Gran Turismo – A comparison across the series (part 2 added)

  • Thread starter Scaff
  • 97 comments
  • 12,340 views

Scaff

Moderator
29,029
United Kingdom
He/Him
ScaffUK
Right, its not very often I start a new thread, but as normal when I do is quite a size, so settle down and have a read.

This time I have set my mind to taking a couple of cars (one FF and one FR) that feature in each Gran Turismo game in the series. Using a series of standard measurable tests, along with my own opinion (which is just that – an opinion) looking at what has changed.

I would also like to point out that I am only looking at the GT series here, comparisons between GT and other games such as Enthusia and Forza have dedicated threads, so let’s keep this one for GT only.

So the two cars that have fallen victim are:

FF – Honda Integra Type R (DC2)
FR – TVR Griffith 500

These two will be tested using both the standard tyres they are supplied with when bought and whatever type of ‘real’ tyre is available in each version of the game. The obvious exception is the original GT, which had no ‘real’ tyre option; additionally GT4 has three ‘real’ tyres on offer and I have taken N2’s for this comparison.

To keep things fair across all versions of the game I have stuck to using a DS2 and no driver aids at all, all games were run on a PS2, but no ‘enhancement options were used for GT and GT2 (just in case). All the tests were carried out in the GT mode of the relevant game, again to ensure the tests are easily comparable.

The tests are as follows:

0 – 100 – 0 MPH test
A good test of off the line traction and acceleration combined with braking ability and stability. The best overall time of three runs is taken.

Grand Valley lap time
The best time of three laps around the Grand Valley Speedway, I chose this track as it does appear in all four games in the series and has had only minor changes over the series. It also has a good length at 3.1 miles and features a range of corners from hairpins to sweepers with a couple of nice technical sections.

I have attached all the figures from the two tests above in a .pdf file at the bottom of this post to allow easy comparisons between the games, as the sections below will cover each game in turn.

Donuts & Handbrake turns
I was initially unsure about including these two tests, but as I’m sure someone would have asked why low speed manoeuvres were missing I have included them. The donut test will only apply to the TVR (for obvious reasons) and the handbrake turns tested at a range of speeds between 30 & 60 mph.

In addition to the above I will also be commenting on my opinion on the handling characteristics of each game, I will also (were required) compare this to my real world driving experience, which consist of many years in the motor industry, over 8 of which cover automotive training.



Gran Turismo
Well this is where it all began, as mentioned above no ‘real’ tyres were available in the original GT, so these tests are all carried out using the Normal tyres supplied with the cars.


0-100-0 Test
Integra Type R times were as follows:
0-100 15.28
100-0 3.64
Total 18.92

TVR Griffith 500 times were:
0-100 8.2
100-0 4.2
Total 12.4

Neither car had any problem leaving the line; no wheel spin even with full throttle launches. No drama no fuss, just hammer down the track till 100mph (161kmh) is reached and then stomp on the brakes.

It’s also worth noting that the times achieved are quite silly when compared to reality, the Integra time of 18.92 is close to that of a ’05 Mustang and the Griffith with a 12.4 would beat a Ford GT by half a second!!!


Grand Valley
Integra Type R 2.16:284

TVR Griffith 500 2.00:311

This test really brought me down to earth with a bang, the rose-tinted glasses were removed and the arcade bias of the handling was revealed. Gran Turismo was ground breaking at launch, nothing else like it was around and in comparison to its contemporaries it was a ‘sim’, but time has not been kind to the handling. Very little wheel spin or understeer is present and oversteer can be initiated at will on either car. The lap times clearly show the forgiving nature of the handling.


Donuts & Handbrake turns
Donuts first, and the TVR managed these with ease, just turn the wheel and plant the throttle, keep this up until you want to stop. It may look pretty, but its just too easy, in reality donuts do require good throttle control and balance.

Handbrake turns were carried out in both cars with ease, speeds from 30 through to 60 caused no problem at all.



Gran Turismo 2
Two years down the line, a lot more cars and tracks; yet many people say that GT2 is inferior to the original GT. I have to say viewing them side by side, the originals graphics are slightly sharper (but not by much) and the GT2 replays look unfinished in places (Tahiti Road and the floating cars is a good example). Additionally, to this day people discuss and debate if GT2 was rushed out and PD missed out chunks of the game.

However very few people comment on how (and if) the manner in which the cars handle had changed, so lets look at the numbers.


0-100-0 Test
Integra Type R times were as follows (Control/Normal):
0-100 20.127 / 19.487
100-0 5.801 / 4.439
Total 25.928 / 23.926

TVR Griffith 500 times were (Control/Normal):
0-100 12.095 / 11.003
100-0 5.431 / 4.727
Total 17.526 / 15.730

The Integra had little problem leaving the line with either tyre type, with little wheel spin present. The Griffith had some wheel spin present with the normal tyre and slightly more on the control tyre. However both were minimal and easily controlled. The times are now the opposite of GT, with both sets of times over that you would expect to see in reality, with strangely the Normal tyres being the closest.



Grand Valley
Integra Type R 2.31:593 (Control) / 2.20:334 (Normal)

TVR Griffith 500 2.22:873 (Control) / 2.09:798 (Normal)

The difference between GT and GT2 became very apparent on the first lap with the Integra on control tyres, for the first time I had to actively think about the workload on the front tyres, particularly under braking. Understeer was much more clearly present under braking, but still minimal under acceleration. Switching to Normal tyres reduced the understeer considerably and when it was present it was easily corrected with the throttle. Neither tyre type would allow the fronts to spin under full throttle when exiting a corner.

