"Light cars"-VP review of rF2

Discussion in 'General Discussion' started by Nieubermesch, Jan 31, 2022.

  1. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,702
    Likes Received:
    3,623
    Thanks for understanding :D Of course I could be wrong, especially since I am possibly the only idiot in the world talking about it lol

    They can add things to their simulation and improve it in every way they can, and it is amazing. But the simple truth is that whole simulation is rooted from that little area where tires meet the surface, unless the car is in the air, or flipped over. Any simulation could be amazing, and yet suck at the same time due to just being utilized incorrectly. But what if people reject authenticity, because it does not compliment their abilities, or because they have messed up perceptions, or no perceptions at all.

    I guarantee that more than half simracers base their perceptions only on simulations, they are all already doomed, and are anchors of simracing into abyss of mental torture chamber of ignorance, and endless reddit posts of rigs, and meaningless racedepartment articles, and liking gamermuscle physics talk (just kidding lol, wanted to make comedic twist on this, but its accurate). I guarantee that no less than 90% of simracers doesn't have an interest to look at actual true footage closely in observant way, trying to understand what is happening in there. I guarantee that 95% of sirmacers doesn't know that cars can actually have net lift force, even when with aero devices, and most cars actually has net aero lift + that tires work is based on static and kinetic friction relation + that tank slap is result of good grip at the rear and slow hands. I guarantee that 99% of simracers never put any time and effort in attempt to study vehicle dynamics, and prioritize "just feeling good about it, bro", over "I want to know if that is true", and are probably even repelled by too much nerding-out.

    Also a fact. No matter how much devs would nerf cars physics down to most uninspiring and undemanding, adrenaline destroying, sleep activating, fully raceable in complete subconsciousness, driver passivity promoting, twitch chat reading enabling, wife conversation holding (while racing), children and animals supervising (while racing), eating and drinking while racing, noob to pro equalizing levels... No matter, how easy it would be. There would still be half of public servers grids online, frequently spinning out and crashing before Turn 1, and blaming game for being undrivable.
     
    Sim_Player and Nieubermesch like this.
  2. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,702
    Likes Received:
    3,623
    @Nieubermesch

    Never underestimate the aerodynamics. I wish to look at Dissenter physics. I am not sure, have to check, but I am afraid they could be encrypted. At higher speeds, for such non-aero car, aerodynamic lift could have notable effect on its responsiveness to steering. It should still steer alright though, I am sure it doesn't go that fast to get crazy lift, but I think it actually should have forward lift balance, just guessing.

    I think there could really be some issues about solid axle dynamics.

    Aerodynamics also does have an effect on how car might move over larger undulations at higher speeds, at track like Bridgehampton for example, but absolutely everything in a car relates...

    I also have an issue about actually many cars in simracing, that their rides often lacks that special factor, and feels too simple in terms of stability, not that they should be unstable, but there should be more extra little nuances to its ride. It is hard to explain to me too. It could be down to limitations of simulation, but also it could be down to physics being made on safer side for a car, with not as dynamics tires as it should be, not as dynamic aero as there should be, not as dynamic suspension as there should be, not as dynamic inertia modeling, not as dynamic chassis, not as dynamic brakes.... but then also perhaps some of the cars are actually not so dynamic, for example my 2013 Toyota Auris :D, I had Fiat Palio weekend and and Opel Omega and both were more dynamic, with more special factor. But Toyota beats them at simplicity, quality and reliability :D
     
    Nieubermesch likes this.
  3. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    10,617
    Likes Received:
    5,422
    @mantasisg As said elsewhere, I don't yet see proof that solid axles have an issue, based on calculated roll rates and what the game does.

    I don't know that photos are enough research.
     
  4. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,702
    Likes Received:
    3,623
    Yes I remember. I will come back to that, feeling interested again. It was great to have your input, only pictures plus just feeling handling surely isn't enough for fully credible claim. But everything should match up, if calculations doesn't detect an issue, but practice shows something is wrong, then research must be continued. I do hope that I am just wrong, and there are no issues about solid axle in ultrachassis, no restriction of whole chassis compliance, but I must get these cars to work how they should.

    I think every piece is important including having matching cars roll and pitch to available authentic media resources. I'd even say that that next to most basic requirement - correct laptimes, realistic observable visuals should be next thing to match reality: diving under braking, squat at acceleration, body roll, lifting tires off the ground, suspension travel, and all these mentioned can be documented in still pictures, even understeer and oversteer can be observed to some level, not as good though as in, ideally, slowmotion moving pictures. Also fortunately many old cars were bold at these effects. Then after that comes handling, and it is fair to have expectations for cars to have a way to be confidently controllable, especially bellow limits.

