FM4 Physics - Houston, We Have a Problem

[Wolfe]

The other night I was racing with my newly acquired Lancia Stratos when I noticed something strange -- "This is really, really easy."

As I toyed with the car's dynamics, an issue came to light: we have no lift-off. Lift-off oversteer, that is.

Alright, the effect is certainly present to a degree, and power oversteer is so prominent that it's no wonder nothing seems amiss. But upon digging deeper, I discovered an isolated and surprising flaw -- one tangible enough I felt it needed the attention of its own thread, rather than a passing mention in the FM4 vs GT5 physics thread.

Try this experiment:

1. Take any RWD car you like. It can have a rear-biased weight distribution (MR/RR), narrow crappy tires, anything you'd expect in a dangerously tail-happy car. The RUFs have pretty sticky tires, but there's also the rear-engined DMC-12.
2. Equip a Race (2-way) Differential. Add Race Weight Reduction if you think it'd make the car easier to throw around.
3. Go on a Test Drive (to adjust settings in-race) in an open area, and set the differential to 100% accel and 100% decel. A fully locked differential, or close to it.
4. Accelerate up to the redline of 2nd/3rd gear while turning -- mind the power oversteer, we already know that works fine, so neutral cornering -- and lift the throttle suddenly.

Lift-off...UNDERsteer?? I'm not exaggerating. A 2-way differential should be a sure way to make a RWD wild in corner entry. It's preferred by drifters for this reason. For the sake of citation:

An inexperienced driver can easily spin the car when using a 2 way LSD if they lift the throttle suddenly, expecting the car to settle like a conventional open differential.

That's not what we're getting.

Now set the differential to 0% accel and 0% decel. Totally open. Try the same thing again. Suddenly lift-off oversteer is back, if not particularly strong. Fairly close to realistic for an open differential, at least for balanced cars. MR and RR cars, I'm not so sure.

At the very least, this provides a lesson for Forza tuners -- to combat understeer with lift-off oversteer, use an open "decel" setting on a 2-way diff. Similarly, you can increase the "decel" setting to add stability. The "accel" setting only affects power oversteer, so it's unrelated to this phenomenon.

This is not realistic. With the support the community has received from Turn 10 so far, I think it would be worth calling their attention to it. I bet plenty of people are feeling their tuned car understeer a bit on corner entry, without knowing the real reason why. I felt it, and I had been equipping 1.5-way differentials for their real-world dynamics without a second thought.

I don't know if the differential simulation is the only thing to blame. Perhaps there's a lack of tire "underspin" (when a driven tire spins slower than it would roll freely). It kind of masks whether we're getting proper weight transfer out of MR/RR cars, too. The Stratos I was driving did not have a 2-way differential equipped.

Don't take my word for it. Try it for yourself, and share your thoughts.

 

[buttsneeze]

I've noticed plenty of lift off oversteer in FWD and AWD cars. It has a lot to do with suspension tuning.

 

[Luminis]

Dunno, might be because I'm using the clutch, but I'm usually setting for a decel setting around 40 percent. Higher values will often result in lift-off oversteer... At least to my experience. Not as much in RWD cars as in others, but then again, I don't know which amount of lift-off oversteer would be realistic It was enough to invalidate some nice laps around Fujimi Kaido, though

That said, I've found that most of the road cars are rather understeery thanks to their default setups (which seems realistic) and once you've added a race suspension, things get a little blurry as the handling can be influenced dramatically by the suspension tuning, as buttsneeze mentioned.

 

[buttsneeze]

If anything's off it's the suspension physics. There's something very fake and screwy about the way cars go over bumps, especially in extreme situations (like hitting curbs hard or going airborne). It's like they have too much damping. I've never seen a car in Forza bounce after landing a jump but real cars do all the time. Cars don't get unsettled over bumps as easily as they should either. NFS Shift 2 is a good example of excellent suspension physics (though the tire physics are awful). You really feel the consequences of an overly stiff suspension when you're constantly going airborne and losing grip. That just doesn't happen in Forza.

 

[X-Ray-Cat]

I agree, there are aspects of different physics modules that arent perfect. But that said its a game and no game is, I am atleast greatful that the imperfections in this arent as many or as drastic as the imperfections in the alternatives. I expect with each new release things will get better....FM3 they concentrated on the tire model. FM4 they rebuild and perfect the tire model and started on the suspension model. I expect in 5 they will spend more time on the suspension model

 

[Killer46]

Um, what?