The Griffith also demonstrated understeer under braking with the control tyres and a similar reduction when the normal tyres were used. The control tyres also provided lots of power oversteer when exiting corners, this was difficult to control with the digital throttle on GT2 only allowing two settings (closed and wide open), much more manageable with the normal tyres.


Donuts & Handbrake turns
The Griffith does allow you to perform donuts with ease, a slight on/off motion with the throttle was required to balance it, however the front tyres did push the car out of line.

Handbrake turns are possible, they are not neat and tidy and require a lot of room, but can be done at a range of speeds in either car.



Gran Turismo 3 A-spec
Another 2+ years and a new platform; a lot less cars and a lot better looking. Some people rate the first two games over GT3, so let’s see what the numbers tell us.


0-100-0 Test
Integra Type R times were as follows (Sim/Normal)
0-100 21.30 / 20.14
100-0 4.02 / 3.24
Total 25.32 / 23.38

TVR Griffith 500 times were (Sim/Normal):
0-100 12.92 / 11.54
100-0 4.24 / 3.70
Total 17.16 / 15.24

First I have to admit that using the DS2 here has caused me some major issues, I normally do not touch the damn thing for GT3 (or GT4 for that matter) and fine throttle control is needed to launch cleanly for the first time in the GT series. This is particularly true of the Integra, try a full throttle launch and with either tyre the car will bog down, losing a lot of time.

The same is to a degree true of the Griffith, however with the weight transfer under acceleration it is easier to modulate using the control; the greater grip can be felt.

The times themselves are slow compared with reality, however the two elements can be looked at separately as the 0-100 times are slow compared to the real world models and the 100-0 times unrealistically low (particularly for the Normal tyres).


Grand Valley
Integra Type R 2.34:905 (Sim) / 2.22:859 (Normal)

TVR Griffith 500 2.28:104 (Sim) / 2.10:498 (Normal)

The Integra on sim tyres for the first time in this series of tests actually started to feel like driving a FWD car on a track. Understeer was very present in both braking and acceleration if you overwork the front tyres, however it was very difficult, even on the sim tyres, to break traction and spin the fronts out of low gear corners such as the hairpins. Switching to the Normal tyres lowered the level of understeer quite dramatically, it was difficult to produce unless you got very silly with the cornering speeds, even then it was quite straightforward to control with the throttle.

The Griffith on sim tyres was a handful, even with analogue throttle on the DS2, power oversteer was immediate out of the slower corners. It was easy to catch, but unlike a car biased for the road no understeer was present before the transfer to oversteer. Switching to Normal tyres reduced oversteer to the level were it was only really present at full throttle exits of low gear corners. Understeer on corner entry and under braking was represented in a very similar manner to the Integra.


Donuts & Handbrake turns
The TVR allowed reasonable Donuts to be performed as long as good throttle control was maintained; this is far more difficult with the DS2 than a wheel & pedals.

Both cars allowed handbrake turns to be carried out with ease at speeds of approx. 30mph, however at 50-60mph you need to tap the brakes to transfer weight forward before using the steering and handbrake. Unless you do this to get the weight over the front wheels the turn will not work. Not realistic, but it did get the turns done.



Gran Turismo 4
The latest and biggest game in the series; with more cars and tracks than ever before. However, how does it look in terms of the numbers.


0-100-0 Test
Integra Type R times were as follows (N2/S2)
0-100 19.02 / 18.46
100-0 4.64 / 4.02
Total 23.66 / 22.46

TVR Griffith 500 times were (N2/S2):
0-100 11.42 / 10.34
100-0 5.06 / 4.38
Total 16.50 / 14.72


The times posted in GT4 are quicker overall than the GT3 times and closer to the times from GT2, however the make up of these times is very different from the two previous games in the series. With both cars the 0-100 times are closer to the real cars (Integra 17secs and Griffith 10.2secs – Autocar) and I’m sure could be matched with better launches, Additionally the 100-0 are much more realistic; while I do not have real world figures for the two cars above; but most modern cars will brake in the 4 – 5 second range from 100mph.


Grand Valley
Integra Type R 2.37:043 (N2) / 2.26:564 (S2)

TVR Griffith 500 2.29:790 (N2) / 2.16:210 (S2)

Now the Integra on N2 tyres really forced me to think about what I was doing and how much work the tyres were being asked to do. Attack a corner too quickly and you will be rewarded with more understeer than any of the previous games, back off the steering and/or the power and as long as you were not going to fast you can get it back; but you will lose time.

The Griffith will also understeer badly if you enter a corner too quickly, again in a manner that none of the previous games in the series has recreated. Power oversteer is also present; power smoothly out of the slower corners and the Griffith will present a little understeer that leads to quite quick oversteer.

Again this is not what we are used to from previous games in the series, in the past it has been a case of wide-open throttle from the apex and the back will step out. However do this in GT4 and the Griffith will first settle into slight understeer, before a rapid transition to oversteer.

Get the braking point and your line right however and it possible to flow from corner to corner in either car; contrary to some opinions I have read I do not believe that GT4 has got understeer wrong. Just that for the first time we are being asked to deal with it in a very direct manner. What the N2 tyres do ask of you is that to be quick, you must be smooth with all of your inputs and very careful with your line. The divide between progress and understeer is slight, but it should be when you are taking the chicane between the two Grand Valley hairpins at 90+ mph.

