• Bookmark hashtag #RDLMS on Twitter, Facebook and Instagram
  1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Assetto Corsa developers, are they really this childish?

Discussion in 'Assetto Corsa' started by Tommi Ylén, Jan 10, 2017.

Thread Status:
Not open for further replies.
  1. ouvert

    ouvert
    Premium

    well he is saying he dont understant the tyremodel by numbers .. just know how car should feel (which I tend to believe based on his exp .. ) I is like when I`m doing lightning setup for some photographers that can`t talk to me in f-stops, proper teminology or explain them self in technical way but rather describe how ti should 'feel' ..

    Point is Stefano could just ignore him or consider looking at his idea (not numbers, but general behaviour) ... it is 10th itteration of tyre model and every one is the best most realistic one .. next one could be TM11 with behaviour changes mentioned by Fred but resulted from math not feeling ... I guess that was just one of those days when Stefano was woken up by his cat too early or something as he can be farily nice guy ...
     
    Last edited: Jan 10, 2017
  2. The guy is accurate on some of the things that he feels. On some cars, the nose does not tuck in from mild understeer with light braking. On some cars, the rear does not plant itself on light throttle application. These are things that can be EASILY fixed with SLIGHTLY different tyre parameters.

    There is a huge step between stating what feels wrong and asking how to improve his mod car vs trying random complete stupid values that for a random reason feels ok and then implying Kunos doesn't understand Load sensitivity. I mean that's just retarded. Look at the tyre tester app while driving his "tyres that work", it's proper hilarious.

    Everyone that has been there knows what happens when you go full retard about physics on Kunos forum...tbh I wouldn't want this to change in any way.

    Stefano completely ignored his first non-sense thread iirc. That was very impressive to me.:roflmao:
     
    • Agree Agree x 7
    • Like Like x 2
    • Sad Sad x 1
    • Angry Angry x 1
  3. ouvert

    ouvert
    Premium

    well he said he is not regular over there ... so I can see how this might come to him as surprise ;)
     
  4. fsr

    fsr

    When did I ever say I don't know how sims work? I've been modding sim physics for 13 years and I know very well how it works. That's why I think outside the box. I know from experience that math has to be interpreted. It's just one part of the sim. Stefano himself once said:

    "What we really don't have is the part where the car starts sliding. Because there is no data, there is no accepted theory how you do that. And I think it's interesting becaouse each one of us is coming with a different approach and a different solution to that. The problem is that we cannot write a scientific paper on it because there is not data to validate what we are trying to do. So it's really this sort of black art."

    I am new to AC, and the lack of documentation means that the only way to figure stuff out is to reverse engineer all the parameters and try to figure out what they do. I never claimed to have a solution or make random suggestions. My observation was very specific and I provided a clear way to test it. That's all.
     
    • Like Like x 4
    • Agree Agree x 4
    • Love Love x 2
  5. fsr

    fsr

    In certain situations (like compressions or crests), the Kunos street tyres seem to have an inverted reaction to load somehow. It's like it reacts correctly in one direction, but not the other. Once you combine them, you lose grip fast. So LS_EXPX=0 is not a fix or a solution, but a way (incorrect no doubt) to eliminate that one parameter so it doesn't mess up the other one. It's not crazy and it's not random. It's just math. Having said that, I believe Kunos parameters are correct. It's just how the tyres react to them that is strange. So by eliminating one it was better than having two correct parameters fighting each other. In other words, LS_EXPX=0 seems to make the car react more realistically in terms of slip under load in a very particular situation. I never said it was more realistic in general. But people read what they want to read. Then take stuff out of context.
     
    Last edited: Jan 10, 2017
    • Haha Haha x 3
    • Like Like x 2
    • Love Love x 2
  6. fsr

    fsr

    I think you get me wrong on purpose here, dude. I never asked them how to improve my mod. How did you get that impression? Maybe I came across as demanding or cocky in my first post, but I assure you that was unintentional. I simply misjudged the mood of the official forums.