More lock on deceleration should increase understeer, not increase lift off oversteer.

 

[imported_JJ72]

being a karter for a long time (which has 100% locked axle all the time), locked rear diff doesn't promote oversteer at all.

Locked diff act against yaw movement, the result will almost always be understeer, in short your assumption is fundamentally incorrect.

take a look at this link:

http://diy-racetuning.net/Forza-Motorsport-4/DIY-limited-slip-differential-tuning.html

There is lift off oversteer alright, ever tried the Rossion Q1?

 

[VXR]

Heck, there's lift-off oversteer in something mundane like the Ford Ka. Had a brilliant moment on Maple reverse during a race a few weeks back, where I lifted to tuck the nose, the short wheelbase caused it to get sketchy and it rotated on me. Managed to stop it from hitting any walls, but forgot to watch the replay after the race. I'd have saved it otherwise.

 

[imported_Tiddy]

I was pretty sure this was the exact reason you added a 2 way diff to a car? To help prevent lift off oversteer?

 

[Melting Brains]

being a karter for a long time (which has 100% locked axle all the time), locked rear diff doesn't promote oversteer at all.

Locked diff act against yaw movement, the result will almost always be understeer, in short your assumption is fundamentally incorrect.

take a look at this link:

http://diy-racetuning.net/Forza-Motorsport-4/DIY-limited-slip-differential-tuning.html

There is lift off oversteer alright, ever tried the Rossion Q1?

Hey, thanks for that great link JJ!

And on-topic, I really haven't noticed much reaction from different diff settings... I tried it with a few cars on test drives and I couldn't tell much difference between a 0% and 100% setting for either front or rear...

 

[Spagetti69]

Hey, thanks for that great link JJ!

And on-topic, I really haven't noticed much reaction from different diff settings... I tried it with a few cars on test drives and I couldn't tell much difference between a 0% and 100% setting for either front or rear...

Do you use telemetry?

Call up the Friction tab and monitor the inside driving tire.
I've managed to dial it out.

Accelerating out of a corner RWD. If my inside rear tire spins on accel I dial it back. Just check your RB's aren't to stiff either because that will lift your inside too if too excessive.

 

[Wolfe]

First question I have for anyone with real-life experience in high speed driving of any sort -- do you drive like you do in a videogame?

This is an important and overlooked question. When playing a game, we always want to maximize horsepower output, stay in the best gear for when we apply throttle, and spend as little time coasting or on half-throttle as possible, because we can always brake later into a corner without endangering ourselves. We also don't feel any G-forces or vibrations, and mechanical sympathy is nil.

This promotes very high RPM, and abrupt braking, throttle, and steering inputs, whether you use a wheel or a controller.

Any developer that understands this also knows that mainstream gamers are oblivious to the changes in handling dynamics this will cause. Regular people drive at 65-90mph all the time -- it's not scary or uncontrollable at all, and that's what they expect. Regular people don't drive at 65-90mph at 6000+ RPM in third gear. So it's entirely possible Turn 10 actually incorporated this flaw on purpose.

For a mid- or rear-engined car, such rough driving is a surefire way to find yourself upside down in a tree. Cars like the old Porsche 911s or even the front-heavy Cobra didn't get their reputation from the frankly mild oversteer we get in this game. Modern cars tuned for understeer and with very wide rear tires are better, but again, almost no one drives like in a videogame.

Because of G-forces, vibrations, and mechanical sympathy, not to mention the threat of bodily harm, all but the most hardened of hobbyists/racecar drivers keep the RPM relatively low and prefer to drive smooth and a bit slower, rather than harsh and a bit faster. They might not even realize they're doing it.

Even with smooth driving, it doesn't take a whole lot in real life to swing the tail out. Not as much as Forza requires. I once watched a friend of mine with a late SW2x MR-2 (after they were tuned to understeer) spin out on a tiny autocross course at less than 50mph. Probably less than 40. But he was inexperienced, a gamer, and racing against the clock, so I'm sure he was in the top of 2nd gear.

Here's a separate example:



This is not a handling trait to be taken lightly.

 

[Wolfe]

I've noticed plenty of lift off oversteer in FWD and AWD cars. It has a lot to do with suspension tuning.