Changing to the S2 tyres that come with the cars as standard has a predictable effect, as the grip levels rise your need to be aware of the tyres limits drops and so do your lap times.


Donuts & Handbrake turns
And now for the bad news, as much as I tried and with all tyre compounds I could not get the Griffith to perform a donut or handbrake turn, the same with the Integra and handbrake turns.

The results were to be honest quite pathetic, with the donut the front tyres would not grip and the car just pushed its way around the track with its backside billowing smoke; very undignified.

The handbrake turns were just as bad, pick either car, any tyre and give it a go. At 30mph the car just stopped half way around, at 60mph it just slid diagonally into the barriers.

I tried for about an hour to get just one half-decent handbrake turn, and just ended up with a headache.

Not GT4 finest hour by a long shot, but it is a strange quirk. Now the GT4 series has never been the last work in driving/racing sims, anyone saying otherwise has either not played enough true sims or is a fan-boy.

However in what is in so many other areas the most realistic game in the GT series yet, this is a very poor thing to miss. Either that or the fun or ruining a perfectly good set of tyres in the pursuit of going nowhere does not have the same appeal in Japan (and that I just don’t believe).



Conclusion

So there we have my number and my opinion on the changes and evolution of the driving side of the Gran Turismo series, and in my opinion the balance of change has been for the better. The main issue for me in the latest version of the series is the low speed manoeuvres, such as handbrake turns and donuts, but as these are hardly used in races or hot laps they have only a minor impact for me.

I have deliberately left out issues on cars, layout, structure, etc. On the grounds that these are principally personal areas and what one person like another will not; for example I love the Fiat 500 (all three models) tiny little RR that can be chucked sideways with glee, others I’m sure hate it with a passion. Subjective, you see.

All I ask is that you read the above, comment and discuss, but please try and reserve your posts to the GT series, comparisons with other games have threads dedicated to them and this was not the reason I started this one.

Regards

Scaff


I have now added the second part of the comparison in this post, it takes a look at the Nissan Skyline R33 GTR V-Spec and Honda NSX across the GT series, enjoy.
 

Attachments

  • Gran Turismo Figures.pdf
    10.8 KB · Views: 84
:bowdown: :bowdown: :bowdown: :bowdown: :bowdown:

If I had a decent clapping smiley, I'd use it.


Thanks for that read. Sort of confirmed how I felt initially about GT4 and how I feel about it now. I was initially disappointed with the understeer and handling balance in GT4, but after comparing it to real-life, I realized that I was just to used to the unrealistically direct handling in GT3. For the first time, cornering behavior on FF cars, at least, is close to real life. I've just read an article on MotorTrend.com that corellates my experiences with the new Mitsu Eclipse... and here I was worried that my view on the Eclipse in-game was unrealistic, as previous reviews have said otherwise.

You're right, donuts don't really matter in racing terms, but they represent some physics quirk that people can sense, hence the complaining.

Great work, man. As ever, I'm impressed. 👍 👍 👍
 
Wow! That's some serious investigating. Really interesting.

I agree with what you said about the donuts and handbrake-turns. It sucks! Can't believe PD "missed out" on it. Of course you don't need donuts in a race, but it's fun! Doesn't bother me that much, although it is so strange to me.

I'm gonna save your pdf until GT5 comes and see what changes have been made. Hopefully I'll be able to burn a couple of tyre-sets...

Good work 👍
 
Thanks guys, this one really was a major work for me; but it was fun going back over the whole series and playing them one after the other. It does bring into focus the changes that have been made and how the series has evolved.

I was originally going to include GT:C and GT4 Prologue, but without Grand Valley and the right cars the comparison would not have been valid, additionally it would have confused the situation for members in parts of the world where these two were not released.

Regards

Scaff

BTW With this and the GT4 & Brakes thread I'm begining to feel like GTP's resident numbers geek! LOL
 
I think the biggest difference between GT1/2 and GT3/4 is the feeling of weight in the cars. In the early GTs, the cars felt like toys, but from GT3 on, they felt more solid and substantial, like they actually weighed a ton and a half.
 
Very nice work! 👍

Having never actually played GT or GT2 I can only say that your description of the differences between GT3 and GT4 seem spot-on. Wierd that the low speed manouvers seem to get worse though.. PD didn't consider them important?

A good, interesting read. :cheers:

Cheers!
 
That's a good writeup, Scaff. Where the lack of donuts causes a problem for me while racing is when I've spun and end up facing the wrong way. I'm used to spinning the car around to face the right way, now I have to do a slow circle or even perform a 3 point turn! :crazy: Now THAT has no place in racing!

I've had more luck with handbrake turns, and have had lots of fun doing J-turns, 180 degree handbrake turns and even 360 degree spins up and down the straight at Tsukuba in, among other cars, a V16T, and even the Formula GT car (an F1 car with a handbrake??). For a good wheeze, perform an emergency start in the FGT car, absolutely full throttle, then brake full force with the left foot while keeping the throttle pinned when the car hits about ten MPH. You will crawl up the road at 3MPH, front wheels locked, rear wheels spinning. A bit like what the F1 guys were doing at Goodwood this year. Getting back to handbrake turns, when going a bit slower I do need to scandinavean flick it to get the rotational momentum, and it does need to be done on a closed throttle or even very light braking to get some weight over the front wheels, but it can be done. I'll try it this weekend in a Grrrrrrrrrrriffith to see if handbrake turns are not possible in that car. My entry speed is always at least 50MPH on N2s, which I agree is too fast.

