“ are rfactor2 physics broken” video

Yeah more roll at the rear means more load shift at the front, which means rear end has a benefit. Having more front roll resistance is good for more responsiveness of steering. Also as more roll happens at the rear, and more tire unloading happens at the front it has good effect for attacking kerbs as unloaded inside tire hits them less hard. It is interestign how old cars often used to have such stiffness difference per front vs rear, that inside front tires would lift up completely, allowing very aggressive kerbs cutting (also looking awesome )

Chassis roll balance is important, so is the overall chasssis roll. Also must not forget that spring and dampers also take big part in dealing with roll motion.

Also tires is the main influence, if tire bites in, then chassis roll. If tire slips, chassis does not roll, then it also needs less roll stiffness....

 

You could hardly be more wrong than what you just wrote.

Okay, the tire model is probably a little too permissive, so check it out. But coming to talk about iRacing which is on the other hand clearly too punitive, LARGER too punitive is stupid, iRacing is far from realistic in terms of the glide, the difficulty is artificial and absolutely not credible, it even ends up disgusting people who want to drive other than as a tightrope walker, there is only one way to drive : soft and without attacking, not to mention the fact that the slightest contact with another car is immediately an uncontrollable accident, so when you also touch grass (well, snow at least a priori), it just becomes anything!

rFactor 2 is much more credible and realistic, at least once the tire slips because before, both have exactly the same behavior.

In short, comparing the difficulty with iRacing is the worst thing you can do.

Both have great physics, I think rF2's one is better but not way better. Tire model on the other hand is way better in rF2 than iRacing. Physics in ACC are impressive as well.

EDIT : oh, and actually iRacing is more grippy than rFactor 2, like you said MOTEC reveals it, I tested it, again and again and again. Difference is over the limit, iRacing is slippery as hell, even on wet surface real cars lose less grip than iRacing in perfect conditions. You probably learned to handle iRacing's flaws so when you try something else you think it is too easy, reality is iRacing is not realistic at all when the car starts to lose even 1% of grip, what MOTEC shows is that even a single % lost on a single wheel provokes a full glide of the four wheels, how logical is it ? It is not difficult, it is retarded. Check this very thread, earlier, I posted a video of the iR-01 losing grip on ONE front wheel, the least used in the situation, the car wasn't turning because this wheel was sliding...seriously, how logical is this ? And the wheel never got grip back despite I wasn't breaking anymore, isn't it clearly a clue that iRacing is over slippery ?

 

Physics of ACC sometimes makes you want to cry

 

@Slip_Angel you're safe with the plugin mate. Though reading this thread I could have had a lot of fun if I faked some values

 

Well, putting speculation apart and talking about what we do know is wrong*, here is what I kind of think is going on with the tires:
First, I think there might be some flaw with tire deformation, and no matter how much you compress it, the middle of the tire never "rises". So compressing the tire just spreads it wide, making the contact patch bigger.
Then, once again possibly due to "flawed" deformation: more deformation does not mean more heat generation, as happens in real life. I do think tires can respond properly to temperatures (and most do), they just don't respond to stuff that rises the temperature, EXCEPT SLIDING. This makes me think the sliding itself is not too permissive, just that, since there aren't other heat generators in play, JUST the temperature from the sliding is not enough to make the tire give up as early as one would expect.

I would love to know you guys's take on it.

I'd also like to know if anyone has any theory for what might be wrong with aero (if anything)

 

I am not sure if such generalizing statements make any sense. Most of the statements are based on aliens running GT cars in rF2 vs. other GT cars in other sims. You have cars in rF2 that would lead to people roasting their tires if they applied full steering lock, especialy FWD tintops or the FISI and Marussia. To make any feasable judgement you need a very large database of consistent laps driven by a very competent driver and not a video of a single hotlap.

I am not sure if ACC physics are impressive tbh. They are certainly better than in AC but there are times when the cars feel very disconnected, wich also relates well to what Niels Heusinkvelt pointed out in his video. First of all there is quite a bit of input delay. It could very well be that aliens would drive in insane ways if they had the same confidence that they have in rF2. You just don't feel the limit as well as in rF2 wich makes it very difficult to compare the two games as you sometimes even simply have the fear of hitting a curb or just turning in a bit too fast as cars rotate into endless spins with zero feedback. Then there are other issues with the tire pressures and curbs, wich aren't quite the same issues like in rF2. But that's another story. Still a very good sim though, don't get me wrong.

Regarding iRacing: I only tried the MX5 during my 3 month trial time and all I can say is that it was tremendous fun. Granted I have no idea how the other cars drive like in iRacing but I had no issues feeling the limit nor other effects that would make me put in the iceracing category and as a big fan of Empty Box I see people do the wierdest things in that sim that they wouldn't get away with in reality. Still a fun product from wich other devs can learn alot.

To sum it up: I think there are certainly cars that are good and bad across the different sims, but I would never go as far as "Are sim XYZ physics broken?". I mean, seriously? Those devs are allways going to look for ways to improve their titles or why are we constantly seeing updates to physics and tire models across the different platforms? There is no dev sitting back like some lazy idiot drinking Tequilla on their golden Yacht. Why are we seeing constant updates during the Holidays while we are enoying (or maybe not ) the time with our families?