There is lift off oversteer alright, ever tried the Rossion Q1?

It is there, and I said as much in the third line. I don't know offhand what differential the Rossion starts with, but an open decel setting (eg. 1-way LSD) certainly allows lift-off oversteer to occur.

However, there's no doubt in my mind that it's lacking.

Locked diff act against yaw movement, the result will almost always be understeer, in short your assumption is fundamentally incorrect.

A locked diff resists yaw movement but it is this resistance that can overwhelm the tires, particularly when weight is transferred to the front of the car. To explain for other readers, because the outside wheel covers a greater distance when cornering, it wants to roll faster, and because the inside wheel covers a shorter distance, it wants to roll slower. The locked diff obviously won't allow this, and so when you lift the throttle abruptly, the outside wheel is liable to "underspin" and break traction, if not both wheels.

A kart isn't a particularly great analog for a roadcar because the suspension travel and overall weight are so miniscule by comparison, not to mention the wider tires (with respect to diameter) which introduce differences in the contact patch and slip angles. It was my understanding that racing karts can bite hard if you're not careful with corner entry and throttle control -- I have no experience with that hobby but Live for Speed and even GT5 support this. And again...videogame driving.

More lock on deceleration should increase understeer, not increase lift off oversteer.

I was pretty sure this was the exact reason you added a 2 way diff to a car? To help prevent lift off oversteer?

This is the way it works in Forza, obviously, and JJ72's link is specifically written for Forza's physics. If the wikipedia quote in the original post isn't a good enough reference (skepticism of that source is certainly understandable), the reasons to use a 2-way differential in drifting are well-documented. If it only increased understeer, no one would be welding the differential on their 240SX to go drift.

As JJ72 pointed out, a locked diff will inhibit yaw. This causes understeer when trying to navigate a slow corner, but anytime the tires exceed their available grip a locked rear diff is only going to contribute to oversteer, whether on- or off-throttle.

Forza's decel setting appears to prevent the rear tires from breaking traction in the first place. This is part of the problem.

I do appreciate everyone's input. 👍

 

[rocketcaron]

First question I have for anyone with real-life experience in high speed driving of any sort -- do you drive like you do in a videogame?

This is an important and overlooked question. When playing a game, we always want to maximize horsepower output, stay in the best gear for when we apply throttle, and spend as little time coasting or on half-throttle as possible, because we can always brake later into a corner without endangering ourselves. We also don't feel any G-forces or vibrations, and mechanical sympathy is nil.

This promotes very high RPM, and abrupt braking, throttle, and steering inputs, whether you use a wheel or a controller.

Any developer that understands this also knows that mainstream gamers are oblivious to the changes in handling dynamics this will cause. Regular people drive at 65-90mph all the time -- it's not scary or uncontrollable at all, and that's what they expect. Regular people don't drive at 65-90mph at 6000+ RPM in third gear. So it's entirely possible Turn 10 actually incorporated this flaw on purpose.

For a mid- or rear-engined car, such rough driving is a surefire way to find yourself upside down in a tree. Cars like the old Porsche 911s or even the front-heavy Cobra didn't get their reputation from the frankly mild oversteer we get in this game. Modern cars tuned for understeer and with very wide rear tires are better, but again, almost no one drives like in a videogame.

Because of G-forces, vibrations, and mechanical sympathy, not to mention the threat of bodily harm, all but the most hardened of hobbyists/racecar drivers keep the RPM relatively low and prefer to drive smooth and a bit slower, rather than harsh and a bit faster. They might not even realize they're doing it.

Even with smooth driving, it doesn't take a whole lot in real life to swing the tail out. Not as much as Forza requires. I once watched a friend of mine with a late SW2x MR-2 (after they were tuned to understeer) spin out on a tiny autocross course at less than 50mph. Probably less than 40. But he was inexperienced, a gamer, and racing against the clock, so I'm sure he was in the top of 2nd gear.

Here's a separate example:


This is not a handling trait to be taken lightly.

I get your point and I understand you wanted to illustrate it with this vid but the guy makes a terrible mistake there by completely lifting AND downshifting (releasing the clutch and totally disengaging the drivetrain) at the worst moment possible. I will have to test it but replicating what he did should give similar results in FM4. And I'm not saying hes not a good driver, I do some lapping IRL myself and I'm subject to similar mistakes believe me!

