Unrealistic physics/features in GT5P

  • Thread starter The Outlaw
  • 66 comments
  • 7,232 views

The Outlaw

Premium
3,042
United States
Bay Area, California
GTP_timeattack
I thought I'd start a thread that would accumulate some of the unrealistic physics or features found in GT5 Prologue of which you guys have experienced. Hopefully we can provide some constructive criticism for PD to further improve the final version of GT5 đź‘Ť After all, this game is supposed to be the "ultimate driving simulator". Anyways, hopefully PD comes across this thread shortly. :)

Here are a few of my criticism:

1. A select group of cars (specifically the Tuned Skyline) brake better when you use massive engine braking to aid in slowing the car. Meaning, while you are under braking you rapidly downshift to the gear you will use at the apex of corner, and allow the engine to bounce of the rev limiter while braking to help slow the car down dramatically. In some cases engine braking can decrease stopping distances in reality, but this is usually just with 4wd cars when braking from high speed, in a STRAIGHT LINE. You're not going to see benefits from bouncing off the rev limiter in a RWD car while braking around a radiused turned, or while trail braking. You would just end up completly upsetting the balance of the car, likely causing a sudden loss of rear grip. Trust me, I've done this a time or two on the track. :dopey: Luckily I didn't blow the engine or worse yet, spin out and trash my car or myself. I also find it quite strange how only a select few cars in the game do this.

Here's a lap I did at Daytona Road with the Tuned Skyline that shows what I'm speaking of. I did a perfect demonstration of the extreme engine braking going into T4, which clearly shows how quickly the car slowed down under braking thanks to this unrealistic affect.


2. The R35 GT-R when set-up with a 45/55 torque split and proper suspension accelerates dramatically quicker when slightly powersliden out of a corner. It also helps to upshift a bit sooner when using this technique...at least while the tires are still slipping and the revs jump up a bit. This flaw is best demonstrated at Daytona Road, where using this technique can yield a good 1 second a lap advantage. There's absolutely no reason why the GT-R should accelerate quicker in a powerslide/yaw angle versus when accelerating without any slip angle coming out of the corner, especially given the GT-R's rather broad torque curve. I'll try and post a video of this shortly, as it is a bit hard to explain in words unless you've experienced it first hand.

3. Some of the cars in GT5P have extremely unrealistic handling at low speeds. A lot of the cars in the game, including the tuned cars won't even do donuts from a stand still. This is quite unrealistic especially considering these cars certainly have a limited slip or posi differential.

4. The HP/TQ graphs in the tuning department are far fetched when the power levels are adjusted from the stock value. This makes maximizing your gearing among other things a guessing game to an extent. First off, torque and horsepower ALWAYS cross at 5250 rpm, given the engine revs that high. That's just a simple part of the equation in which you take the engines torque curve throughout the rpm range and derive a horsepower curve/plot for the rpm range. After all, horespower is just a calculation of torque. Torque is all a dyno really measures. When you start playing with the power levels, the point in which hp and torque meet, ends up floating all over the place.

Also, it would be nice to have a much more in depth power graph, with 1k increment in rpm's on the x axis. This would help to tell what in the hell we're doing in terms of gearing our cars to really maximize the engines powerband.

5. The tuning in GT5P doesn't really seem to correspond with how certain changes would affect the car in reality. I have a good amount of experience in set-up, from setting up my track car, kart, and rc cars. It's frustrating trying to tune the cars at times, as the particular changes to certain cars don't have the affect they would in reality. It makes for a lot more trial and error than is really necessary. I remember seeing a 430 tune with a 6/2 spring rate that handled quite nicely on S2's at Suzuka. In reality this set-up would understeer like a pig with a spring that hard in the front, and a spring that soft in the rear.

6. A lot of you probably know about the "toe bug" on the Sti's. I believe when you set the toe to F: +.50 R: -.50 the car gains a massive acceleration advantage. I believe this yielded something like 2 seconds a lap at Suzuka. This again is extremely unrealistic, considering having that much toe would only scrub speed from frictional losses. This further helps to prove my point in criticism 5.

Anyways, that's it for now. :) If I experience or think of anything else I'll be sure to post it. Feel free to share any bugs or unrealistic features that you guys have experienced! đź‘Ť
 
4. The HP/TQ graphs in the tuning department are far fetched when the power levels are adjusted from the stock value. This makes maximizing your gearing among other things a guessing game to an extent. First off, torque and horsepower ALWAYS cross at 5250 rpm, given the engine revs that high. That's just a simple part of the equation in which you take the engines torque curve throughout the rpm range and derive a horsepower curve/plot for the rpm range. After all, horespower is just a calculation of torque. Torque is all a dyno really measures. When you start playing with the power levels, the point in which hp and torque meet, ends up floating all over the place.

