“ are rfactor2 physics broken” video

Discussion in 'General Discussion' started by GTClub_wajdi, Dec 29, 2020.

  1. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Agreed. Time has to be taken into account and other priorities, but how much time does it really take to look into the tyres and make some parameter changes? I know Rfactor2 tyre model needs to go trough a process of generation wich I believe then produces the real time complex numbers that computers can't really do right now, and borrows the help from look up tables to produce a simillar result with much more effeciency of resources, right? It would not be the first time they recognized the tyre model was wrong... I'll the post and edit it here, but I think you're probably already aware of that if you've been an Rfactor2 for some time.

    I was just here going at it with some cars I might give a compilation and ask some people who are finding some nagging issues in some cars to drive, to see how they feel to them. There are some cara that to me and other people are quite more edgy and requires more precision from the driver and that seems to correspond to what can not only be seen in real life but even backed up by some data (slip angles, grip drop off, temperatures of tyre surface...). Anyway, yeah, I was having a blast and today I tried AC once again and man, no, there is no coming back to that and ACC is too limited on content for my taste. Also grip behavior past the limit there isn't also the best, but it doesn't seem to be possible to driver under the limit for the sim wich doesn't seem to correspond to real life, just like Rfactor2 under some conditions do.

    I'll have to take a look at that GT series. The thing is I believe I can drive Rfactor2 under "my limits", meaning I know I can exploit it in a way but I don't let myself do it, so it always feel quite good, minus the occasional time I made a mistake and then I take advantage of it to not lose distance of the AI... There are things one can't really control...

    I know some of the more realistic aspects of damage would help the issue, but what if the issue wasn't there in the first place? It will not completly get rid of it I don't think, as some hotlaps will still be possible to exploit like that. People who know how to drive fast like that might actually not be that great at driving propperly, but I don't think that's a majority.

    All in all, I want them to aknowledge if there are issues. If there are and they should be qualified enough to see themselves, forming a strategy that if possible to fix, they should fix. If it means no releasing some content for some time or not so much so be it. Better less and quality than a lot and poor quality... Just like PC2 case. Tracks and cars are nice, but if I had a race car and a simulation to chose that was really expensive, I would probably go with the car even if I would have endless options on a simulation that's even more realistic then Rfactor2 or any other professional one.

    I think Rfactor2 when 100% damage is on, doesn't take much to wreck the engine and some wheels falling off, but I know what you mean, I think RBR really did it hardcore.

    Oh and yes, that drivetrain simulation should already have been developed, no excuse for that.
     
    avenger82 likes this.
  2. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173
    I still fail to see where the exploit is and where the problem lies.

    All we have is some tiny videos from some pilots that are overstressing tires for a single lap, consequences aren't in the video. GT's soft tires are very grippy when at perfect conditions but :

    1) those guys are using very low pressure (which may be the only issue I can see so far) so under heavy load, the contact patch is MUCH bigger than with proper pressure
    2) those tires are very grippy...for some wear level, as soon as you reach something like 80 to 85% of tire wear remaining (and even before that point actually), tires are WAY more slippery.
    3) as it was stated before, those pilots don't have to suffer other consequences real pilots are, it makes a HUGE difference actually.

    Post a video of a guy emptying his full tank driving like that and yes, I would admit there is a problem. until then, it is just a pilot attacking like crazy for a single hotlap and even there, this doesn't even prove he couldn't do better without sliding like this.

    There are TONS of youtube video with real cars testing tires that are attacking like you can do in rF2, Pirelli tire test sessions for example, where you can see civil cars (so not even close to GT3 performance) that are sliding like hell with lower performance tires (but those tires have a wider range of usable slip angle)...they still slide way more and keep the car on track without any issue.

    ...but NO, some of you WANT there to be a problem so they FORCE the line and abuse the playback of some videos.

    No kidding, they even ended up explaining to us that the telemetry was bad, borderline falsified, that what we saw might seem correct but that the physics that used this data (which proves that they don't understand how it works, these data are a consequence, not a cause) was wrong and therefore the telemetry data was not a proof. SERIOUSLY ????