With that said, I think you are right though, it is not perfectly implemented in every car and/or situation.

But I guess my track time makes me drive like in real life.. I will deliberately do a couple of warmup laps just for the sake of emulating reality as much as possible and I think FM4 makes you pay quite a bit for sloppy, careless ''It's just a game'' type of driving. Especially since the last update. This was a big gripe from FM3, the game was much more permissive.

 

[VXR]

Has the update altered settings for pad users on sim steering? Cause it took me a while to get used to the 458 Italia earlier and I'd driven it before - albeit in free run rather than a race - and something felt different.

 

[imported_JJ72]

This is the way it works in Forza, obviously, and JJ72's link is specifically written for Forza's physics. If the wikipedia quote in the original post isn't a good enough reference (skepticism of that source is certainly understandable), the reasons to use a 2-way differential in drifting are well-documented. If it only increased understeer, no one would be welding the differential on their 240SX to go drift.

As JJ72 pointed out, a locked diff will inhibit yaw. This causes understeer when trying to navigate a slow corner, but anytime the tires exceed their available grip a locked rear diff is only going to contribute to oversteer, whether on- or off-throttle.

Forza's decel setting appears to prevent the rear tires from breaking traction in the first place. This is part of the problem.

I do appreciate everyone's input. 👍

First thing first in the vid you posted, your friend downshifted in the middle of a corner, pretty much everything RWD will spin in that situation and that isn't because of liftoff oversteer. I concur the MR2 isn't any snappy as in real life (there are better videos of MR2 doing that online) but it is more a problem with individual cars.

secondly I am going to add a few links that is not Forza based for your reference:
http://strikeengine.com/limited-slip-diff-tech

http://forum.miata.net/vb/archive/index.php/t-168817.html

http://www.motoiq.com/magazine_arti...art-2-tuning-a-limited-slip-differential.aspx

the last one is the best. his description is the same as what I think, which brings to point three:

locked diff inhibits initial oversteer but once you break the limit of traction, it dramatized a slide, and in forza it is pretty much how it should be.

in a gripping situation, the forces of the tires wants to go in a straight line, hence the understeer effect, while the inside wheel being less loaded will give first if the limit of grip is reached. once one start to slide, the overall road holding ability at the rear drops and the outside tire will soon follow. (unless the car was initially setup to corner on 3 wheel of course.)

Now on a open diff car each tire has more freedom to regain traction at their respective wheel speed, letting it to settle easier. (assuming no throttle input at this point, since that will mess thing up big time) But on a locked diff they are more inclined to spin at the same speed, which makes them harder to get back in line, basically the more locked a diff the less likely the car will follow a natural line. it will tend to swing between extremities.

drifters wielding differential is just a ghetto way to get a locked diff effect for powersliding with low cost, and since their driving style forcefully and abruptly introduces oversteer, basically what we have been discussing does not matter to them, what matters to them is the consistency and length of the slide, which is more related to the acceleration side of the diff.

I will skip the part about real world driving vs game driving. because it all comes down to personal skill level and commitment, but I don't think I drive any differently in both realms, when I race in Forza I rarely drive at the absolute limit, but granted the sensation is diminished in the game and for inexperience drivers, they won't attempt the same driving in real life from the get go. on my own car I wouldn't push it since it's basically part of the family, but on a go kart though I can go all out without worrying about consequences.

 

[Wolfe]

1. That isn't my friend in the video. Hence "separate example." It doesn't matter if he made a noob mistake -- we're talking about physics, not driving technique, and if a real MR behaves that way when you make that mistake, MRs in Forza should do the same thing with the same mistake. In the game, you can throw your car around as violently and inappropriately as you like -- unless you break traction with power oversteer or the handbrake first, you will not unsettle it as easily as in that video. Regardless of the differential settings. And not just the MR-2.

Everything may seem fine because you generally don't try to drive like that, but it's a hidden safety net, and that detracts from the sim experience. It's practically an undefeatable assist. You need to drive "poorly" to see what I'm talking about. Note that it took me two months to discover something was wrong.

2. If you take one thing away from the "videogame driving" point, it's that we tend to race at the top of the tachometer. There's little reason not to. Racecar drivers frequently do this too, but they're professionals. The average enthusiast out for a "spirited drive" is unlikely to punish their car that way -- not if they want their powertrain to last. This is important because a car is easiest to unsettle at the top of its RPM range. Again, regardless of the differential setting.

