Can we actually call GT5 a simulator?

[TurboProp]

Probably because you don't even realize he was responding to your off the wall comment:


Ironic you forgot your own post that started your own conversation about something so irrelevant and not previously mentioned in this thread.

So now I'll ask: Why would you try driving on the street like you play GT5?
You brought it up, so you must think it's questionable in order to feel a need to say it's not a good idea.💡

They are not the same...game vs. real life

 

[LMSCorvetteGT2]

your american , you obviously don't know what the real meaning of troll is.

I have tried iracing specifically in suzuka circuit , the layout of the track is there but the bumps and the tarmac type or the in other terms realistic details are not there and physics wise , its trash I'm afraid. No question for me.

I have tried http://www.youtube.com/watch?v=udoydg7vtx0 , that's a real Ferrari simulator in abu dhabi , the physics you can relate to the F10's physics online in GT5 with tire wear and fuel load. The driving position in ferrari world's f1 simulator is just like a real f1 car because the simulator's seat is actually real f1 seat model (carbon fiber tub) although with some base forces simulated , the driving perspective is totally and completely different.

I have tried http://www.youtube.com/watch?v=kCwH8fQWSFI&feature=related as well and also you could relate to GT5's race car online physics , I'm not sure anymore but the physics before 2.0 came was much better.

actual weekend racer? for sure in america iracing is "#1" I mean everyone from there just to describe a few exactly like you thinks they know everything anyway but nascar doesn't even use it and formula one has their own simulators way more simulated than anything america can give , you can't even compare any kind of detail of iracing to a real team simulator.

GT5 is a real simulator , the older GT's you can call arcade , Vettel or Coulthard themselves can answer you if they even bothered to.

Wow xenophobic some what? No one is claiming it's the best because it's American and I'm American...Nascar drivers and IRL drivers all use iRacing and swear by it, so not sure where you've got your info. I never said race teams use iRacing any where in my post, sorry to break it to you but you twisted my words. I said (to paraphrase myself) that if Gt were this grand simulator like you've made it out to be then why isn't every team going out and buying it to train their drivers on? Why aren't these F1 teams asking PD to build their simulators if it's so good? I'll tell you why, cause the ability to build a true sim that can grow isn't something PD or T10 or even EA know how to do.

Simbin/ISI which work on rFactor have their tech in the RBR pro simulator, rFactor is probably the closest challenger to iRacing. So if GT is better then why not use it rather than rFactor a game quite similar to iRacing? Also many would say that iRacing is better. The FW31 seems more real feel or typical of what a car would do at fast speeds more so than GT5. Also Simraceway is another one to look out for as a more authentic sim. I'm a fan of GT but at least I can be honest when a game isn't as real as others, not sure why you have to get defensive.

Also troll doesn't have one universal meaning and I find it highly ironic that you claim I act as if I know everything, yet you're rash enough to come out the woodwork doing exactly what you accuse me of. Hypocrite much?

 

[jtv90069]

Polyphony Digital Inc. has produced simulations and demos for many major car manufacturers, including Toyota, Subaru, Nissan, and Honda, for both in-house and promotional use.

http://www.worldwidestudios.net/polyphony

 

[sinbad]

They are not the same...game vs. real life

There's an obvious difference, of course, or else the only simulation of real life would be.... ....real life.

But you can't put down a sim because you aren't scared of crashing when you use it.

 

[peter_vod69]

I can't help but note just how seriously most of the Americans in this thread take themselves. I can only surmise that most Americans are experts at everything.

A couple of points:

1. iRacing et al publicise the fact that real races use their game to train on. It's called marketing. GT5's marketing is pretty poor in this regard, but several racers have used GT for training in the past.

2. The main thing preventing GT5 from being a comprehensive "sim" is the loungeroom. I have a racing cockpit setup with a G25 and clutch, and realistically, that is pretty much where the physical immersion ends for most people, because that's what finances and ongoing relationships will allow.

3. People in this thread should note quote examples like the FW31 - because unless you are Rubens Barrichello, how the hell would you know how it is supposed to feel?