The other thing I am going to try is a donut, in a FF car like the 147GTA, going forwards. I have a sneaking suspicion that I'm in for a surprise.
 
wow--good job... spot on, nice and in-depth... makes me want to go back and see for myself how the games compare to one another... hmm--just need to hunt down a copy of GT that i traded-in for GT2 back in the day (doh)... :dunce:
 
Great read Scaff and well done for doing the kind of work I could never be arsed to do :lol:. 👍.
 
Great read thanks 👍

The times in the original GT really surprised me, although i havn't been on that game since GT2 it shows how unrealistic it was
 
Fine work, sir, I was wondering what you'd been up to (polite applause).

Interesting that lap times have got progressively longer as the handling experience has got closer to reality.

I remember feeling distinctly put out that I now had to brake in the middle of that fast chicane when I first got GT4, and I also used to call it GT4: Understeer City. Now we know why :) Thankyou.

On going back to the older games, were the changes to the esses before the first tunnel really noticeable? At the start they really threw me, but I suspect that I'm taking remarkably similar lines, now that I'm used to it. Hard to guess the effect on lap times.

Again, good stuff, and thanks!
 
Righty ho. I got into my Alfa 147 GTA, trotted off to Tsukuba, took off all the engine mods I'd wasted money on and planted it on the track, on N2 tyres, and with 280 BHP showing.

I torque steered my way up and down the pit straight, trying handbrake turns at speeds between 40 and (roughly) 70MPH. The Alfa failed miserably, only managing a half spin before the front tyres started sliding as much as the still locked up rears were. I then parked the car in the middle of the straight, put the steering on full lock, and buried the throttle. What I got wasn't quite a donut, but was a damn sight closer to a donut than I had managed in my 500bhp Corvette. With the front wheels spinning madly, the Alfawould drag its nose around almost on the spot, and the rear would lose grip and slide out wide, helping the FF car around half a revolution. So my suspicions were half proven: The FF Alfa does a better donut than my FR Corvette :odd:

I then hopped in one of my 6 TVR Grrrriffiths (I couldn't decide which colour I liked the most so I bought most of them) and happened to pick the one that had been treated to an oil change and chassis refresh. The griffith was also rubbish at handbrake turns, only managing a half spin, and was even worse at donuts than the Alfa... it just understeered while both rear tyres were spinning for their lives. :confused:

Knowing that I had little trouble with handbrake turns in the Cizeta, I moved on to a well balanced MR car - The Esprit V8 SE (no chassis refresh on my second hand example). It pulled off a semi decent handbrake turn at any speed that it was capable of achieving, and would even lose lateral grip at the rear instead of the front when attempting a donut! I managed a series of sort of half donuts, where the tail scoots around 180 degrees and I then lose momentum, which is as good as I've generally managed in most car sims I have driven. Finally, I tried a powerful MR car with an oversteer bias - my Protomotors Spirra. I plugged in the supercharger, chip and exhaust to pop it up to around 500bhp, put the N2 tyres on, and managed a performance similar to the Lotus's. :boggled:

Watching the replays was illuminating. When attempting handbrake turns, all the cars would show more loss of grip from the front tyres than I would expect. While the rear would slide past the front on the MR cars, the front still seemed to slide further than I would expect. I have experienced a similar thing in a real car, when attempting a 90 degree handbrake assisted turn on a road that had less grip than I anticipated. Instead of the rear neatly swinging around and lining me up nicely with the exit allowing me to drive out of the slide, I ended up in a four wheel drift which had me sliding wide of my anticipated apex. Since I never had this problem trying the same trick on slippery gravel, I put it down to my speed and steering input being mismatched to the conditions and asking for more grip than the front tyres had available. This is kind of what seems to happen in GT4, except if attempt the turn at a slower speed, you don't have enough momentum for a complete 180 degree rotation. In the replays it just looks wrong, as though the front tyres are losing grip too quickly and the momentum of the swinging tail does not have enough effect. It's as though halfway into the handbrake turn you suddenly find yourself dealing with the physics you would get from entering a corner too quickly in that particular car. GT3 was far easier for handbrake turns.
 
Great thread and great read Scaff,

I think as the GT series goes on your findings will be similar, maybe one day we will have true to life laps round the Nurburgring when realistic damage and perspective is increased,

But i think your post has shown that for now, what we have got really aint that bad at all......
 
FastEddie12
Fine work, sir, I was wondering what you'd been up to (polite applause).

Interesting that lap times have got progressively longer as the handling experience has got closer to reality.

I remember feeling distinctly put out that I now had to brake in the middle of that fast chicane when I first got GT4, and I also used to call it GT4: Understeer City. Now we know why :) Thankyou.

On going back to the older games, were the changes to the esses before the first tunnel really noticeable? At the start they really threw me, but I suspect that I'm taking remarkably similar lines, now that I'm used to it. Hard to guess the effect on lap times.

Again, good stuff, and thanks!

Always busy and up to something, thats me.

In regards to the changes to the track, they are actually quite slight at Grand Valley and the line I took was quite similar between the games.