What bothers me about the differential simulation is that with the same driver (me) performing the same maneuvers in the same car, an open rear differential is always more prone to lift-off/snap oversteer than a locked rear differential. I can't seem to get a locked rear differential to lift-off/snap oversteer at all...

This is where I would like to point out, JJ72, that I agree your last link provides the best description (emphasis added):

A 2-way differential will lock equally under acceleration or deceleration. This makes the car drive more like a fully locked "spool" or go-kart rear end. A lot of rally racers and drifters prefer this sort of feel because the wheels don’t want to travel at different speeds and will either cause the car to go straight or break traction and slide easier and longer. 2-way LSDs often have too much deceleration lock for road racing/grip driving.

You and I clearly both understand how a 2-way or locked differential can inhibit yaw and maneuverability. That's the "cause the car to go straight" part -- I'm not ignoring it. 👍 Where we seem to differ is in how likely it is for a 2-way/locked rear differential to unsettle the rear end. With respect to the quote above, I again call into question the idea that drifters or rally racers would prefer a differential setup that overwhelmingly promotes understeer. There's more to rally-racing-oversteer and drifting than booting the throttle and letting the engine do the work. There's also a red flag in the fact that a 2-way is preferred by real drifters, but an open decel setting provides the best oversteer control in Forza...

Each of your links highlighted a 2-way's tendency to resist turn-in on corner entry, but in my Forza tests I'm already turning from the start. I can stomp and lift the throttle all day, feint the other direction, even use the brakes, and if I don't give the game enough time to "build" power oversteer, I'll just keep driving in a big circle.

Unless I apply wheelspin, the rear end is pretty much planted. That's not correct. It's easy. It's "console sim." And from the way Turn 10 talks about this game, I don't think they wanted to settle for that.

 

[imported_JJ72]

Take a rabbit GTI, head to the skidpad (on benchmark ring), drive near the limit and suddenly lift the throttle, don't you get lift-off oversteer?

I am damn sure if you mess up your downshift into a corner you can lose a car in Forza too.

If not, something with your testing method has a problem. Because losing a car on entry is pretty common when I have cold tires or turned in too violently.

And what about "performing the same maneuvers in the same car"? you know exactly what differential setting he is running? and suspension tuning etc?

I have never heard about anything about a car being looser at the top of the drive range, except when you have really serious engine braking and that factors in, when you downshift really aggressively, and that does affect your car in forza.

I think both your observation and hypothesis has some problem, for one I don't feel what you feel (and seems most of the people so far don't?) and two your comparison method doesn't seem very scientific. I would be very surprised if you can find a real life video with someone running 30% rear dece-diff and not understeer like a bugger. Only drifters run that high and they use handbrake/shiftlock feint etc to unsettle a car.

the way you are testing, suspenion setting is a factor too, in real life people would run more aggressive setups to balance out the effect of the diff, if you just pick a standard car and throw a diff in it would give nothing but understeer.

 

[imported_Tiddy]

Yeah people weld diffs when drifting but that's because otherwise it's impossible as the inside wheel just spins. You don't just slap an LSD on a car and suddenly it drifts, A lot more work is required than that.

LSD's can be used to promote over or understeering characteristics in a car, it just depends what you're aiming for. Maybe forza got it backwards, I don't know?

 

[Dopplegagger]

Some folks are sooo desperate to find faults with this game, you can almost sense the excitement in the opening post by the OP!!!
Big dramatic thread title......

But alas he's been proved embarrassingly wrong and also proved to be left wanting with his knowledge of LSD and car physics.
The least you could do is research before trying to mock Turn 10.

Having 2 brothers who race cars and having driven them on private tracks, and having been the owner of an Alfa 75 RWD LSD track car and having had a V6 Alfa GTV with Alfas famous 'Q2' LSD (a must!) i can confirm that LSD in this game is pretty acurate in so far as its such an underrated mod (like in real life).

Yes not as exciting as bolting on a turbo, or and engine swap or ratcheting up the BHP, but a fantastic mod and once i realise its value in game i have sorted out a great many of my tunes and im no tuner.

Its about as good as i'd expect considering this is a game, which people sometimes it appears, forget.