4. This "true sim" stuff is utter bull****. All racing games are at best an approximation of what they are trying to simulate. Some do it better than others in some areas, but none are perfect. WHAT GT5 DOES BEST IS REPLICATE THE FEEL AND CHARACTERISTICS OF THE REAL ROAD CAR.

5. If you don't occasionally drive a car hard on public roads, if you've never felt understeer or oversteer, or indeed explored the boundaries of such, or if you've never done a track day, how are you qualified to comment? Indeed, if you don't drive like that occasionally, you shouldn't consider yourself a driving enthusiast.

 

[Exorcet]

3. People in this thread should note quote examples like the FW31 - because unless you are Rubens Barrichello, how the hell would you know how it is supposed to feel?

5. If you don't occasionally drive a car hard on public roads, if you've never felt understeer or oversteer, or indeed explored the boundaries of such, or if you've never done a track day, how are you qualified to comment? Indeed, if you don't drive like that occasionally, you shouldn't consider yourself a driving enthusiast.

You should know that math exists which can predict the behavior of a vehicle. I don't think anyone here is going to write a simulation to show how car X handles to 99% accuracy, but that's OK. Instead they can apply general physics knowledge and also seek out sources that explain how that car behaves.

 

[MSTER232]

I want to start out by saying I'm not bashing the physics engine. I'm not saying these cars aren't reacting to the road properly, I won't whine and complain about a ton of parts that I think should be in the game, and this isn't a rant about engine noises or standard cars. The vast majority of stock cars are near perfection when it comes to performance. The area of concern has to do with the upgrade system.

The chassis/suspension aspect has had me tearing my hair out since day one. There is no way to tell your vehicle's actual ride height or drop, and there's no way the spring rates are accurate. A 0mm drop with the full customization kit is still lower than stock, and unless you guys know something I don't, there's no defined standard drop when installing the kit. I did some research on spring rates and found that basic sport springs were firmer than is allowed with the premium suspension kit on quite a few vehicles. Again, we're dealing with a magical and undefined number over stock.

The second issue is the power achieved from upgrades. You can somehow get 40hp from bolting a cone filter onto a stock naturally aspirated engine, but a baby-eater turbo (that spools up to the same amount of boost no matter what) often struggles to produce 100hp? As mentioned in another thread, you just can't push realistic numbers no matter what you put into the car, as is most evident with the Toyota Supra. There are also the compatibility issues, but I won't get into that.

At this point, it's readily apparent that modifications are simply not accurate or well-simulated. And, since about 99.99% of racing is done with modified vehicles, very little of the driving performed in-game can be considered realistic. To prove it, I decided to run a little experiment on the Top Gear Test Track...

Stock TVR Tamora '02 with racing soft tires- 1:15.019
Ariel Atom V8 (as tested on Top Gear)- 1:15.1x

Are you telling me OEM-size racing soft tires make an otherwise stock Tamora (a car with 150 less hp and double the weight of the Atom V8) faster than the fastest car Top Gear has ever tested?

Realistically, does PD need to overhaul the upgrade system to ensure that this game can actually be called a simulator by all groups, or is a stock vehicle simulator good enough to earn the title of "The Real Driving Simulator?"

----------------------------------------
Note: TVR was new with no oil change, and the lap was started from a complete stop about 10ft before the line.

The choice of tyres can make all the difference. By adding on tyres that have a higher overall grip, you have a higher cornering speed, quicker acceleration off the line and shorter braking distances.

Do a lap with the tyres the Tamora originally came with and this will prove my point. I can guarantee you will never get a time that fast with the normal tyres, not even 3 seconds behind the time set with racing tyres.

 

[peter_vod69]

You should know that math exists which can predict the behavior of a vehicle. I don't think anyone here is going to write a simulation to show how car X handles to 99% accuracy, but that's OK. Instead they can apply general physics knowledge and also seek out sources that explain how that car behaves.

You mean like GT5 for example?

Right.

 

[Exorcet]

You mean like GT5 for example?