    Aliens are exploiting while racing ? Show it, not a single lap, a full race. Good luck, I tried, it is near to impossible to handle the car at the same pace for more than 3 laps on Spa or Silverstone, so Le Mans or Nordschleife, let me laugh...
     
    Last edited: Jan 24, 2021
  3. Stefan_L_01

    Stefan_L_01 Registered

    Joined:
    Nov 18, 2012
    Messages:
    595
    Likes Received:
    386
    I know from one who knows from someone. At what load? what tire size? what road? What camber?
    As a matter of fact noone has ever brought up some real data for real slick tires so far.
    The only real data I could find is mentioned here, an interesting report about a simulator comparing rfactor (1 ?) F3 standard tire model and some real data put into a pacejka tire model (Dallara 312 Hankook tire) (link at end of page pdf).
    https://repository.tudelft.nl/islandora/object/uuid:4cb0dc14-0ca0-4296-be74-51d0fd752fc5
    As said its rfactor 1 tire model but it shows up some model differences. Some major differences but of course the question is whether both models are build up correctly. In one of the image the dependency of wheel load to slip angle is shown. It should be noted that a pacejka model does not work well or at all below 100N. But the slip angle shows a minimum of roughly 4° at a wheel load of 2000N with a static load of 1200N (120kg) approx. The increase in slip angle with load is less for the real model. However the decrease of grip with more load is more pronounced for the real model.
    Then there are differences with camber and so on.
    So without knowing exact background of data throwing a number into the arena is useless. SlipAngle characteristics are highly dynamic and depend on many factors.
    But if you sum the rf model up, more intial slip angle, and more benefit with more load (turning creates more load, aero) and less overall loss in contact area will make your turn your wheels to much higher angles than the real model would allow for
     

    Attached Files:

  4. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173
    Ah because now the question is no longer whether the tire model is realistic, credible...it must now be EXACTLY what the real model proposes?

    Is this where we are, seriously? Do you want there to be a problem at this point? The bad faith seems more than obvious to me...

    No, they are not exactly the same as the real tires and you know why? Simply because in the real world there are not two RIGHTLY identical tires to begin with, they are all slightly different, very marginal but you will never find two exactly identical ones...so the virtual versions must be the exact copy of which one ?

    Go and ask Aristotelis, he worked for years with real GT3 teams and Pirelli, he is under NDA but he have the official data, ACC tire model is pretty close to real counterparts.
     
    Last edited: Jan 24, 2021
  5. Stefan_L_01

    Stefan_L_01 Registered

    Joined:
    Nov 18, 2012
    Messages:
    595
    Likes Received:
    386
    Why dont you go to him and ask him why you dont see overturning the steering by 7° and more is not seen by real drivers, especially not in fast turns?
    Someone should tell Mr. Hirschi from the protoyp video that he can go eau rouge flat just by turning the wheels another 5-10° more (steering ~: +45°?) (in case he is suffering front wheel grip
    )
     
    Last edited: Jan 24, 2021
  6. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173
    You know what ? I'll go and test the Porsche Cup or Oreca like he does, just to check :

    1) if there is a special setup involved (especially steering ratio)
    2) if it is efficient (will I gain time ?)
    3) what are the consequences...


    Will record it and post video with telemetry results (unless like some others you think telemetry is falsified) and maybe we will all agree on something, dunno ?
     
  7. ATQ

    ATQ Registered

    Joined:
    Jan 15, 2021
    Messages:
    52
    Likes Received:
    33
    I think he should ask Mika Häkkinen for advice.