Oh well... I benefited from his assumption. After reading that RWD cars tend to perform better with the 1.5 LSD instead of the 2 way, I threw one on my B class Mustang GT. It's still tail happy but this time there us a metric ton of warning signs that the rear is about to step out...

Honestly... I would have never tried to use a 1.5 LSD on that particular car but I am glad that I did.

Thanks OP. You inadvertently rule.

 

[t.o.]

Thanks OP. You inadvertently rule.

This made me laugh. You guys gave me some ideas too. I got a Stang that I don't touch anymore because it's ridiculous to drive, also my BMW's can be a handful but now I might have to take a look at this LSD.

 

[Wolfe]

Couple things to clarify...

1. Lift-off oversteer exists in the game but it isn't enough and does not work properly. Lift-off oversteer exists in the game. I admitted this in the third line of the first post, but the effect is mild enough that in some cornering situations it might as well be completely absent. I said there's "no lift-off oversteer" because it's so weak you can't utilize it for cornering like you would in real life.

2. The MR-2 video was nothing but a visual aid. I was never trying to recreate his exact mistake, setup, and conditions. I don't know why anyone thinks his mistake discredits the video, because if Forza wants to be "real" it should do the same thing with the same input. The laws of physics don't change based on driving habits or skill.

I have never heard about anything about a car being looser at the top of the drive range, except when you have really serious engine braking and that factors in, when you downshift really aggressively, and that does affect your car in forza.

Take your real car on a country road and go around a bend at 55mph in 5th/6th gear. Do the same at 55mph in 2nd gear. If you're not careful you're much more likely to get loose in 2nd gear.

Why? Positive torque/power will be much greater, obviously, because you're up in the powerband. But engine braking is more severe as well, because you're in a gear with greater torque multiplication, which works both ways -- this is why a novice manual driver can jerk the car pretty violently trying to get the hang of launching in 1st. It's a pretty simple but widely overlooked thing.

-------------------------------

On that subject, I believe I may have actually figured out the root cause of the problem.

It's pretty clear Forza 4 is programmed with a series of values representing the torque curve of each car's engine. I'm guessing the torque multiplication effect of gearing is simulated by applying a literal multiplier to those values, derived from the ratios of the transmission and final drive. What may have happened is Turn 10 programmed the physics engine to apply the torque multiplication effect upon acceleration, but on deceleration the game erroneously defaults to a 1:1 torque ratio in the transmission.

The torque curve of acceleration and the "torque curve" of deceleration are not the same -- the engine behaves differently. But if the deceleration "torque curve" is the same in every gear, it would help explain things.

While I was checking out that GTI, JJ72, I tested this theory by attempting lift-off oversteer with a variety of methods at 60mph in 2nd gear, and then doing the same at 60mph in 4th gear. I chose 4th because it's near or at a 1:1 true drive in most vehicles, so I could see if the lift-off oversteer was more realistic in that gear. I repeated the experiment with a variety of cars.

The severity and onset of lift-off oversteer in both cases (2nd and 4th) was near-identical in the GTI, and at minimum very similar in other cars. Some discrepancy is to be expected given that in 2nd and 4th the engine is at a different place in the RPM range (a different spot in the deceleration "torque curve"), but compared to reality the differences were minimal. Furthermore, the oversteer in 4th was pretty reasonable for that gear and speed.

This would explain the differential issue, because in real life, by the time you're up in 4th gear the torque of engine braking is relatively unlikely to break the tail end loose on a locked or 2-way rear diff -- it's no longer multiplied in the transmission. So if 2nd gear produces the same engine braking torque, you won't get much oversteer, especially since the tires have a lower velocity/inertia range to contend with.

 

[X-Ray-Cat]

Even with smooth driving, it doesn't take a whole lot in real life to swing the tail out. Not as much as Forza requires. I once watched a friend of mine with a late SW2x MR-2 (after they were tuned to understeer) spin out on a tiny autocross course at less than 50mph. Probably less than 40. But he was inexperienced, a gamer, and racing against the clock, so I'm sure he was in the top of 2nd gear

SW20's are notorious for lift-off oversteer. They are notorious for spinning, and thats probably about the only way to get them to spin under most conditions. Because as you mentioned, they are HUGE for understeering under power.

At any rate. With a 900deg wheel, and simulation steering. Everything feels about right enough for me to enjoy it.