The change in lap times in one if the most noticable things, looking at the Griffith it ranges from 2.00:311 to 2.16:610 with 'Normal' tyres to 2.22:873 to 2.29:79 with the 'sim' tyres. Changes of 16.299 and 6.917 respectively, although the original GT's lack of 'sim tyres does throw that and if you discount GT the range is 6.812 and 6.917, which equate to similar levels of change.

Its also interesting to note that while the 0-100-0 has changed up and down, with GT4 being the closest to reality in this (and if your interested you can find more of these over in the GT4 & Brakes thread), the lap times have increased and the physics model improved.


Alfaholic
Righty ho. I got into my Alfa 147 GTA, trotted off to Tsukuba, took off all the engine mods I'd wasted money on and planted it on the track, on N2 tyres, and with 280 BHP showing.

BIG & INTERESTING POST FROM ABOVE THAT I WILL NOT QUOTE ALL OF

I quite agree with you, it just looks crap on the replays. Since I finished the above I have been thinking about running the above tests with an MR car as well (should have done that from the start), I have some time on my hands next week so I may start that as well.

Regards

Scaff
 
Oh yeah I'd just like to add that though my lengthy ramble may indicate that I think the physics of GT4 are completely up the spout, let me state here that I do overall find the physics engine to have a pleasantly complex and depthy feel to it, as though its always processing a lot of data and giving a solution well tailored to all influences at the time. I feel it's overall GT4's strongest point, better than GT3 (in my opinion) and I already thought GT3 was very good. I find the cars in GT4 have a lovely complex, fluid, organic feel to them. It's one of the very best physics engines I have experienced in a driving sim. It's just rubbish at doing donuts. :P
 
Scaff
I have been thinking about running the above tests with an MR car as well (should have done that from the start), I have some time on my hands next week so I may start that as well.

Regards

Scaff


How about an NSX vs a Scooby? I'm sure they appear in all games. If a job's not worth doing well, it's not worth doing. Something like that.
 
Alfaholic
It's one of the very best physics engines I have experienced in a driving sim. It's just rubbish at doing donuts. :P

My entire first post in two lines, Kudos Alfaholic and thats just got added to my sig.


FastEddie12
How about an NSX vs a Scooby? I'm sure they appear in all games. If a job's not worth doing well, it's not worth doing. Something like that.

OK, I will ID the exact cars and put this one together, you just need to give me a bit of time, as I'm sure you can imagine, its not a quick thing to do.

Regards

Scaff
 
Thanks Scaff. It's just as well there are people like you around to help answer these questions, cos we've all asked them.

We'd do it ourselves if we could find the motivation.... :)
 
Yeah it feels kinda weird for me too playing GT4.

At 1st it feels great but after I played enthusia, it just doesn't feel right. At least in low speeds and in areas like oversteer.

While enthusia has some kind of guardian angel assisting in saving slides at high speeds.
 
@Alfaholic: FF cars do donuts just as I expect them to... dial in full lock on an FF from a standing start and the car will get partway through before settling into understeer... sort of what I can get with an FF in real life... the car will swing around, but not that much... FF cars do fine donuts in reverse, though, JUST LIKE in real life. Only tried it a couple of times, but it seems right... you ought to give it a shot. I think the problem with donuts is that to get realistic understeer, they decreased the coefficient of friction for the front tires just a wee bit too much.

@Scaff: Good luck with the NSX, be waiting to read about that. :D
 
Great read, I've noticed significant changes to the GT series. The pulling donuts with FR cars is extremely hard, next to impossible. First it starts off right, your car slowly starts the donut, but then you notice the front end moves then the donut just breaks, and you will just whip 180º and lose the entire donut. All control was lost, kind of annoying. I wonder about reverse gear in this game, it's horrendously slow acting to the point where i loathe reverse gear. Hopefully PD gives realistic gearing doubt we'll see a need for Park, but it should have the usual suspects N, 1-5,(6,7) and R, no more one button reverse. I think I'm really liking the progression, the N tires are the way to go, as you have to be more or less perfect when entering steering inputs and braking and gas inputs as well. Hopefully the next game really focuses on tire physics, and realistic tire wear.
 
niky
@Alfaholic: FF cars do donuts just as I expect them to... dial in full lock on an FF from a standing start and the car will get partway through before settling into understeer... sort of what I can get with an FF in real life... the car will swing around, but not that much... FF cars do fine donuts in reverse, though, JUST LIKE in real life. Only tried it a couple of times, but it seems right... you ought to give it a shot. I think the problem with donuts is that to get realistic understeer, they decreased the coefficient of friction for the front tires just a wee bit too much.


's odd.. what I was experiencing with the Alfa was OVERsteer! Not enough to complete a full circle but the back end was certainly losing grip and swinging around. I did try it in reverse as you suggested. Flicks around quite violently, lots of fun :D . I also enjoy doing j-turns... reverse up to a respectable speed, flick the steering to one side, and floor the throttle to accelerate up the road facing the right way. GT4 makes that stunt very easy indeed.
 
Scaff, I seem to remember that on GT2, if you had a highly tuned FF car, it would almost power oversteer, once settlled in the bend. Have you found this? or do I have an overactive imagination?
 
FastEddie12
Scaff, I seem to remember that on GT2, if you had a highly tuned FF car, it would almost power oversteer, once settlled in the bend. Have you found this? or do I have an overactive imagination?

The Integra I used in the test was stock and as such did not dothis. However when doing the tests I did have a play around with a few other cars, including some tuned FWD's and yes you can get them to kick the back out a bit under power.