That's an example of the math, though it's not all that accurate. Real time simulators can't process the actual math in most cases.

But the point was, having experience behind the wheel is no prerequisite for this discussion. Someone who doesn't even know what a car is could figure out how one behaves from physics alone.

 

[JDMKING13]

If you are playing GT5 with ABS on you are not getting the best experience in my opinion. With ABS on you need no skill when it comes to braking, something just doesn't feel right when braking with abs(1) With it off it actually feels like braking a real car.

 

[LMSCorvetteGT2]

I can't help but note just how seriously most of the Americans in this thread take themselves. I can only surmise that most Americans are experts at everything.

I haven't seen any Americans make that claim, could you give some examples?

A couple of points:

1. iRacing et al publicise the fact that real races use their game to train on. It's called marketing. GT5's marketing is pretty poor in this regard, but several racers have used GT for training in the past.

It's safe to say that PD probably has more money to pay Loeb and Vettel to say things rather than iRacing simulations. I think I'll take the word of Will Power, Marcos Ambrose, Dale Jr., Scott Speed and Villeneuve among others than just the few GT have. However, that then becomes preferance, but I bring it up because of the feedback they give to iRacing that helps me and others safely say that it's probably more real than your average off the shelf console racer.

2. The main thing preventing GT5 from being a comprehensive "sim" is the loungeroom. I have a racing cockpit setup with a G25 and clutch, and realistically, that is pretty much where the physical immersion ends for most people, because that's what finances and ongoing relationships will allow.

I have to agree, but at the same time even in that area it lacks to an extent. We also must remember once again that GT will never be a PC racer for a few things. It's on a console and secondly it still has to be used by those who don't use a wheel. iRacing, rFactor and GTR games don't have to do this.

3. People in this thread should note quote examples like the FW31 - because unless you are Rubens Barrichello, how the hell would you know how it is supposed to feel?

Well seeing as another quoted the Ferrari F1 sim being close to GT5...that just isn't true. I've used rfactor and the pro version is what was used for the Ferrari sim as well as the RBR sim. rFactor pro is similar to iRacing so I ask again if GT is so good why doesn't PD going into the profession sim business, that way they'll get racer feedback as well as track feedback, and more money to higher people. Then the fans (us) will get a better GT6 or GT7 game at the end of the day! I mean it's so much cheaper to go out and buy GT5 then have ISI or SimBin build them a simulator. Also Rubens didn't drive the FW31, I'd rather as Nico Rosberg or Kaz Nakajima what the car feels like since they raced it. Rubens was too busy winning with the BGP001.

4. This "true sim" stuff is utter bull****. All racing games are at best an approximation of what they are trying to simulate. Some do it better than others in some areas, but none are perfect. WHAT GT5 DOES BEST IS REPLICATE THE FEEL AND CHARACTERISTICS OF THE REAL ROAD CAR.

Once again when several drivers on the professional level claim by it and actually use it and you can race them...that means something. Vettel also seems to swear by his RBR simulator which once again is rfactor pro. So it's not utter bs, it may be that way because of bias for GT. However if a racer says one game over another or says this is as close as real world gets without getting in the real car, that says something. Also the last part is subjective, there are people that claim their car was done well, and other claim that they aren't even the same.

5. If you don't occasionally drive a car hard on public roads, if you've never felt understeer or oversteer, or indeed explored the boundaries of such, or if you've never done a track day, how are you qualified to comment? Indeed, if you don't drive like that occasionally, you shouldn't consider yourself a driving enthusiast.

I've done that.

If you are playing GT5 with ABS on you are not getting the best experience in my opinion. With ABS on you need no skill when it comes to braking, something just doesn't feel right when braking with abs(1) With it off it actually feels like braking a real car.

This I have to 100% agree with, GT5 did that quite well.

 

[peter_vod69]

But the point was, having experience behind the wheel is no prerequisite for this discussion. Someone who doesn't even know what a car is could figure out how one behaves from physics alone.