 

[Wolfe]

I am damn sure if you mess up your downshift into a corner you can lose a car in Forza too.

I was thinking about this, and tried intentionally downshifting too far heading into corners, bouncing off the rev-limiter. I'd never really done that yet. It gave me the sort of oversteer I've been expecting, but it's not something that should require a mistake. I'm not sure it's the best "solution" either, particularly in races with damage enabled.

Gran Turismo 5 had issues with lift-off oversteer, at least with front-engined cars, and I remember the GT Academy replays highlighted the handbrake as an unrealistically effective fix. It was a magical "fix the physics" button, and I had already found that out myself (from the license tests) when those replays came under scrutiny. I'm not sure whether the handbrake actually makes you faster in this game too, but it's similarly helpful in a lame way.

At any rate. With a 900deg wheel, and simulation steering. Everything feels about right enough for me to enjoy it.

I agree, this stuff isn't a game-breaker. It's not going to be another FM3, where I stopped playing it after less than two months because the physics were so off.

Driving in this game is just so...easy. The excellent power oversteer and countersteer and all that just completes the package to make you feel like a hero. It's fun! But Forza won't be challenging Live for Speed anytime soon.

{end of post clipped}

 

[Dopplegagger]

^ You've brought up an interesting topic, and I would not back out of the discussion. Just pay the witless idiots such as myself no mind and continue to contribute as you see fit. There is nothing wrong with discussion.

 

[NA]

Driving in this game is just so...easy. The excellent power oversteer and countersteer and all that just completes the package to make you feel like a hero. It's fun! But Forza won't be challenging Live for Speed anytime soon.

About 15 years ago my brother-in-law did a project for his psyc class. One of the museums near his school had a full motion driving simulator at the time, and to simplify what the project was about, one of the things they did was divide everyone who they polled by age groups, and ask them questions along the lines of how much they enjoyed it, and how good a driver they thought they were. Age groups of 10-15, 16-21, 22-27, etc.
The interesting thing about this is that groups ranging from 16 to 37 tended to consider themselves "awesome" drivers as a large majority, more so than any other groups,(highest belief at 16-21, declining with age from there, with a noteworthy dropoff after the 32-37 group). Crashes occured most frequently at 16-21, declining with age groups at roughly the same rate has "belief in personal driving awesomeness", and "satisfaction with the experience" followed a pattern where it was highest at 10-15, and spiked again after the 32-37 group. However, durring the 16-37 demographics satisfaction seemed directly tied to whether an "awesome" driver had his/her belief reinforced by the simulator, with crashing being the number one reason for dissatisfaction, and "the simulator isn't realistic" being the most common given excuse for said crashing.
Granted, this is just a rough summary of a relatively small student project, but i'm betting it reflects something game makers have seen reflected in research of their own over the years. What it points to is that the prime sales demographics for console based driving video games is made up of people who want the illusion of realism, but don't deal with actual realism very well if it also happens to burst the ego bubble in the process.
In other words, console devs may imply that they are on par with PC sims, they may even outright claim it to be true, they may even say they've surpassed them, but chances are they arn't even trying to acomplish any such thing in reality. It's much more profitable to stroke the ego, and sell a myth of total realism, while stopping short of it in order to provide that "drive like a hero" experience that provides the real sales.

 

[t.o.]

About 15 years ago my brother-in-law did a project for his psyc class. One of the museums near his school had a full motion driving simulator at the time, and to simplify what the project was about, one of the things they did was divide everyone who they polled by age groups, and ask them questions along the lines of how much they enjoyed it, and how good a driver they thought they were. Age groups of 10-15, 16-21, 22-27, etc.
The interesting thing about this is that groups ranging from 16 to 37 tended to consider themselves "awesome" drivers as a large majority, more so than any other groups,(highest belief at 16-21, declining with age from there, with a noteworthy dropoff after the 32-37 group). Crashes occured most frequently at 16-21, declining with age groups at roughly the same rate has "belief in personal driving awesomeness", and "satisfaction with the experience" followed a pattern where it was highest at 10-15, and spiked again after the 32-37 group. However, durring the 16-37 demographics satisfaction seemed directly tied to whether an "awesome" driver had his/her belief reinforced by the simulator, with crashing being the number one reason for dissatisfaction, and "the simulator isn't realistic" being the most common given excuse for said crashing.
Granted, this is just a rough summary of a relatively small student project, but i'm betting it reflects something game makers have seen reflected in research of their own over the years. What it points to is that the prime sales demographics for console based driving video games is made up of people who want the illusion of realism, but don't deal with actual realism very well if it also happens to burst the ego bubble in the process.
In other words, console devs may imply that they are on par with PC sims, they may even outright claim it to be true, they may even say they've surpassed them, but chances are they arn't even trying to acomplish any such thing in reality. It's much more profitable to stroke the ego, and sell a myth of total realism, while stopping short of it in order to provide that "drive like a hero" experience that provides the real sales.