For serious laughes in this area you want the original GT, in that you can get the rear of a stock Integra to kick out under power on just about any corner.

This set of tests really did bring home how far the series has come.

Regards

Scaff
 
J-turns = Fun! :lol: yup!

@Scaff: As per research and lapping in stock FF cars, I find that the "unrealistic" power-oversteer some members complain of usually happens in situations where the car is already unsettled and/or with the weight balance of the rear-end swinging around and little traction on the rear tires. So again, it's pretty realistic.

But thinking about it, what I said about the coeff of friction at the front earlier... I just realized... we often get really poor 1/4th times in FF cars in GT4 (as compared to real life), while FR cars get relatively good 1/4th times... is this correct? because it might also be another data bit pointing to GT using different physics for front and rear traction... or maybe GT just can't model front traction very well?
 
From GT to GT4:

Lots of Oversteer - Oversteer - Understeer - More Understeer

It felt like each installment just added understeer into the car. I hate understeer, so I can hardly say I was amused. But then the realism is vastly greater in GT4 and I appreciate that. I had my first true tank-slapping in an M5 in GT4-Prologue. That to me marked progress. GT3 - it was soo difficult to get a car to spin (ASM off before you ask), now the cars really do seem to act like real-life beyond the limit. GT1 and 2 were arcady, yes. Maybe that is why GT4 has the least in all of them in the arcade mode - perhaps PD finally realise that an Arcade Mode in a driving sim is a tad oxymoronic. It's a bit like putting an Arcade Mode into Microsoft Flight Simulator.
 
niky
J-turns = Fun! :lol: yup!

@Scaff: As per research and lapping in stock FF cars, I find that the "unrealistic" power-oversteer some members complain of usually happens in situations where the car is already unsettled and/or with the weight balance of the rear-end swinging around and little traction on the rear tires. So again, it's pretty realistic.

niky, I would agree with you in regard to this, in most cases that I have encountered this in FWD cars in Gt4 has been when the rear of the car is unsettled. I have had a similar thing happen to me in a number of cars when testing at proving grounds, particularly when braking hard into a corner to get the front end to 'bite', it unsettles the rear.

niky
But thinking about it, what I said about the coeff of friction at the front earlier... I just realized... we often get really poor 1/4th times in FF cars in GT4 (as compared to real life), while FR cars get relatively good 1/4th times... is this correct? because it might also be another data bit pointing to GT using different physics for front and rear traction... or maybe GT just can't model front traction very well?

Yes everything I have tested so far (and other peoples tests) seem to indicate that while realistic 0-30, 0-60, 0-100 and 1/4 times car be achieved for stock RWD cars, Its very difficult to get within 0.5 - 1 sec of an actual time in a FWD car. It could be to do with slightly different physics for the front and rear, bu I also believe that the very short throttle travel in the GT series does not make it easy to launch a FWD car.

RenesisEvo
From GT to GT4:

Lots of Oversteer - Oversteer - Understeer - More Understeer

It felt like each installment just added understeer into the car. I hate understeer, so I can hardly say I was amused. But then the realism is vastly greater in GT4 and I appreciate that. I had my first true tank-slapping in an M5 in GT4-Prologue. That to me marked progress. GT3 - it was soo difficult to get a car to spin (ASM off before you ask), now the cars really do seem to act like real-life beyond the limit. GT1 and 2 were arcady, yes. Maybe that is why GT4 has the least in all of them in the arcade mode - perhaps PD finally realise that an Arcade Mode in a driving sim is a tad oxymoronic. It's a bit like putting an Arcade Mode into Microsoft Flight Simulator.

Good points, particularly with regard to the arcade mode, thinking about it, the arcade mode has become less important with each game.

Personally I'm all for the understeer, but then I'm a realism Junkie in my driving games/sims, just what apeals to me; but the bit about the M5 in GT4 Prologue brings back memories. :)

Regards

Scaff
 
Gran Turismo – A comparison across the series. Part 2

After a number of requests to provide comparison data between the GT games, using 4WD and MR cars. I have once again thrown all my spare time in the bin, risked divorce and avoided decorating; all so you can see the following two cars go head to head across the GT series.

So the two cars this time are

4WD Nissan Skyline GTR (R33) V-Spec
MR Honda NSX (base model in all but GT3, which uses an Acura NSX)

The testing conditions and requirements are to be the same as the original test and again I have attached a summary sheet of the results to the bottom of the post.


Gran Turismo


0-100-0 Test
Skyline (R33)
0-100 11.28
100-0 4.00
Total 15.28

Honda NSX
0-100 11.36
100-0 4.24
Total 15.60

As with the first GT 0-100-0 tests neither car suffered from wheel-spin off the line (not that I would expect the R33 to), once again very simple and straightforward.

The times for the two cars were very close and hopefully this will allow me to compare both the cars and the games.


Grand Valley
Skyline R33 2.05:993

Honda NSX 2.07:507

Once again the original GT game’s age is brought into focus here, almost no understeer from either car and on-demand oversteer. The only real difference between the two cars was that the NSX was the slightly more unstable of the two when you get on the brake while cornering.


Donuts & Handbrake turns
Donuts first, and the Skyline has to rate as one of the funniest things I’ve seen in a while. It does do-nuts, but the car rotates around a central point about where the gear lever is. Very unrealistic, but very funny as well. If you have GT then you must give this a go. The NSX just got on with the job, do-nuts on demand, just keep the throttle buried and away you go. As with the Griffith in the last test, its not realistic as no throttle control at all is required.