    Right now I am very interested in Dissenter, first of all I need to counfirm that it has solid axle.
     
    Nieubermesch likes this.
  5. Devin

    Devin Member

    Joined:
    Dec 30, 2011
    Messages:
    208
    Likes Received:
    134
    True, it sort of really looked like it wasn't, and since similar posts routinely happen on the forums I wasn't particularly surprised to think that either.
    I think that is defined somewhere indeed. Pretty sure it makes a difference but you'll hardly be able to see let alone measure any difference that can be linked to temps. Maybe worth checking some telemetry with cold and hot ambient temps to compare.

    Also to address the whole video: ACC cars have very high inertia for some reason, what I think to be way too much. Going from ACC to any realistic sim will make it feel light in comparison. For that reason, don't take for gossip what is said in some random youtube video. Either they are made by the experts working with the tech and mostly go off of objective data so you can form your own opinion, or they are made by people who just want to present their subjective opinions and create clicks by making it seem outrageous
     
  6. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    There are an in between... There are things that you are also even saying without the propper values of ACC to be saying it has very high inertia... Or do you? Some behavior is discernible enough by the "naked eye". If for example we apply the kind of logic of outrageous to a popular video like Ermin he still was quite on point that the tire phsyics were completly stupid in terms of slip angle and sliding behavior. He didn't need to compare it to any empirical data if available to the real car to tell it's simply not realistic. Some people in this forum actually defended the behavior... That's why we can't have nice things... People are pleased with less so it's all fine at the end of the day.
     
  7. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    I'm not sure what could be at play for the cars to lack a bit of that "humf" compared to onboards I see where cars suspension compressions feel more substancial and dangerous to distabilize tires grip and such. One of them is actually the tires itself maybe too... We all know that DLC cars have those "easy" mode tires little sensitive to loads and such... Might be too I guess. Many cars in RF2 have that great humf with the right setups and all, and it's a great sensation! GroupC MAK Corps is one that comes to mind.

    Sure, some cars will have it more smooth and slow and that is perfectly fine too, it just has to be car and setup dependent, if not it's not really immersive to be driving a vehicle that can't even emulate a little bit of how that car should feel to drive.
     
  8. Devin

    Devin Member

    Joined:
    Dec 30, 2011
    Messages:
    208
    Likes Received:
    134
    Inertia is one of the things that doesn't require a lot of data to see - the cars will just look a lot lazier on load changes and take a lot longer to rotate. This is why in ACC you drive these super smooth lines cutting the hell out of the track. Even footage of a lap from different sims side by side shows this wonderfully because of how big the difference is. It will also make the cars carry undesired movement for much longer, such as bouncing over kerbs which is basically a death sentence there, or stepping out which is near impossible to save partially because of the high inertia. This is why "safe understeer" is the preferred and fastest way to go on ACC.
    Slip angle is actually not as easy to explain by the naked eye. And Ermin wasn't correct with that explanation either. He merely picked it up from people who thought they understood it, and presented it as facts. Most cars released before the M4 Class 1 have optimal slip angles in the range of 6-8°. That is quite a bit too high, but still far from drifting the cars. So why do a lot of people still drift them? Two reasons:
    • They can. It's not faster, but sliding beyond optimal slip is just not slow enough to cost you so much time that you wouldn't do it, and getting grip from all four tyres is still slightly quicker than being bottlenecked by two of them.
    • In slow corners, inertia is the bottleneck, and because sliding doesn't cost you enough grip it's preferred over the inertia bottleneck, as it helps you get around the corner that much quicker.
    This can be shown quite easily in motec, but I'll let that be your homework for today.
    Except for those slow corners, where in cars past the Class 1 it's only preferable for hotlaps due to thermals, driving clean and right on those slip angles is objectively faster, contrary to what some people want you to believe. I assure you, you'll never see a Redline car take a corner sideways.
    The latest tyres fixed the optimal slip angle for the most part, but you still have too much grip beyond it, so it still feels like you can go slightly sideways in slow corners. In fast corners that's just not possible anymore, not without major obvious time loss at least.
    But anyway, that's a different topic.

    On that note, I'm pretty sure you can check out inertia with telemetry as well in ACC, but its telemetry output was very limited for some reason so it may not be that easy.
     
    Nieubermesch and Sim_Player like this.
  9. JamesB

    JamesB Registered

    Joined:
    Dec 28, 2020
    Messages:
    269
    Likes Received:
    180
    My god but you chaps are refreshing, a proper debate and conversation without descending into personal insults and incoherent rants! What a great read this thread is. I to watched the video and was not convinced by some of it, but I am not as knowledgeable as you chaps.
     