That is an utterly ridiculous statement. That's like saying someone knows what an orgasm is just by reading about the chemical reactions in the brain.

 

[Exorcet]

That is an utterly ridiculous statement.

It's something that every auto (or any type of) engineer does. You model something as a system, specify an input, and get an output. I hope you don't think that manufacturers build a car before they know what every part of its design is supposed to do.

The difference between an orgasm and a car is that we have equations for the latter. The former isn't understood quite as well.

 

[jtv90069]

It's something that every auto (or any type of) engineer does. You model something as a system, specify an input, and get an output. I hope you don't think that manufacturers build a car before they know what every part of its design is supposed to do.

Prototype build then huge amounts of real world testing and refining with feedback from high level test drivers.

 

[peter_vod69]

Prototype build then huge amounts of real world testing and refining with feedback from high level test drivers.

No you don't need drivers or feedback. Apparently.

 

[Exorcet]

No you don't need drivers or feedback. Apparently.

According to who? Validation is another important part of design.

But all of this strays from the point. There's no need to have experience behind a wheel to understand cars.

 

[Raven_WET01]

The choice of tyres can make all the difference. By adding on tyres that have a higher overall grip, you have a higher cornering speed, quicker acceleration off the line and shorter braking distances.

Do a lap with the tyres the Tamora originally came with and this will prove my point. I can guarantee you will never get a time that fast with the normal tyres, not even 3 seconds behind the time set with racing tyres.

Last time I'm making this point...

The idea behind this comparison was to show that modifications aren't accurately simulated.

Because modifications aren't accurately simulated, cars with modifications don't have realistic physics.

Because the majority of cars used to drive with are modified, the majority of driving in-game is not simulated very well.

If the majority of in-game driving is not well simulated, can we call this a simulator?

Please note there is nothing in there about stock cars not being well simulated. In fact, first paragraph:

The vast majority of stock cars are near perfection when it comes to performance.

Great discussion otherwise guys, I'm starting to like what I see 👍

 

[Ammothief41]

This is all anecdotal, but between having played the game myself, read a bit about discrepancies in performance between otherwise similar cars, particularly the Ford GT and Ford GT (no stripe), and a few of the odd little things missing in the game, I can't help but think that maybe the physics models for some of the cars have been intentionally tweaked where they weren't necessarily running lap times or performing similarly to the real world cars.

That could explain why modifications to the car aren't going to work the way you'd expect as the base physics engine isn't necessarily accurate and so the modifications wouldn't be either.

Again, no hard facts to back that up, but I think that's the most likely reason why, say, the game may have no idea what the actual ride height of a car is, or why there's no clutch / launch control. Or why the GT-R is so amazingly good. It's one of those cases of slow as molasses on paper, but great on the road. I really doubt that the ps3 (or any other console or home computer) has the extra processing to take into account all the minor nuances of the design of that car that when added up make it one hell of a performer. The only logical solutions I can think of are that either the performance data of the real life car is directly used to set how the car behaves in the game (And so what tires or ride height or even weight or hp of the vehicle doesn't matter), or that the basic stats of the car are tweaked so that it gets near the performance of what the real thing does. Either way it's gonna make any modifications not work quite right.

 

[Tenacious D]

I haven't seen any Americans make that claim (they're experts at everything, could you give some examples?

You have to admit that Americans are arrogant in different ways to those from the rest of the world. Just talk to any liberal, they know more than God.

By the way, fascism came to America in the name of "change." Twice.

As for the post asking PD to make a PC sim, there's no need.

The PS3 is more powerful than many gaming PCs already, even with the skimpy ram and old fashioned RSX GPU.

PC racing sims sell a paltry fraction of what a console game will sell.

However, Both Gran Turismo and Forza are striving for this level of realism, and perhaps in the sequels, we'll have that. And with the wealth of cars both games offer, PC sims seem miniscule in comparison, and have a much narrower performance region. I know many gravitate to the 1000hp monsters and race cars, but I really get a kick out of racing my Supra, realistically, without having to spend the small fortune to do so in real life.

