is that cars with an understeer bias enjoy (if that's the right word) a multiplier effect from understeer, while cars that are more neutral (or oversteery) are not affected as much.

This idea is supported by the fact that B-Spec drivers (yours is fairly typical, even "perfect" B-Spec drivers (max points) drive much the same way) can drive the balls off of a Ruf, an Alpine or any other RR car, but simply can't seem to drive an FF car without plowing off the course if you increase the speed even a tiny bit. They seem to be programmed for a different physics model than what GT4 currently uses.

From personal experience on and off track, and from local touring car races, I've catalogued instances where an FF should break loose in the rear and spin, and that doesn't seem to happen in GT4 as often as you'd expect... and in some cases, not at all.

Add to this the fact that to GT4, weight=traction, then you start to see where the more powerful (and usually heavier) cars have an advantage.

The gross weight vs. manufacturer's weight is a very valid and irritating complaint. I've found that ballasting my car to match what I'd expect the actual weight to be has a drastic effect on handling and laptimes... as much as 1-2 seconds per lap on a race course. Take note, though, that fuel apparently has some weight, as different fuel levels have an incremental effect on tire wear, but not as much effect on overall speed as you might expect. If GT4 has actually implemented "wet" weight (+oil,+water,+driver), there is no way for us to find out.

Great study... great article! 👍

And +rep from me for Dave, too. :dopey:
 
Gabkicks
a real life racer doesnt let fear hold him back...
Click to expand...

That's true to a degree, however I have seen a few interviews after a professional driver has just set a new lap record or over-cooked it into a corner, and one thing they are asked, is "How did it feel?" and the response in quite a few cases... "I was scared sh**less, I won't be trying THAT again for a while" or words to that effect.

Both sides are valid points, sure, pro race drivers have a high confidence in their safety equipment in the event of things not going right (and rightly so), however, they are still not going to push the limits of their car as readily as someone playing a game, in which screwing up a corner, or totalling your car has no real life effect. This is one reason why pro drivers will set faster times on the simulator (and yes, most Aussie V8 teams have a simulator set up with V8 supercars 3) than on the real track, their are no real consequences.

Well thats my $0.02
 
Gabkicks
its just i find it funny that fear is brought up as more of an explanation for lap times being off rather than physics/suspension differences in gt4 when compared to real life.

a real life racer doesnt let fear hold him back...
Click to expand...

I mentioned the "Fear factor" as it is quite often quoted on here as being one of the principle reasons for gt4 being faster than real life.

I respect that factor to a certain degree and also acknowledge that there are differences that gt4's physics system cannot replicate.

If you look back at the article and specifically at the NSX-R lap you will see that to match the quoted real weight I had to add in nearly 200kg of ballast and still ended up 3 seconds behind the real time, with some more laps I could probably equal this but Im not sure I could go much quicker.

regards

David
 
David - You are right there, many factors come into it, and programming limitations do play a big part. However those same limitations can have the opposite effect, in that there may be some positive handling characteristics that connot be reproduced digitally at this stage of software development. (Why do I always wind up back at coding and it's limitations) Just remember that GT is, after all by definition, a GAME, and as such will never be able to 100% replicate the real life challenges and behaviour of a real car, be it a race car or a road car, although it gets damn close at times!
 
Gabkicks
its just i find it funny that fear is brought up as more of an explanation for lap times being off rather than physics/suspension differences in gt4 when compared to real life.

a real life racer doesnt let fear hold him back...
Click to expand...

From my reading of the piece in question and my own observations on the point I don't think anyone has claimed that GT4 is 100% accurate in regards to the physics and suspension.

I would however have to disagree in regard to the fear point, a real life race will use 'fear' in a different way to a lot of people, but it is still a limiting factor.

A number of drivers have even spoken about the differing ways that fear affects you as a driver, a good example of this is the book "Working the wheel" by Martin Brundle, which discusses it quite a bit. Martin Brundle talks about how its a vital tool at times to let you know how far you can go in a given situation, such as spinning in the rain and not knowing which way you are facing when you start off again, fear raisesd its head here without a doubt (and he says as much), but he carried on more aware of the potential risks of the situation.

Other examples would be Lauda's choice to not race in Japan in 1976, when he retired during the second lap, saying that the world championship was not worth dieing for.

Or Fangio's win at the Nurburgring in 1957 in which he repeatedly smashed the lap record after loosing the lead, afterwards he said

"I believe that on that day in 1957 I finally managed to master the Nürburgring, making those leaps in the dark on those curves where I had never before had the courage to push things so far."

and

'I did things I've never done before, and I don't ever want to drive like that again.'


So it would certainly appear (from their own words) that drivers do let fear hold them back at times, not just fear of injury or death, but also fear of losing the lead or a place. In most racing sims a quick trip across the grass is a minor distraction, in reality its a major issue that could well cost you a place or a win.

Regards

Scaff
 
Hi

wolfe2x7
Excellent work, Dave_George! I can tell you put a lot of effort into this. +Rep!

I'm impressed that the times matched up so well, but unfortunately, laptimes don't tell the whole story...not only can laptimes be skewed by an inaccurate track model, they also don't represent how a car handles and behaves during cornering, and can only tell us how fast a car corners.
Click to expand...

Thanks very much:tup:

Concerning how the car handles and behaves during cornering, your right, lap-times dont inform us of how a car behaves, but Like i mentioned (perhaps not enough) in the article, every car clearly had its own personality and "feel" on circuit, each car had different cornering traits and abilitys too.

gooseter
David - You are right there, many factors come into it, and programming limitations do play a big part. However those same limitations can have the opposite effect, in that there may be some positive handling characteristics that connot be reproduced digitally at this stage of software development. (Why do I always wind up back at coding and it's limitations) Just remember that GT is, after all by definition, a GAME, and as such will never be able to 100% replicate the real life challenges and behaviour of a real car, be it a race car or a road car, although it gets damn close at times!
Click to expand...

Yep, I realise gt4 is never going to be 100% and that there are many things that cannot be reproduced in the digital world. However, when doing a comparison like this, all you can do is take the variables that gt4 allows you to change, weight, power, correct car variant etc and match them up as closely as possible.

I still believe that based on my results gt4 makes a damn good case for itself. Its a shame there isnt a N "2.5" tyre available, and I would love to test some more with some very very accurate real life weight data available.

Once again thanks for all the feedback so far wether good bad or indifferent:)

regards

David
 
don't get me wrong david, I'm seriously imressed with your efforts, I just think too technically sometimes. And yes, I missed the point of the thread for a while there! My bad!

Kudos (I give you a positive red tick, but I'm still under 100 post requirement)
 
gooseter
don't get me wrong david, I'm seriously imressed with your efforts, I just think too technically sometimes. And yes, I missed the point of the thread for a while there! My bad!

Kudos (I give you a positive red tick, but I'm still under 100 post requirement)
Click to expand...

Cheers gooseter, I completely understood where you where coming from by the way.

regards

David
 
no worries, I would love to get the chance to test a few of these cars around the real tracks one day (not likely, but I can hope). I have test driven HQ's (a class of racing here in OZ, which uses only HQ Holdens), and go-karts (real race ones, not just your local track ones), but never a high end race car.
 

Latest Posts

Back