  10. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    Not that knowledgeable also and I can enter into some unhealthy rants about S397 and the state of rF2 because honestly that's been just a reaction to all the BS that has transpired from their part. First time I came to this forum was to have discussions like that and still remember threads of the time reaching more then 50 pages... Ah, good times. We got to still see the fruits of our endless discussion from S397 though and I'm sad and sad and sad...
     
  11. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    Good read.
    So you agree what I've said, because it doesn't have dwelve into thecnical aspects of it even going to motec to prove it when observation can be accurate enough to determine. You although first said - in the context of VP video - that he based it on just that, observation and saying it wasn't valid to do so... But whatever, I agree with you on ACC and can see some points as to why some cars in rF2 feel a bit strange, wether official or not, in terms of weight transfer dynamics and inertia.

    It is the faster way to slide the cars in situations where driving at such speed and dangerously would result in a huge spin and probably without possibility for saving it, so I think people are defending parts of it without understanding the full repercussions of a permissive tire/phsyics combination like that. I saw a twisty section of Daytona by Henry Sinik on the C8R and OMG, that was sending the car like a toy and wasn't really drifting, it was kind of careless driving that can only happen in a car like that.

    Why can't the new tires just drop the sliding friction? It's one of the things I've experimented with with the files. I also tried some adhesion diffusion stuff but it didn't work or something, but it looked promising, referring to accelaration of slide and what not... Actually, there is adhesion without diffusion and with diffusion and maybe you can tell me what the difference is? Makes me wonder if one is more for the initial slide, as for example the SkipBarber file doesn't have any of that.
     
  12. Devin

    Devin Member

    Joined:
    Dec 30, 2011
    Messages:
    208
    Likes Received:
    134
    I didn't word that very well. In very few circumstances, video footage can be pretty good reference. However, it is only valid if you know the circumstances, and in case of sim to sim, if you have worked with real data before to back why one of them is correct. What I meant to say was that he was used to ACC so he just presented that as the real deal with inertia, then used that to call other sims too light. When in fact he should have compared that to real cars and/or real data. Since you can hardly compare ACC data to real data, you'd have to pick another sim, compare that data, and if that's correct you can safely say the sim is right and ACC is too heavy if it still actually looks heavier.

    Unfortunately Henri only ever learned how to drive cars with basically full lock. This works in some mod cars that really got things wrong. Very well in fact. But in many cars it's not just tenths but full seconds off the pace. And I understand why he does it in some cars, as it just works there and I would totally do the same. But in many others, it's just the wrong approach.

    I wish I could tell you the answer to any of this. Unfortunately I don't work in the physics department so I could only guess. It's definitely an observation I have made in the past as well. However, that observation truly was only an observation, and not very scientific, so I can't just present it as the correct solution either, even though personally I'd think it would be an improvement.
     
    Nieubermesch likes this.
  13. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    There is another theory I also have about what VP is referring too. If you notice he is playing with the 3rd person camero behind the car and in rF2 when such camera is close enough to the car, it will give a distorted perception of the cars inertia movements being faster then they are. When you back it up a little you can see how much of a difference that can make to the overall feel of what is happening.
     
    Last edited: Feb 1, 2022
  14. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    Yeah, I thought you worded that wrong and glad you aknowledge, but no issues there. I'll have to climb up again in AC and ACC to refresh my memory of how it drives, but I do remember them feeling less nimble and "bloated", at least many of them but who can say whats at play? Maybe it's about their suspension simulation or something too and also for sure about the tires reactions acting on inertia of course, it's all connected.

    It is the wrong approach and glad it is, I' just not so glad that it is the majority of paid content from you guys. But I don't want to go into that again here ahah.

    Oh ok ok, I thought you might have a role in physics too, no problem. I have to go back to messing around with it, but stresses on another parts of my life haven't allowed me to be messing around all those files and all without trully knowing what I' doing.
     
  15. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,702
    Likes Received:
    3,623
    Yeah that view does not compliment physics. One of my issues with rF2 is that overall it is rare that tracks has replay cams, that compliments spotting actual cars in detail from various angles. They just don't do that in most tracks. And I guess 3rd person cam being fixed to car like that also isn't great. But I think it VP just used it to illustrate what he felt.

    I have a speculation that VP might have accidently used some of driving helps, or should have reseted FFB via in game function, I find myself having to reset FFB in rF2 often, as FFB sometimes loads wrong, undetailed and light.