    None of my values are random, and I can see what they do in the tyre tester (not all that crazy the way I look at it). No matter what the numbers say, they don't change the behaviour of the car though. And no, the issues can not be easily fixed with slightly different tyre parameters. I can get one or the other, but not both. With Kunos' values I get understeer when turning in, no matter what. And as soon as I apply throttle I get oversteer. In compressions I lose grip and the grip will increase over a crest. That makes no sense, but it's all consistent with my theory that grip drops when load is added in both directions at the same time! Or it might just drop too fast. No way for me to verify this. Keep in mind, the car I am working on is a low grip car with a live axle and a welded diff. As soon as I open the diff up and give it a bit more grip, the issues are way less noticeable.

    Anyway. This is not something that can be covered in a forum post, but I'm be happy to learn more and get a better understanding of the tyres. It's a black art, and even if my numbers are wrong, they are giving me the answers I was looking for.
     
    • Agree Agree x 4
    • Like Like x 2
    • Love Love x 1
  7. Turk

    Turk
    Premium

    Just to avoid confusion on my part, are you the Fredrik Sørlie from the picture in the OP?

    If you are, you say "i understand car physics, just not in numbers". That gives the impression you have personal experience of the physics of being in a car, but that's a long way from being able to program the maths of that physics into a computer model.

    My point is basically, if your not involved in the behind the scenes making of the game you don't know what's been done, what's possible, what's not possible, you can't really give advice. If a company gets a consultant in for some reason the consultant has to spend his first few hours/days/weeks, learning how the business actually works, he's not just going to start throwing out general advice and assume some of it might stick.

    I can appreciate that many of us want to help but we really don't know enough to give any kind of helpful advice, we're likely to be going over stuff they went over years ago. Kunos likely have a plan of stuff they have to do and if we were sitting in on those planning meeting we'd probably agree that's the best, or only course of action. Then we show up a few months later and start telling them they need to be doing what they're already doing.

    It's annoying to be corrected by people who you know couldn't really have a clue what's going on. And that's not meant as an insult, nobody outside of the Kunos office knows what's going on. Even the guys at rFactor who make a similar game are in no position to criticize or tell Kunos what direction they should be taking because they don't know the particulars of Kunos method even though their doing something very similar..

    Yeah, Stefano probably shouldn't be getting into arguments with customers, but it's difficult to put up with day in, day out. Plus an explanation is boring and people forget it, an argument get's more attention. The alternative is we get a bland interaction with Kunos were they make non offensive statements with loads of words but no real content.

    It's a two way street. We can criticize the way things work, but it's up to Kunos after that, there's no point us telling them how they should fix the problem because we haven't a clue. Kunos staff need to take on board that the modern consumer doesn't know how ignorant they really are, we think because we can look up stuff on google and watched a couple of episodes of "how it's made" that we're all encompassing engineers.

    I'm one of those customers too, even though I've embarrassed myself with my assumptions in the past, I still can't stop myself from making those assumptions.
     
    • Like Like x 7
    • Agree Agree x 2
    • Sad Sad x 1
  8. fsr

    fsr

    Yup, that's me. And I agree with you for the most part. I just feel like I get **** for stuff I didn't do or say. I was simply offering my feedback and a way to test it. I wasn't even criticizing. Sure, I probably stepped on some toes and my suggestion probably turned them off completely. It's a shame though. It was just a way to demonstrate it. Not a permanent fix.

    And no matter how stupid my suggestion was, this is not an acceptable reaction from the developer. But this is a bit like with Trump. If you are rude and ridiculous enough, you don't really have to answer to it. You go from being an asshole to becoming a character or an eccentric.
     
    • Agree Agree x 6
    • Like Like x 2
  9. But then you go on the forums, say you have experience IRL and in sims, that you're a World Champion in both (why not eh ?) or some ****, explain the handling "errors" (even show them in videos with inputs showing) you're feeling and then you have dumbasses asking for data still.