I'll agree with you there. In gT1 & gT2, power and torque bands would change to different RPMs. They weren't just these static, shape-unchanging graphs.
 
OMG :eek: I didnt even know you could break like that in GT5P. I know there is no damage in GT5P, but if there was that motor would be toast! What should happed though is the balance of the car should of been way off, and the car should of lost control. Thanks for posting up this video though, i have never did that type of breaking ever, and i dont see me using it.
 
hey ta, i have one that i have found to be massively harmful in races. sorry i dont have the time right now to write it up in as splendid a fashion as you have. :)

it is: the loss/flattening out of the advantages that should be gained by accomplishing a better drive and/or turn finish than an opponent you are following.

i do believe this is all-pervasive, even to include a ghost you are chasing during a time trial. it seems that if you do a better job coming out of a turn, you do stand to gain on the car in front of you a bit, but if the period of acceleration is long enough (in other words, as long as you are on a portion of a track where you stay in the throttle for a long period of time and are not confronted by a turn very soon) you rather suddenly stop gaining on your opponent.

the obvious exception is when drafting- although it is somewhat apparent here as well. the effect, of course, is offset by the advantage gained by the draft effect.

i think one other exception is when there is a massive speed differential. the problem i am describing seems to be most apparent in cases where you may have earned a few mph advantage coming off a turn.

if you havent noticed this, try running against a ghost, or if you are in a race try taking a slightly different exit line than the car in front so that you are out of draft, and notice that if you have accomplished a better drive out of the turn, you will begin to gain and may even pull up next to the other car, but then your speed advantage flattens out.

its almost as if, after having accelerated for longer than a certain period of time, the game equalizes the rates of acceleration for like cars.

has anyone else noticed this?
 
What's shown in the vid is not engine braking, but engine breaking. You really should be hitting rev limiters during a race. All engine braking is is shifting into a lower gear to change your rolling momentum. Shifting down reduces your trasnmission's moment of inertia, which results in more deceleration due to rolling resistance. You don't even have to touch your brakes to engine brake.



after having accelerated for longer than a certain period of time, the game equalizes the rates of acceleration for like cars.

has anyone else noticed this?

That is perfectly realistic. Acceleration goes to 0 as a car approaches top speed. With the PP system, top speed is pretty even, minus the occasional unbalanced car.

My problems with the game:

Tires are horrible, too little longitudinal grip, tire list is all wrong. The tire fully determines a car's stopping distance and lateral g.

Aero is poor, tuning causes no change in drag, changes is aero seem very small (might be because they're not in real units though)

Suspension tuning as a whole isn't very effective, difference should be much bigger.

4WD has huge advantages over RWD that shouldn't exist.
 
Last edited:
That is perfectly realistic. Acceleration goes to 0 as a car approaches top speed. With the PP system, top speed is pretty even, minus the occasional unbalanced car.

It looks like you've zeroed in on a portion of my post that wasn't communicated properly, and missed my overall point- sorry about that. My point relates to losing advantages gained by exiting a turn faster than another car. An extra 3mph would have one car travelling at 100mph when the other is going 97mph, and I have noticed this advantage disappears in the game.

It sounds like you're raising a related point about how equal machines top out at the same speed, and cease to accelerate - I think we all undertsand that - but I am speaking about a point much earlier than this. In the case I am describing, the cars are nowhere near top speed, and the car that is travelling faster loses its mph advantage while both cars still have a long way to go before reaching top speed. Is it clear to you now?
 
I think I get you now. My confusion came from that fact that I've never seen it, and have actually seen the opposite (normal car behavior).

I guess I'll keep an eye out for anything odd.

One possible reason though, could be gearing. If someone has a bad gear setup, where they have to keep shifting repeatedly, it might slow their car down. Like wise a good setup wouldn't. Some people also learn their car's powerband really well and know when to shift (when it's not the redline). That wouldn't explain it on ghost cars though.
 
I have noticed the first one a lot, especially with the G25 using the H and clutch. Downshift some cars from 6th to 2nd and they will brake so much quicker than usual.
 
It would all depend on tires and gearing. Once you reach a certain speed, acceleration advantages even out. Would have to provide video evidence.