    Well.. to add to a topic some more meaningful posting I did some decent driving with Dissenter since a long time. I had to use reset FFB function, and then it was very awesome, even though there is this issue of steering degs mismatch, it still feels good.

    I couldn't relate to VP experience right from beginning. I would never describe the car as feeling light. Boaty ? Yess it is infact soft compliant, and tires although does have some grip, they do develop sliding progresssively and early, and IMO that very correct for 70s slicks.

    Next I have minimised tire pressures, springs stiffness, slightly softened dampers, minimised ARB stiffnes, and here is what this car did, it was fun sailing, and I'd never call it a light car handling:

    [​IMG] [​IMG] [​IMG] [​IMG]

    NOW, CHECK THIS OUT !
    [​IMG]

    I was also doing weaving in straights, and I could feel inertia, and after some weaving, chassis begins to oscilate, and it needs some time to get stable, of course with soft suspension it should. But it was similar in defautl setup. This made me feel that VP might have had some driving helps turned on, or FFB not fully functional.

    Now last but not least, I actually checked if I can open physics files, and I did. After experiencing such good chassis compliance I suspected this car isn't using solid axle ultrachassis physics. But it does ! So I might have been wrong @Lazza , and happy about it.
    I will be studying whats happening there, and perhaps will break through with ton of 50s and few 70s cars and one 30s car physics with solid axle that I have. It uses trackbar + 3 links, perhaps its one that works well in rF2, as most of my cars use four link originating from broken ISI quad, but I have not had much success with three link + trackbar setup either at gainign compliance...

    Another thing I though when checking physics was that this car probably has too little aero lift. It has some downforce set to front and rear wings, which is obvious and alright, but the body lift value is not very big for big old car, and is almost compensated by wings values... So it ofcourse means that car probably is more stable at high speeds than it should be, but thats just my opinion.

    Anyway, I was very enjoying driving it, and I have same opinion about it as I had years ago, it is one of best rF2 cars. VP must have had something enabled that spoiled the driving, or else I don't know why he felt like car doesn't behave as heavy car. In fact, it is heavy - 1.5ton with driver and fuel.
     
  16. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    2,702
    Likes Received:
    3,623
    haha thanks, I basically just crop and downsize screenshots from replay, best way to do rF2 screenshots, since comfortable free roaming cam + zoom + focus is not implemented

    @Lazza Another find regarding opening Dissenter files, is that it actually uses all those late HDV steering parameter lines, so reasons for mismatching steering angles must be different. I wonder could it possibly be just graphical.
     
    Last edited: Feb 1, 2022
    Sim_Player and Nieubermesch like this.
  17. Lazza

    Lazza Registered

    Joined:
    Oct 5, 2010
    Messages:
    10,617
    Likes Received:
    5,422
    I never looked at the quad, but I found both methods described by Bristow in this post from Emery seemed to work:

    https://forum.studio-397.com/index....etails-about-rf2-geometry.67438/#post-1042190
     
  18. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    I'd like to invite people to try to drive the Ferrari GTE of S397 against the Ferrari GTE from Enruracers. I know they are from different years and different cars, but try to analyse the tires behavior and the lateral bite they seem to give compared to the S397 one. This is why I think some cars might feel weird in terms of roll inertia, because you drive this cars in yaw angles where the tires grip up the same with excessive rotation and the lateral grip doesn't need to be high enough for the car to mantain the trajectory of turn, wich is pretty weird.
     
  19. Nieubermesch

    Nieubermesch Registered

    Joined:
    May 18, 2019
    Messages:
    792
    Likes Received:
    401
    I also didn't dind anything weird with the car honeslty, felt pretty solid, and the first cwr of the hwuston group seemed to be even better, with a much bigger focus on aero of course. Don't remember the name.

    Maybe he expected a little bit of a different behavior from the suspensions? In a way I'd imagined the car would be a bit more (or should I say) less effective. As it is in current default setup there isn't possibly that worn out factor or something of a older car, possibly more compared to road cars of the epoch wich you see in movies have those suspensions that seem to always compress to the max. Can't quite expain it myself that well, sorry.

    I don't no, but doubt he had assists. Now it's even easier to not mess that up with presets.
     
  20. Bill Worrel

    Bill Worrel Registered

    Joined:
    May 1, 2019
    Messages:
    321
    Likes Received:
    203
    [QUOTE="Lazza, post: 1088745, member: 23047"
    I believe 99.6% of @Nieubermesch 's post was a quote. [/QUOTE]

    Why do I think you actually calculated that. ;)
     
    Nieubermesch likes this.

Share This Page