Handbrake turns were carried out in both cars with ease, speeds from 30 through to 60 caused no problem at all.


R33 Vs NSX
Victory in both the 0-100-0 and circuit tests seal the win for the Skyline in the original GT, however this was mainly caused by the NSX suffering under braking when compared to the Skyline, resulting in a 1.514 second defect. The 0-100-0 test was much closer with the NSX only losing out by 0.32 seconds, so while a win for the Skyline, it’s a narrow margin.


Gran Turismo 2


0-100-0 Test
Skyline GTR (R33) (Control/Normal):
0-100 13.851 / 13.851
100-0 4.832 / 4.045
Total 18.683 / 17.896

Honda NSX (Control/Normal):
0-100 13.087 / 12.564
100-0 5.408 / 5.086
Total 18.495 / 17.650

No wheel spin at all for the Skyline, as it should be being a 4WD car; the NSX suffered no wheel spin an the Normal tyres and only a small amount on the Control tyres. The small degree of spin encountered by the NSX on the control tyres was easily managed and again both cars posted very close overall times. The NSX was the quicker to 100mph, but the Skyline better under brakes. Overall the NSX managed to win this one, but by the narrowest of margins.


Grand Valley
Skyline R33 2.24:807 (Control) / 2.14:224 (Normal)

Honda NSX 2.24:939 (Control) / 2.13:872 (Normal)

The R33 displayed almost no understeer on Normal tyres and only a little more on the control tyres. Both sets of tyres would allow oversteer on demand, which was predictable and easy to hold and correct.

The NSX was much more sensitive to the tyres fitted, particularly in regard to braking stability, the control tyres demanded almost totally straight braking, while the Normal tyres were more forgiving. Understeer was noticeable on both tyre type, but transferred to oversteer if required.


Donuts & Handbrake turns
The NSX does allow you to perform donuts with ease, a slight on/off motion with the throttle was required to balance it, however the front tyres did push the car out of line slightly. Handbrake turns were straightforward in the NSX at speeds between 30 and 60 mph.

The R33 was a totally different story, donuts were handled in the same manner as the R33 in GT, with the car rotating around a central point. Handbrake turns were even worse, the R33 simply refused to play. Attempt a handbrake turn at any speed with Control tyres fitted and the car just stopped with a slight (and I do mean slight) change of direction. At first I thought my controller had a problem, however I tried it with a couple of other DS2’s and even an original DS and they were all the same.

The story was much the same for the Normal tyres, donut around a central pivot point and very poor handbrake turns; although you can get around 45 – 60 degrees of rotation with the Normal tyres.


R33 Vs NSX
In the 0-100-0 test the NSX was the quicker to 100mph, but the Skyline better under brakes. Overall the NSX managed to win this one, but by the narrowest of margins (at most 0.246 seconds). The story was almost exactly the same on the track, while the R33 was 0.132 seconds faster on control tyres, the NSX was 0.352 seconds faster on normal tyres.

This one is almost to close to call, but the balance of faster times goes to the NSX, but only just.



Gran Turismo 3 A-spec

OK confession time, I forgot to check when I started that all the games had a base spec NSX, so when I got to GT3 I had a bit of a shock. I have instead used the Acura NSX as the closest possible car, and it does weigh slightly more than the Japanese/Euro spec Honda NSX models.

0-100-0 Test
Skyline GTR (R33) (Sim/Normal)
0-100 13.74 / 13.36
100-0 4.38 / 3.46
Total 18.12 / 16.82

Acura NSX (Sim/Normal):
0-100 14.76 / 13.92
100-0 4.36 / 3.90
Total 19.12 / 17.82

Again no wheel spin from the 4WD Skyline, just straight off the line and through the gears to 100 mph and then hard on the brakes. The NSX was a totally different story, the Sim tyres require a great deal of throttle control to get even a reasonable launch from the car, and the Normal tyres did make life a bit easier at launch. After that the NSX was also just a case of through the gears to 100 and then hard on the brakes.


Grand Valley
Skyline R33 2.27:510 (Sim) / 2.15:091 (Normal)

Acura NSX 2.28:800 (Sim) / 2.15:867 (Normal)

The R33 on Sim tyres was very much like driving with the front wheels disconnected from the steering wheel, with almost no understeer and oversteer by the bucket-load. The car just did not feel connected to the road at all and nothing like a 4wd car or a Skyline (and yes I have driven one). Switch to Normal tyres and it was like the mechanics had been out and reattached the steering. Nice understeer to oversteer progression, easy to control and with excellent stability under braking.

The Acura NSX on Sim tyres felt for the first time like driving a mid-engined car, set it up correctly for the corner and you get gentle understeer on the limit that quickly pushes to rapid oversteer if you step over the line. However be smooth with your inputs, think a few corners ahead and set the line right and the NSX is an easy car to drive quickly. Switching to Normal tyres and the car handled as above just with the limits pushed that much further away.


Donuts & Handbrake turns
The Skyline allowed donuts on either tyre, but would push wide and move around, but at least its stopped spinning around a central point. Handbrake turns were slightly less embarrassing that in GT2, but not by a great deal, still pushing wide and stopping too soon

The NSX allowed both donuts and handbrake turns to be performed with relative ease, you simply need to balance the throttle for the donuts.