    (Reference to Spa 2000, when Häkkinen overtook Schumacher)
     
  8. hitm4k3r

    hitm4k3r Registered

    Joined:
    Jun 15, 2016
    Messages:
    1,320
    Likes Received:
    3,121
    The big issue, and that's what I take from this discussion, is that you first of all need to know what you want to fix. You can't start changing values and parameters out of the blue to see what happens. This costs time and money. Then you need to build the tire as you allready said and start testing it in a way how Lazza described it. The testing is what actually takes alot of time because you need proper real data and see how your model fits after a huge amount of consistent laps. The data that you get from the manufacturer can have issues aswell btw. This is actually one of the big issues with enabling track and ambient temperature. All the tires that are in the game need to be rechecked and tested again. I guess it's obvious how big the task is. And then you have some hillbillies claiming that the whole game is broken. :D
     
    Nieubermesch likes this.
  9. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    I am pretty sure even if tires don't have to be the same as the michelin real life counterpart, I am sure tires have contraints that need to be balanced, you can't make any tire you want and behave as nothing that reality can be reproduced, so that's the problem. The tire not being totally realistic is one thing, some of us just wished some of the limits would be hitted sooner as we think we are seeing in reality. And please stop saying Slip Angel is stubborn and won't aknowledge some important stuff when you go on and on about saves and slides as if that's the same we are seeing. It's not about just controlling a slide, is when and how much man. I agree with you that driving like that for long wont work. But you know what? ACC I like to run sprint races... Guess what, someone with Rfactor2 would be abusing this and winning for sure.
     
  10. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Sure, let's not go to extremes and say the game is totally broken, but sometimes it takes only some little number to make abusing simulation possible and it certainly doesn't look that neat. I wish great success to Rfactor2 and Studio 397 team, I really do though.

    Some people are not even aknowledging that devs themselves (Marcel) has a tweet or something saying they are working in a heat model that seems to get rid of some of the sliding... So I don't know, while some of us might be too harsh some others just want to drive a sim and not want their immersion broken by a possibility of irrealism and so they will try to explain it way, and the opposite side will also exagerate how bad the problem is. We don't want too hard or too easy. All cars are easy and hard in their own way.
     
    Remco Majoor likes this.
  11. Havner

    Havner Registered

    Joined:
    Jul 14, 2020
    Messages:
    359
    Likes Received:
    356
    I'll repeat again, I know little about car telemetry, car physics and the like. I base my findings on comparison, and common sense, so please bear with me.

    But if we compare to ACC (not the complete games, but GT3 specifically). E.g. on Spa I can overturn the wheel most corners in rFactor 2 and gain time on each of them. In ACC I can overturn the wheel maybe on 2 corners and don't gain the time at all during this. In other corners (where I can do that in RF2) if I did that I'd end up oversteering.

    I'm not a very experienced driver. On SPA in ACC I do 2:20. In RF2 when driving like in ACC I can do similar. In RF2 overturning the wheel I do 2:16.

    And yes, the tires will not hold for two 1h stints (at least I assume they won't as they'll be at 20-40% at that point probably, don't know how they behave on such percentage, never tried). But they will hold up easily one 45 minute stint (60-70%). Which is a problem in my book.
     
  12. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Yes it is, holding 45 minutes of what already seems to be sketchy and it's no problem for some people? Cmon guys.
     
  13. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173
    Sorry but I can do exactly the same thing in ACC than in rF2, my question now is "is there any consequence ?". And according to my testings, you lose A LOT of grip beyond 85% of tire wear, which is pretty fast to reach when you slide like this (3 to 4 laps on Silverstone or Spa, figure). My lap time clearly shown this.

    In both game I have serious consequences like tire wear and ffb vibrations...and in no title I actually gain time doing this, can you elaborate a bit about this ? I still fail to find any form of evidence about this, I fail to find anything about time gain by overturning the wheel in any corner, most of you keep stating this but still no evidence, nowhere, why ?

    I want to test this but in my testing I don't gain any time, is there a special setup ? Does it involve something special ?

    Guys, you keep crying about this "exploit" but nobody is able to provide any form of proof or anything repeatable...
     
  14. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173
    Slip_Angel is stuborn but this is not a problem to me, I can be as well. On the other hand, when he insults others by saying that they are "fanboys" and therefore when he tries to discredit their opinion, I have a problem with that.