 

[MTLclllutch]

wow you guys are really going at it. my fav part is where you stereotype an entire nation off a forum ahahaha

 

[mikeonthebike]

That's an example of the math, though it's not all that accurate. Real time simulators can't process the actual math in most cases.

But the point was, having experience behind the wheel is no prerequisite for this discussion. Someone who doesn't even know what a car is could figure out how one behaves from physics alone.

That is an utterly ridiculous statement.

AGREED

It's something that every auto (or any type of) engineer does. You model something as a system, specify an input, and get an output. I hope you don't think that manufacturers build a car before they know what every part of its design is supposed to do.

The difference between an orgasm and a car is that we have equations for the latter. The former isn't understood quite as well.

Prototype build then huge amounts of real world testing and refining with feedback from high level test drivers.

NOT ACCORDING TO THE GUY ABOVE YOU

No you don't need drivers or feedback. Apparently.

this statement is OBVIOUSLY SARCASM I got it. simple reading comprehension

According to who? Validation is another important part of design.

But all of this strays from the point. There's no need to have experience behind a wheel to understand cars.

WAIT A SECOND you just said "I HOPE YOU DON'T THINK A MANUFACTURER BUILDS A CAR BEFORE THEY KNOW WHAT EVERY PART IS SUPPOSED TO DO. Then it was stated that "Prototype build then huge amounts of real world testing and refining with feedback from high level test drivers"

Somehow you agree with that statement then make the statement in RED above.


I'm sorry but reading about something cannot make you know what it feels like. It can give you ideas on how it's done, but doing it is a whole other story.

You are not going to read a book how to weld and then go weld perfectly, it's takes time and practice and patience. You are not going to read a book about snowboarding and the physics involved and hop right on a board and fly down the mountain carving at breakneck speeds or clear some huge tabletop jumps or drop a nice cliff line.

IT ALL COMES FROM EXPERIENCE.

You guys (not to the ones that make real valid points in here and not everyone quoted) think if you argue enough and tell other people they are wrong enough and they finally leave the thread totally pissed off you feel as if you have won something or this validates your argument or debate.


Wise men don't need advice. Fools won't take it.
--Benjamin Franklin

 

[mikeonthebike]

You have to admit that Americans are arrogant in different ways to those from the rest of the world. Just talk to any liberal, they know more than God.

This is hardly the time or place. This discussion if you can call it that has gone through all sorts of stupid and it doesn't need to get political, against or for any country.

 

[Exorcet]

WAIT A SECOND you just said "I HOPE YOU DON'T THINK A MANUFACTURER BUILDS A CAR BEFORE THEY KNOW WHAT EVERY PART IS SUPPOSED TO DO. Then it was stated that "Prototype build then huge amounts of real world testing and refining with feedback from high level test drivers"

Somehow you agree with that statement then make the statement in RED above.

Of course. Because they don't contradict. Notice the word suppose, chosen because it best describes what I'm trying to say.

Manufactures know what everything is supposed to do, because the physics that they used to make the car tells them this, however the real thing maybe a few percentages off from that. That is where validation comes in. Makes sense right?

I should also add that validation isn't just driver feedback. It's using the already mentioned math to check yourself. Including measuring horsepower, drag, weight, temperatures, suspension travel, etc.

So, I say again. Having experience behind the wheel of a car is not necessary since realism can be deduced from physics. In fact going on real world experience without knowing the physics is a pretty bad thing to do because it means that you don't know if you were even driving the car properly.

I'm sorry but reading about something cannot make you know what it feels like.

It can if it's descriptive enough. But what I'm saying goes beyond feel. You can calculate what a mass on eight springs (remember, tires are springs) will do when it goes over a bump of a specific shape, among other things. If GT5 does not match up to the real math well enough, it's doing something wrong.