R33 Vs NSX
The Acura NSX was a second slower in the 0-100-0 run that the Skyline on both tyre choices, no big surprise as the Acura NSX is heavier that the Honda model and the Skyline does have the advantage of 4wd off the line.

Around the track the Skyline was faster by 1.29 and 0.776 seconds for each tyre type, quite a big difference even considering the weight increase between the NSX versions.



Gran Turismo 4


0-100-0 Test
Skyline GTR (R33) (N2/S2)
0-100 12.58 / 12.54
100-0 5.10 / 4.12
Total 17.68 / 16.66

Honda NSX (N2/S2):
0-100 13.42 / 12.92
100-0 5.02 / 4.28
Total 18.44 / 17.20


The Skyline on N2 tyres allowed a fraction of a second worth of wheel spin but that was all from a full throttle launch, with S2 tyres fitted no wheel spin and a very simple launch.

The NSX required a well balance throttle at first off the line, but once the weight was transferred to the rear under acceleration it was simple to get the power down. The S2 tyres were easier to control and required less modulation of the throttle on launch.


Grand Valley
Skyline R33 2.29:241 (N2) / 2.17:998 (S2)

Honda NSX 2.28:393 (N2) / 2.17:289 (S2)

The Skyline on N2 tyres offered very predictable understeer that progressed to oversteer, the progression was easy to play with using the throttle and even small four-wheel drifts could be induced. A very easy car to drive quickly

The NSX was much more of a challenge; the car needs to be balanced for the corner because unless you get you line right correction is difficult. A throttle lift or dab of the brakes mid corner will unsettle the car and as a too fast entry to the corner will lead to understeer, you have to get it right. A challenging car to drive at the limit (as is the nature of the MR layout), but very, very quick.

Changing to the S2 tyres that come with the cars as standard has a predictable effect, as the grip levels rise your need to be aware of the tyres limits drops and so do your lap times.


Donuts & Handbrake turns

The R33 almost lets you do a donut, you can get the car to pull maybe 1½ before it straightens out, it does almost feel like the 4wd system takes over. The NSX again almost pulls off the donut, but its mainly just large circles rather than nice tight donuts.

Both cars will almost allow a handbrake turn, but run too wide and stop too soon.

Its again not a good showing for GT4 in this area, but it’s a lot better than the Griffith managed.


R33 Vs NSX
A very close one again, the R33 claiming the 0-100-0 again by small margins (0.76 and 0.54 seconds) mainly due to its 4wd traction off the line. However on the track despite being a more challenging car to drive the NSX was quicker by 0.848 and 0.709 seconds, again a small margin.




Conclusion

The above test really just helps to confirm the results of the first run through with the Integra and Griffith, and just really reinforces my thoughts from that test.

The most interesting thing about this test was just how close the Honda NSX and R33 Skyline GTR V-Spec are in terms of performance. Across four games the results was 2 to each car, however the Skyline is always going to have the natural advantage in the 0-100-0 test of 4wd, which just make the launch so easy.

Its around the track that I found things most interesting, particularly with GT3 and GT4 using the Sim/N2 tyres, they really did feel like two totally different cars, with different driving style required to get the most out of them.

The R33 being very easy to drive and a car you have to take by the scruff of the neck and beat the performance out off it. Its big and prone to understeer, but you can drive through it on corner exit and use the 4wd to hammer down the next straight.

By contract the NSX will throw you into the nearest armco if you try the same approach, it’s a classic mid-engined car, the limits are high, but once you get to them you have to know what you are doing. Commit to a corner on the limit and that what you have to do, back off the throttle or brake and the NSX is going to get upset with you. It’s a car that needs a smooth and balanced approach, but if you can do that its outlandishly fast for it power output and a joy to drive.

So which one wins overall, well they both do, the lap times are so close across all the games and they are such different cars with radically different driving styles required that what suits one person may not suit another. Guess I’m just lucky, I love them both.

Regards

Scaff
 

Attachments

  • Gran Turismo Figures.doc
    56 KB · Views: 28
The spinning on a central point happened with a couple other cars (early Mitsubishi GTO/3000GT's in particular). I think it was a poor early attempt at working in 4 wheel steering, or something. You're right though: It's funny as hell to watch. Also, I think an explanation for the inability to handbrake with the GT-R R33 in the later games could very well be because of some special physics for the 4 wheel steering combined with 4 wheel drive. I wonder if the Nissan 300ZX ('89-'98) would do the same thing, as I believe it has 4 wheel steering and it is not a 4 wheel drive car.
 
👍 Great work again Scaff 👍 Just a shame the wife doesn't appreciate it as much as we do :indiff: Maybe if you got Jordan to pay you for your endeavours? Send him an invoice, see what happens... 💡 ("You just keep on thinking, Butch, that's what you're good at... ")

Driving in GT4 with a DS2 is a much more delicate affair than it was. Would you say that this has altered your driving style for the older games, or did you revert to old habits?

I always had a problem in the early games with MR and RR cars: some of them (Lotus Europa especially) would oversteer if you merely showed them a corner at speed, and were almost undriveable as a result (Lotus Esprit and Ruf Porsches also bad). NSXs were always one of the better ones but the race versions were very tricky to drive fast (I usually preferred a Supra). This used to p*ss me off, as, power on, most mid- and rear-engined cars will grip till kingdom come (Beetles on wooden tyres excepted, of course). How do you feel about this?

Thanks again,

Ed
 
Back