    To come back to the subject, if I understand the concern would come from the fact that the tires are "too permissive", you would like a limit to be reached sooner. Ok, let's admit it, but "sooner" means when exactly? How do you determine that the limit is correct without a thorough study? And why the present limit is wrong, beside personal assumption?

    Then there is the famous "exploit" that Slip_Angel has shown in some videos where the driver was turning the wheel too much and too abruptly...we have seen the behavior and we find that it is indeed a very bad driving, the question now is to know if :

    1) do we gain time by doing this?
    2) what are the consequences?

    I'd like to test but the tests I've done so far have been done with the default setup, I've been EXTREMELY penalized, both in terms of stability, time and consequences, in other words with the default setup, doing this is suicidal...and yet we're talking about an "exploit" that would give a decisive advantage, I'd like someone to show me what to do, what setup ?

    I'm also waiting for someone to show me a complete race during which it would have been done, it should not be complicated since we are talking about aliens who do it systematically in private league but for the moment, still no proof (which I agree can be "complicated" in the end since it would take an inside view to see what the driver is doing).

    That's a bit my problem, I'm not saying that there is no exploit, I'm just asking for someone to explain me how to test it because with the default setup, sorry but there is no possible exploit, doing what we've been talking about since the beginning is catastrophic at all levels.
     
  15. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403


    Look mom, it's a plane!!! Man, look at this BS and tell me you don't see anything wrong. I don't care about it even that much, as I am never driving that way ever in my life, no pleasure exploiting the game and under those circunstances there is an awsome simulation I love! When you see something like this and compare it to what Alsono does with understeer I have to laugh. Slip Angle is even seems like there's none, just a car hovering the track, somehow not gripping anything but taking a curve as if planted (like a plane...). This might no be consistent at all. This is a driver that at least for shorter stints will do it and I am sure most Esports player will do this.

    Alright, you are not saying there is no exploit, but that's not what you've been sounding like a lot of the time... There are strange thecnics and then there are impossible thecnics that only work in a simulation.

    I've been saying that devs should look at it man, and then test, when and if time exists and if they aknowledge that it needs looking into it. I mean, don't they have to base their models somewhere? Some observation is still required and listening to drivers on this, since sometimes data isn't even given and sometimes unreliable. They have to base it a bit on good sense too... Iracing says somewhere on their site that their simulation is based on car data and all of data, more than anyone else and then you have no problem trashing the grip dynamics there. Cmon man.

    Watch it all, towards the end there are even more ridiculous driving then at the start.
     
    mantasisg and Slip_Angel like this.
  16. Yzangard

    Yzangard Registered

    Joined:
    Dec 6, 2018
    Messages:
    351
    Likes Received:
    173

    Ok, make a simple test then : try to do what he does with a lot of FFB, the strongest you can "handle"...then, try to do this with no FFB, at all...

    You'll see where the problem lies. The exploit is unrealistic FFB.

    Check tire wear too...

    Then, check without doing this your lap time...are you faster, really ?

    And yes, it is EXACTLY what Alonso did.


    EDIT : tried it and really, even before my timed lap it was a no go, my wheel was so vibrating it was impossible to drive. Maybe my FFB is a bit too high but clearly this is an issue here, his wheel don't even seem to move a bit, he is smoothing like crazy or don't have set his ffb high enough.

    This is an exploit indeed but not the one you think, at least not the way you think it is. The problem is not that it shouldn't be possible to do this, it should...problem is consequences shouldn't only by tire wear or temp (his tires are burning sometimes, which should lessen a bit his grip but should ruin them as well and it doesn't seem to be the case, not enough anyway) but should also be huge force returns when he realign his wheels (with higher FFB it is MUCH harder to be that smooth) and the vibrations induced...

    The only real issue beside FFB I can spot is tire wear which seems too low, not sure what tires Senna is using...

    He seems to be using some anti-lock as well or he is VERY precise because braking like he does without any lock, seriously this is almost perfect...
     