Not surprising. At 16 to 20-something people think they have the wisdom of Solomon while having a teaspoon of actual experience in life. And even some believe the experience isn't needed because of their greatness.

 

[Wolfe]

Dopplegagger -- Thanks. 👍 I made a lot of enemies on GTPlanet and YouTube (not that YouTube comments are anything to take remotely seriously) pointing out the problems with Gran Turismo 4. I really like Forza 4 and being able to race online with people you know is a big part of it, so I didn't exactly feel eager to press the issue.

About 15 years ago my brother-in-law did a project for his psyc class...
...
...It's much more profitable to stroke the ego, and sell a myth of total realism, while stopping short of it in order to provide that "drive like a hero" experience that provides the real sales.

Very interesting, thanks for sharing that NA. 👍

My own experiences support that observation. Following what you said, it's worth considering the case of Enthusia Professional Racing, for the PS2. Excellent game, and still my favorite console sim over FM4 for a number of reasons.

The physics are relatively low-fidelity, if you understand what I mean, so it's not a professional-grade simulator and lacks many details Forza has incorporated -- after all, it ran on PS2 hardware and devoted a share of CPU cycles to reasonably good graphics, a 6-car grid with AI, etc. -- but it had the right "building blocks" to feel incredibly intuitive and natural (screenshot from its now-defunct website). Having played the well-respected LFS for a few years already, I dubbed it "LFS-lite." Among the things it simulated well are the very topics of this thread -- lift-off/braking oversteer and the behavior of differentials. Funny, huh.

Accordingly...it was a complete and utter commercial failure. A lesson Turn 10 may have paid attention to, I'm sure. Its physics weren't the only poison -- wacky game concepts, a bland and "geeky" UI, minimal advertising, and the exact same release date as Forza 1 all sealed the coffin -- but they played a very big part.

Gamespot somewhat famously said, "if RWD cars in real life drove the way they do in this game, no one would buy them." It has to do with that expectation I mentioned before, how people tend to drive a real car in 5th/6th gear at 2000RPM, instead of 7000RPM in 2nd. Again, it's so often overlooked, but it makes a world of difference!

Many people say the game exaggerates oversteer, and to some extent that's fair, but a lot of other complaints were about understeer or overspeeding corners, and IMO it should all be taken with a grain of salt. The game wouldn't even cut players slack on RWD burnouts; unless you take off on a perfectly level surface, the rear end will start to swing around. Take a Shelby Cobra for a spin, and you'll know why it was considered deadly. A wheel was almost required; even other fans of the game on GTPlanet considered me odd for playing with the controller.

To conclude my semi-off-topic gushing, I still recommend Enthusia to any driving sim fan with a PS2. It's very cheap and not too rare to find online (yet).

Getting back on-topic, that fearsome uncontrollability with cars like the Shelby Cobra is what FM4 seems to lack, and this thread is my best attempt to figure out what's causing it. Come to think of it, FM4 has the Cobra, doesn't it? I'm gonna try it now.

 

[VXR]

Enthusia scores for some wacky car choices, but it's too typically Japanese in its approach. You just seem to be playing the dev's vision and not your own made up of decent building blocks. At least with Forza you can play it more in-tune with how you wish.

 

[SuperShouden]

As far as can tell from watching some videos, it seems that FM4 is simply FM3 with some minor tweaks to the physics. Cars are still incredibly easy to roll. (There's a particular turn at Sebring that flips cars for no real reason. I experienced it in FM3, and it's kind of silly to see it return in FM4. But seriously...I remember Rage Quitting over that corner a LOT.) The biggest changes appear to have been to the graphics. So...kinda like how FM2 felt like a shiny FM1...FM3 appears to be a shiny FM4.