The power torque curves change intersection because the units are not static. As you increase horsepower in a naturally aspirated car, you're not increasing torque. So while the units on the torque side of the graph stay similar to stock, the units on the horsepower side move to accomodate the taller graph.

I haven't tried the Subaru bug. Although I'd need citation that massive toe doesn't increase acceleration in real life.

The GT-R. Well, it could be a bug, it could not. The GT-R has such good traction that some people turn VDC off in real life to get faster track performance. I don't know whether the performance is higher because it allows more wheelspin or because it allows higher slip angles (you can execute mild powerslides even with it on). There is nothing but anecdotal evidence for this trait.

And, of course... in Standard Physics... powersliding is definitely faster, no matter what the car. In Pro Physics, I haven't seen proof that powersliding in anything is faster (from viewing the time trial leaderboards). Even on the Norwand rally stage, fastest times are with a clean drive.

The tuning thing is the one that's really off-kilter. Due to the way GT models tire physics, many suspension changes don't have the desired effect. Of course, the problem with making declarative statements about specific set-ups in relation to real-life... how do you know how hard the springs actually are? In GT4, we had meters showing how much spring pressure there was... in GT5P, you get a 1-10 rating which could mean absolutely anything.

Longitudinal grip... well... we can only hope that PD finally fixes this... I'm of the mind that they don't properly model tire deformation under acceleration and braking... either that, or the tires are modelled as having a rock hard 50 psi, all around.
 
In terms of tuning even though we don't even have the level that say, gt4 had, I find that things like front and rear spring rates should have more of an effect for sure.
Also something seems oddly opposite about stiffening of the fronts to the rears. For example when stiffening the rear in other sims such as live for speed (which although far from perfect, is known to be a fine sim) I'm going to get oversteer. In gt5p doing this always creates understeer while doing the opposite, even a 10/1 setup, merely frees up the car...slightly, but not anywhere near what I would expect.
Something just feels wrong to me. If anyone knows differently, I can honestly say please correct me.:)
 
You comes off as wanting to disagree just for the sake of causing controversy :boggled:

It would all depend on tires and gearing. Once you reach a certain speed, acceleration advantages even out. Would have to provide video evidence.

The power torque curves change intersection because the units are not static. As you increase horsepower in a naturally aspirated car, you're not increasing torque. So while the units on the torque side of the graph stay similar to stock, the units on the horsepower side move to accomodate the taller graph.


Wrong. As you increase the horsepower value you're increasing the torque generated at higher rpms...this is clearly demonstrated in the graph. Just because the PEAK torque value remains static doesn't mean peak hp won't increase if the torque is elevated or carried into the higher revs. After looking at the graph again I realized that when you increase the horsepower, the line doesn't really carry up any higher on the graph like it should. The only thing that really changes is the actual hp curve and of course the peak hp number. This explains why the values do not statically meet at 5250 rpm when the hp is increased.


I haven't tried the Subaru bug. Although I'd need citation that massive toe doesn't increase acceleration in real life.

Do you really need a citation of this? You come off as knowing a thing or two about cars, yet you can't comprehend that toe does absolutely nothing to improve acceleration? Think about it...it's common sense. An increase in toe (in or out) only increases friction or rolling resistance the tires have with the road. When there is toe, the car goes straight but the tires want to go their own separate directions, creating scrub or friction. This of course ends up reducing the power transmitted to the ground. Here's a simple analogy. Imagine trying to ski down a hill with your skies toed in about 30 degrees on each foot. Do you think that this is somehow going to increase your acceleration or potential speed skying down the hill? If anything it would drastically slow you down.

The GT-R. Well, it could be a bug, it could not. The GT-R has such good traction that some people turn VDC off in real life to get faster track performance. I don't know whether the performance is higher because it allows more wheel spin or because it allows higher slip angles (you can execute mild powerslides even with it on). There is nothing but anecdotal evidence for this trait.

Only a HIGHLY experienced pilot will get the most out of the GT-R with the VDC off. VDC gives a large margin for error, yet can perform within about 1-2 seconds a lap compared to with all aids off and in R mode. With that said, I was strictly referring to the acceleration of the GT-R when in this powerslide...not necessarily cornering speed, entry speed or anything else. I'd like to see a valid explanation as to why a car would accelerate so much quicker when the tires are excessively slipping versus when they are at the threshold of there available grip. I'll be putting up a video demonstrating this in a little bit.