    Last edited: Jan 24, 2021
  17. hitm4k3r

    hitm4k3r Registered

    Joined:
    Jun 15, 2016
    Messages:
    1,320
    Likes Received:
    3,121
    As explained plenty of times, there is allways a video of that single person with a hotlap, a guy who uses a particular technique to get around a corner and suddenly everything is broken. Where are the hotlaps of the 100 other pro drivers? Really, after all the stuff that I explained to you? Do you have any info about his steering weel range and steering lock? No, so I would be carefull judging the game engine based on a 3 minute video of the same person over and over again because I see nothing particular that looks too far off knowing that people can adjust quite a wide range of settings. I could create a similar video using a steering wheel range of 500 or 600 degree and drive in a similar fashion, but I don't do it because it doesn't fit my driving style. There might be a video of a person doing a similar laptime with different inputs abusing his tires less. All we do here is hypothetical and I am just making a guess here, but I suspect the pros won't share their setups for obvious reasons. So unless you know his setup all you do is speculation.
     
  18. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    I don't have a direct drive wheel, only T300, so I can hald whatever FFB and I am sure stupidly high FFB will hinder you, but why would you do that? Anyway, to each his own, but I would like heavier then T300 but have to upgrade when conditions allow me. That is definetly in the equation, I agree 100% with you, low FFB can be helpful but it also can be not helpful...

    I can't understand how reducing velocity with understeer entering a corner has anything to do with the two or three corners before the zig zag at the end of Lemans (the two consecutive slow corners). Alonso is controlling that and slowing the car, when those corners looks like a ball curving in the air with tires seemingly really looking like they are leaving the ground but they still produce a lot of control of the rotation. That to me doesn't have anything to do with what Alonso did, but whatever. And I concede that the behavior of the video can be realistic, but not really what Alonso is doing, not those particular moments.

    I am getting a bit tired of the discussion, since no one really goes anywhere. I can concede that maybe I am too considering some instances where it isn't impossible, but more reproducible than it should in Rfactor2, but some other instances I believe you guys are giving them too much the benefit of the doubt, even when considering low FFB and all.

    All in all, believe me I really love RF2. I just spent the afternoon playing with Oreca, the LMP1 mod, I tried the dissenter and an huston (or something ahah). There are cars with modern tires in mods that I believe are closer to what limit should be like and when one reachs it, but I love Studio 397 because they still feel very freaking good and I don't drive like a maniac...
     
  19. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    403
    Honestly I am not sure sterring locks come into play at all. It has to be enough steering to make cars glide the surface as if oversteering and understeering at the same time but rotation is mantained (I know this is a real state). I think it's observable that the steering is being turned quite a bit (internally for the simulation, if not for the driver on the wheel).

    Could people stop with lap times? That isn't the point.... The driver of the video could have been playing "normally" 80% of the time and then 20% was more of what I consider sketchy to say the least, and the lap would be normal, time wise. People and their lap times man :rolleyes::rolleyes::D:D:D Being cheecky, but it's true.

    Also, I don't want unrealistic driving to not only be slower, I would like to not have it at all, as much as one could!
     
  20. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    12,345
    Likes Received:
    6,572
    I haven't seen a video yet showing conclusively that they're getting extra grip in the turn by extreme-steering. If you aren't at the limit with normal driving, a 'silly driving' comparison isn't valid in terms of grip gain. A car driven neatly at 98% of full speed could quite logically be driven in a slightly messy manner and achieve the same laptime (or even faster) - bearing in mind I'm not giving percentage figures on how messy we're talking.

    Each of the videos showing extreme steering also appear to show increased turn when the wheel is brought back near normal steering lock. I don't think this is a grip exploit. (but I don't think it should be possible to drive that way and be fast, either)

    I don't know if I'm looking at the wrong video, but that car hardly gets out of line (it does, but not a lot). From a TV camera people would say it's on rails.
     
    Nieubermesch likes this.

Share This Page