 

[kimi123]

Wow xenophobic some what? No one is claiming it's the best because it's American and I'm American...Nascar drivers and IRL drivers all use iRacing and swear by it, so not sure where you've got your info. I never said race teams use iRacing any where in my post, sorry to break it to you but you twisted my words. I said (to paraphrase myself) that if Gt were this grand simulator like you've made it out to be then why isn't every team going out and buying it to train their drivers on? Why aren't these F1 teams asking PD to build their simulators if it's so good? I'll tell you why, cause the ability to build a true sim that can grow isn't something PD or T10 or even EA know how to do.

Simbin/ISI which work on rFactor have their tech in the RBR pro simulator, rFactor is probably the closest challenger to iRacing. So if GT is better then why not use it rather than rFactor a game quite similar to iRacing? Also many would say that iRacing is better. The FW31 seems more real feel or typical of what a car would do at fast speeds more so than GT5. Also Simraceway is another one to look out for as a more authentic sim. I'm a fan of GT but at least I can be honest when a game isn't as real as others, not sure why you have to get defensive.

Also troll doesn't have one universal meaning and I find it highly ironic that you claim I act as if I know everything, yet you're rash enough to come out the woodwork doing exactly what you accuse me of. Hypocrite much?

iracing has good marketing because few drivers and many amature race drivers have tried it and probably game on it few times every now and then.

You should know that math exists which can predict the behavior of a vehicle. I don't think anyone here is going to write a simulation to show how car X handles to 99% accuracy, but that's OK. Instead they can apply general physics knowledge and also seek out sources that explain how that car behaves.

No. We do not want generic physics engine. There is a difference between a game having good physics and game having realistic physics. Everyone want to drive cars on tracks that race drivers do in real life.

 

[sinbad]

It's wrong to say that validation of a "sim" should only come from "feel"and familiarity which you associate with some real world experience.

"Feel" is subjective, and often when we look at a screen whilst sitting on a stationary chair, what we expect to happen, and what actually should happen can differ significantly. You wouldn't be able to immediately drive a real car via radio controls, using wheel and tv screen with a camera on the car, in exactly the same way you drive it in real life. Certainly not everyone would be able to switch from one to the other.

Perceptions of realistic often miss the mark, even sometimes those from people with real world experience. For some people a simple misjudgement of speed is the root of the problem. Many people have a "the car has no grip" reaction to sims, but you've got road cars on road tyres pulling well over 1 lateral g. How much grip do you want?

No doubt this is why some people insist on using racing tyres all the time in gt5 too.

The point is that something can feel right, but be quite, quite wrong. Lots of sims have fudged physics with canned effects to give a "feeling". Whilst a detailed, accurate, mathematic approach should invariably yield something that is close to right, but will still only feel right if you adjust your perceptions/expectations appropriately.

 

[JDMKING13]

It's wrong to say that validation of a "sim" should only come from "feel"and familiarity which you associate with some real world experience.

"Feel" is subjective, and often when we look at a screen whilst sitting on a stationary chair, what we expect to happen, and what actually should happen can differ significantly. You wouldn't be able to immediately drive a real car via radio controls, using wheel and tv screen with a camera on the car, in exactly the same way you drive it in real life. Certainly not everyone would be able to switch from one to the other.

Perceptions of realistic often miss the mark, even sometimes those from people with real world experience. For some people a simple misjudgement of speed is the root of the problem. Many people have a "the car has no grip" reaction to sims, but you've got road cars on road tyres pulling well over 1 lateral g. How much grip do you want?

No doubt this is why some people insist on using racing tyres all the time in gt5 too.

The point is that something can feel right, but be quite, quite wrong. Lots of sims have fudged physics with canned effects to give a "feeling". Whilst a detailed, accurate, mathematic approach should invariably yield something that is close to right, but will still only feel right if you adjust your perceptions/expectations appropriately.

Great post man this is so true, I was getting blasted for telling somebody to use the comfort tires to emulate the car in stock form. They insisted I have no clue, and the cars drive as if they are on ice with comfort tires. I don't know it more what you said.

 

[peter_vod69]

It's wrong to say that validation of a "sim" should only come from "feel"and familiarity which you associate with some real world experience.