 

On contact patch load distribution and grip, you may not be aware of this: https://www.studio-397.com/2017/12/6512/

I think it's easy to overestimate the effect of low pressures, especially when they're still within normal setup bounds. In terms of grip I think there isn't much wrong, or haven't seen any reason yet to think otherwise.

Heat from low pressure, again - real teams in general don't seem to feel a need to raise pressures to control overheating (or prefer to keep the grip gains and deal with heat in other ways), so I think in comparison to other heat-generating factors it's probably quite minor.

Temperature vs grip: this is a simple parameter in the TGM file. It's not part of the physical model. For a particular tyre it might be too forgiving. For DLC we can't see it (spotting a trend here?).

Aero: I've seen knowledgeable and skilled rF2 drivers talk of unrealistic aero behaviour especially in the GT3/E cars. We've seen cars somersaulting over slight crests and aero imbalances in the draft, things that were wrong and were picked up because they're noticeable. Aero isn't easily measurable like mass or weight distribution, suspension geometry, etc, and for me easily stands out as an easy thing to get wrong or inadvertently create a time bomb in specific scenarios.

 

Physics of any simulator ever make me want to cry sometimes, because guess what, all simulators have flaws

 

"all simulator have flaws " ; "it's just a game " ; "its same as braking rules in real life "

Sentence like these are not going to help in progress.

 

and posts like that aren't either

 

Telling the truth is never detrimental.
So we have completed the loop.

The difficulty for S397 will be to close these holes in their simulation using physics calculate effects that happen in the real world, and not just a patch applied over the hole. So for starter they will have to identify which differences between reality and simulation trigger that behaviour, then they need to simulate it and insert it in the physics without issues. Is for this reason we still don't have aquaplaning, because they clearly stated that a workaround was out of question, when they will be able to simulate it, they will include it.

 

All in all, I find the physics in ACC to be more than correct, I even find it good. I've never seen anything as crazy as iRacing with this sudden, mind-boggling glide.

My biggest concern in ACC is the FFB which is, I find, far from being rFactor 2 level (well and good) but even iRacing which is only working at 60 Hz !

 

I don't think that one of these 3 simulations really has faulty physics, at least not outrageously. Each one is relatively close to reality (as close as possible) and I don't see anything really crazy when I drive, except the sliding in iRacing which is clearly too sudden and important, it feels like sliding on ice but as long as we don't slide, I find the physics very good.

Of course, it's just a feeling based on my experience of reality, I don't have a huge experience of racing vehicles (2 courses done on GT4 level sports cars and a few laps in Spa with an Audi R8 LMS in excellent track conditions, it feels light). None of the simulations perfectly reflect reality, which is normal, there are too many missing elements but we are still getting close to it seriously.

ACC has a concern of philosophy regarding the FFB, they don't want to "add artificial effects to compensate" but even the real and logical effects are not present, which is a problem, like the weight of the steering when braking (a 1.5 ton car braking from 280 km/h to 100 over 150m would have no reaction in the steering wheel? Newton, Einstein finally all the scientists of the world were visibly wrong), or the movement of the steering wheel when the rear slips but not the front...in short, crucial information is missing which makes the game "disconnected" at times, indeed.

 

Of course, but sentences like that also remind us that we shouldn't look for absolute perfection either, we can be satisfied with a 95% correct behavior knowing that the remaining 5% are not incredibly bad either, just not perfect.

I understand your approach but we should be a little more flexible with what you call imperfections. When it's openly bad, you're right to point it out but when it's marginally different from reality, you're not going to get more royalist than the King either...

 

but RF2 isn't in "good state" . these issues are clear and can improve physics much better. this is why sentence like those are not right for current situation. the room for improvement is much clear in RF2 case.

 

Aaand the circle is round again. "nobody said that the physics should not be improved"

 

Lol.

 

The question is the priority that S397 should give to the different aspects of the simulation.

Shouldn't physics be more important than the graphics engine (which has been improving incredibly slowly over the past 4 years) or the UI (which overall is very disappointing and has been improving extremely slowly over the past 4 years) ?

This is where the folks on this thread disagree.

And again, if a minority of people show a high interest in improving physics (as has been the case for the past 4 years), this does not encourage S397 to understand the importance and the urgency of improving physics.

Users should have a role to play, and S397 should be listening outside of Discord #&!#!

 

If the controls in the ACC keep their way to the gamepads, the physics will be messed up.

 

It is undoubtedly a question of nomenclature, in French we say that it is in "good state" when overall there is nothing clearly penalizing, it is more than acceptable in other words. I even find that rFactor 2 has a very good physics with indeed some problems but nothing very serious, not like iRacing for example which has a sliding problem that is definitely penalizing.

We all agree and admit that there are some slight problems in rFactor 2 but it seems that according to you it's impractical, it seems to me that we're not there either.