And, of course... in Standard Physics... powersliding is definitely faster, no matter what the car. In Pro Physics, I haven't seen proof that powersliding in anything is faster (from viewing the time trial leaderboards). Even on the Norwand rally stage, fastest times are with a clean drive.


Again, this is only with the torque setting at 45/55 when using the GT-R. You can't touch the torque split in time trials, so your point is moot.

The tuning thing is the one that's really off-kilter. Due to the way GT models tire physics, many suspension changes don't have the desired effect. Of course, the problem with making declarative statements about specific set-ups in relation to real-life... how do you know how hard the springs actually are? In GT4, we had meters showing how much spring pressure there was... in GT5P, you get a 1-10 rating which could mean absolutely anything.

You bring up a valid point here đź‘Ť The spring rating system FOR ALL WE KNOW might not be an accurate indicator of the factory/default spring stiffness. Meaning a 6 in the front might not actually be three times as stiff as a 2 in the rear. This might be because the games default settings don't reflect the exact spring stiffness of the stock cars. Meaning a 2/2 default spring setting might be something like a 300lb front spring and 450lb rear spring, and not necessarily an even value front and rear (200lbs/200lbs) Your guess is as good as mine when it comes to this one. Regardless, PD can do a lot to improve in this gray area. And of course, the purpose of creating this thread was to give constructive criticism to help PD improve the GT series đź‘Ť

Longitudinal grip... well... we can only hope that PD finally fixes this... I'm of the mind that they don't properly model tire deformation under acceleration and braking... either that, or the tires are modelled as having a rock hard 50 psi, all around.
 
Last edited:
In terms of tuning even though we don't even have the level that say, gt4 had, I find that things like front and rear spring rates should have more of an effect for sure.
Also something seems oddly opposite about stiffening of the fronts to the rears. For example when stiffening the rear in other sims such as live for speed (which although far from perfect, is known to be a fine sim) I'm going to get oversteer. In gt5p doing this always creates understeer while doing the opposite, even a 10/1 setup, merely frees up the car...slightly, but not anywhere near what I would expect.
Something just feels wrong to me. If anyone knows differently, I can honestly say please correct me.:)


Exactly my friend. đź‘Ť In general and in reality, increasing the rear spring stiffness will only increase the oversteer tendencies of the car. In GT5P you can put 1/10 spring combo on a car and still drive it as if it weren't set up half badly. The other problem I've experienced is that some cars respond in a realistic/predictable fashion (when tuning the suspension) and then you have some cars where you end up trial an error'ing different settings to get the desired affect you're looking for, without any type of scientific approach.

Also I agree that the range of settings don't have a large enough impact on the handling characteristics of the cars. With some of the settings we're capable of dialing in, the cars would be absolutely undriveable, or they certainly would not handle as well as they do :lol:
 
Last edited:
It all kinda worries me a bit, not to be a negative guy. I would have hoped that by now: YEARS after gt4, that even basic prologue tuning would be much more realistic....ah well...:indiff:
Thats why we can always go and fire up the pc and........nah, won't get into it here.
 
It all kinda worries me a bit, not to be a negative guy. I would have hoped that by now: YEARS after gt4, that even basic prologue tuning would be much more realistic....ah well...:indiff:
Thats why we can always go and fire up the pc and........nah, won't get into it here.

I agree. It is a bit discouraging considering the basic tuning in this game is so out of line, and the GT series is already some 12 years old :nervous: It's not exactly rocket science either :dunce:
 
Those unpenalised downshifting on overreved engine is bugging me too. It will survive till day PD will implement real "engine engine" with stalling, real clutch slipping, realistic rev limiter, innertias inside engine, damage on overreving, engine shaking in silentblocks etc.
 
How about no torque steer, launching from a halt on a cambered road for all cars? How about most cars inability to perform a donut?

The aerodynamics seems completely off, changing the imaginary wing unit to alter downforce sure down force increases but the increased drag isn't noticeable, it becomes a huge advantage to run with full wing all the time no matter what the track.

Then of course we have to notorious super draft, every car seems to displace the same air of a truck in its wake which gives a huge tow for the car behind to leapfrog the guy in front.

Niky I can vouch for the validity of the subaru bug. The bug works only for the two sti's it works by maximising both front and rear toe, the result is a large acceleration gain.
 
1. The steering strength settings needs to be increased. I'm on 'power steering off' and on maximum strength setting of 10, but it's still not comparable to real life.

2. The 'Yaw' movement is exaggerated and needs to be toned down. It's almost as if PD tried too hard in implementing some driving physics and just over did some aspects abit.