"Feel" is subjective, and often when we look at a screen whilst sitting on a stationary chair, what we expect to happen, and what actually should happen can differ significantly. You wouldn't be able to immediately drive a real car via radio controls, using wheel and tv screen with a camera on the car, in exactly the same way you drive it in real life. Certainly not everyone would be able to switch from one to the other.

Perceptions of realistic often miss the mark, even sometimes those from people with real world experience. For some people a simple misjudgement of speed is the root of the problem. Many people have a "the car has no grip" reaction to sims, but you've got road cars on road tyres pulling well over 1 lateral g. How much grip do you want?

No doubt this is why some people insist on using racing tyres all the time in gt5 too.

The point is that something can feel right, but be quite, quite wrong. Lots of sims have fudged physics with canned effects to give a "feeling". Whilst a detailed, accurate, mathematic approach should invariably yield something that is close to right, but will still only feel right if you adjust your perceptions/expectations appropriately.

So what if the physics are "right" but it doesn't feel right? Does that make it a better sim? Or should we just trust the mathematicians and ignore countless hours of experience?

 

[jtv90069]

There are generally two classes of physics engines: real-time and high-precision. High-precision physics engines require more processing power to calculate very precise physics and are usually used by scientists and computer animated movies. Real-time physics engines—as used in video games and other forms of interactive computing—use simplified calculations and decreased accuracy to compute in time for the game to respond at an appropriate rate for gameplay.

In most computer games, speed of simulation, as well as so-called “gameplay” (simplified, it can be explained as “how fun the game is to play”), is more important than accuracy of simulation. This leads to designs for physics engines that produce results in real-time but that replicate real world physics only for simple cases and typically with some approximation. More often than not, the simulation is geared towards providing a "perceptually correct" approximation rather than a real simulation.

source: http://en.wikipedia.org/wiki/Physics_engine

No. We do not want generic physics engine. There is a difference between a game having good physics and game having realistic physics. Everyone want to drive cars on tracks that race drivers do in real life.

We may want accurate physics but the math required is too complex to be done in real time on the PS3 and keep all 1000+ cars and have upto 16 different vehicles on track at the same time. If GT5 had only 20 cars total they would likely be able to have a lot more detailed physics engine maybe even rivaling iracing and rfactor although i think that even those games would still use real time physics where rfactor pro would likely use the high precision type (and require much more processing power for limited amount of vehicles)

the generic tyres and modifications are necessary, to accurately model the hundreds of real world tyres and mods for each car would be massively time consuming and expensive. so using a simplified generic approach (which it has done since the first) is fine.

^^ include physics

 

[Exorcet]

So what if the physics are "right" but it doesn't feel right? Does that make it a better sim? Or should we just trust the mathematicians and ignore countless hours of experience?

If the math is right, it's going to feel right. Feel isn't all that useful of a thing to base a simulator on unless you define feel objectively.

We may want accurate physics but the math required is too complex to be done in real time on the PS3 and keep all 1000+ cars and have upto 16 different vehicles on track at the same time. If GT5 had only 20 cars total they would likely be able to have a lot more detailed physics engine

The number of cars modeled has very little to do with accuracy. The only thing a physics engine does it take an input and then give an output. How many different inputs you have is not a factor as long as you can take those inputs, and since most cars are fairly similar, the physics engine should be able to take most cars.

There are things wrong with the physics that indicates that the lack of realism isn't always due to hardware. Examples being the aerodynamics and apparently damped weight transfer.

 

[jtv90069]

The number of cars modeled has very little to do with accuracy. The only thing a physics engine does it take an input and then give an output. How many different inputs you have is not a factor as long as you can take those inputs, and since most cars are fairly similar, the physics engine should be able to take most cars.

The number of cars has everything to do with it. Having to process x amount of data to output the physics of one car then it makes sense to me that it would take x times 16 amounts of data to output the physics required for 16 different vehicles at the same time, similar or not.
The more realistic the physics you want the more data that will need processing.
With the PS3's 256 mb RAM there is a limited amount of space to hold all the data required, not just for the physics but everything needed for the game.