3. Road tyres(N3) don't feel natural and needs to be looked at. Maybe its a combination of steering strength, too much yaw movement and oversteer. The engine note when the car is pushed past it limits on a bend is too sedate - almost as if you at 2000rpm when it actually reads 5000rpm and should sound as if its under 'load'.

I think this will go a long way to improving the best driving sim on the market:tup:
 
And, of course... in Standard Physics... powersliding is definitely faster, no matter what the car. In Pro Physics, I haven't seen proof that powersliding in anything is faster (from viewing the time trial leaderboards). Even on the Norwand rally stage, fastest times are with a clean drive.

Try watching replays of cars like the Z06 at Suzuka and London (top times) and you'll find your pro physics powersliding. It's like that because the Corvette (amongst other RWD cars) is unfairly taily.
 
@timeattack07:

Fair enough... I was thinking more along the lines of real-life tuning. You get to a point in NA tuning where you can't increase the torque any further... you simply move it up the rev range. What I was getting at was that the torque units don't increase on the graph at the same rate as the hp units, which means that it won't look in any way like a regular dyno-graph, which usually uses the same units for both measures... and yes, I agree, PD should put a matrix so we can actually tell what we're looking at. It kinda helps to know what rpm 90% power comes at, if the car loses a lot of power at redline, so you can set your gearing up to reflect this.

RE: Subaru: Just saying... is all... :lol: ...What I mean is, nobody has tried it in real life. Maybe it actually works. :lol: But I can accept that it is probably a bug.

The GT-R and VDC off... everyone who uses it says they see a gain in performance from this. Again, I don't know what factor of VDC slows the car down, but who knows? It might even be psychological. In my experience, cars with more grip than power can get a boost in acceleration from spinning the tires a little... but there also may be something in the way GT5P models the non-existent clutch or torque converter on cars. There's a built-in torque-limiter and rev-matching in GT5P... one that cuts the power on shifts and prevents unsettling torque from reaching the tires (probably also the cause of the engine-braking everyone is complaining about). Maybe the loss of traction gets around this behaviour, or maybe it's just a way of preventing the engine from bogging down in corners.

I think the springs and shocks are a scalar value based on stock values... so... with "stock" springs at 2, 10 is 5 times stiffer than stock... and with "stock" shocks at 1, 10 is 10 times stiffer than stock... whatever stock is. Which means that cars that are stiff to begin with have an unfair advantage in-game.

And yes, I'm aware that rear stiffness never has the desired effect. I'm wondering whether this is due to the lack of sidewall flex modelling or the lack of tire-skip (I've only seen it in rare situations) due to the anti-flip-over nazi hidden in the game program. That's actually the one HUGE physics bug still left in GT5P. And the sad thing is, it's in there by design.
 
Then DON't play it

The whole point of this thread was to discuss the Physics problems and other things, still in GT5p and to provide constructive criticism for PD. Who are you saying that to and is it really necessary. Afterall the same could be said to you, if you don't like this thread, Then DON'T post here.
 
Afterall the same could be said to you, if you don't like this thread, Then DON'T post here.

Nuff said đź‘Ť

Some of the GT fanboys get so upset if you even have the slightest criticism toward the GT series...you would think you were talking about their mother :lol:
 
Nuff said đź‘Ť

Some of the GT fanboys get so upset if you even have the slightest criticism toward the GT series...you would think you were talking about their mother :lol:

So true:lol: Afterall, it's not as if your comments are unfair in any way. You are not bashing the game, just pointing out areas of the game that need improving. For instance the GT-R with torque set at 45/55 does indeed have at least a second advantage on a GT-R with say 40/60, this is not realistic, especially as the sliding and wheel spin is what makes it faster.
 
Then DON'T play it and buy the real cars since they are perfect.
The reason i am saying don't play it is because everyone now seems to start threads to moan.
 
Then DON'T play it and buy the real cars since they are perfect.
The reason i am saying don't play it is because everyone now seems to start threads to moan.

Again, you say everone starts threads to moan, while this is partly true, don't you think by moaning about people moaning, in a thread about moaning as you put it, you are infact no different to everyone else. Except you have not posted a thread to moan, just a couple of moaning comments that you didn't need to post.
 
Dude...just go back in your cave...please đź‘Ť Since you don't seem to care about realism maybe I'll send you my copy of GT1 that I have no reason to touch. I'm sure you'll be quite content with the impressive physics in that game đź‘Ť
 
Back