    Back to zero. And all that is assuming that those damn numbers work as intended. Remember here that it's a game that takes those and decides what to do with them. It's not real life. It surely can't be as smoothly as always being 1:1 now can it ?
     
    • Wow Wow x 2
  10. Helmut Skrdla

    Helmut Skrdla
    Premium

    I think your mistake was:

    "But you would have to go away from the default Kunos values. Which may not be a bad thing for realism anyway"

    In there you basically said what they were doing wasn't realism. Which sounded a bit arrogant to my ears, to be honest. Stefano obviously takes criticism on his physics more personal than many others would, but I've been banned from the ISI forums for criticing the move to Steam, so other devs aren't exactly Angels either.
     
    • Like Like x 4
    • Haha Haha x 1
    • Agree Agree x 1
  11. Turk

    Turk
    Premium

    There may have been a certain amount of the straw that broke the camels back and you were just the guy in the wrong place at the wrong time. There's certainly been a lot of people telling Kunos they should be doing this or that when they were in no position to do so. I'm not really informed about the whole thing, I just quickly read through what was in the OP which hasn't got the beginning of the conversation.

    Forums can be a hard medium to communicate through. It's very easy to misinterpret someones intention, many internet forums can be confrontational at the best of times and Kunos have attracted their fair share of **** stirrers which makes it even harder.

    Plus they're Italian, I'd almost be disappointed if they didn't fly off the handle every so often. :p
     
    • Like Like x 3
    • Sad Sad x 1
  12. Helmut Skrdla

    Helmut Skrdla
    Premium

    I've just read a few more posts on the thread past Frederic's "change the values, the Kunos one's aren't realism anyway" comment, and it's pretty obvious Stefano didn't like that line of argument. Especially when it wasn't fed with numbers or data and "only" feel.

    The screenshooted escalation is at the far end of a longer discussion, where it seemed pretty obvious Stefano wasn't getting into it. Frederic sure tried to communicate very maturely and non-provocative, but he still didn't understand that Kunos just doesn't like people with no hard data making claims about the realism of his baby.

    Classical failure to communicate.
     
    • Agree Agree x 3
  13. Why are people continuously getting behind the "I'm a customer" wall whenever they don't like the situations they got themselves in?

    Doesn't matter how much you want to sweet talk the person and say you don't want to offend, and that you want to talk politely, actions > intentions > sweet talk. Even the worst dictator can say all the right things, speak in an educated manner, be polite, doesn't mean you forget and react softly to the things he did.

    This thing basically is: I'm a customer, I talk about the core of your product and say is wrong, and you need to oppress your emotions.

    This post from Chazz Ranger in the AC forums explains well what I'm trying to say.
     
    • Sad Sad x 1
    • Agree Agree x 1
  14. Finally got around to actually testing it, with LS_EXPX=0 it actually replaces the equation and uses a linear one (can't find what controls the slope of that, seems to be using FZ0 and DX_REF to set the intercept) so that particular value behaves nothing like LS_EXPX=0.01 or LS_EXP=-0.01

    I did mention in the thread that LS_EXPX=1 is the "no effect" value, just confirmed that in game, you'll get a constant peak DY at any load.
     
  15. One person uses it as his playground to make himself feel better about himself, since he can't be banned or warned there.
     
    • Wow Wow x 2
    • Agree Agree x 2
    • Like Like x 1
    • Haha Haha x 1
  16. Sort of like how Blimey's not getting banned here so he shitposts in every thread? >_>
     
    • Agree Agree x 5
    • Love Love x 3
    • Like Like x 2
  17. fsr

    fsr

    I know. But it was a joke in context and never really aimed at Stefano. I never even thought he would read it, never mind respond to it.
     
  18. The thing is, Fredrik Sorlie didn't say that Stefano doesn't understand load sensitivity (which btw, he clearly didn't one year ago, as it was almost non-existent).

    A lot of Assetto Corsa physics issues have been uncovered over the years, it wouldn't surprise me that there is something wrong in this case.

    Also, a few years ago, Stefano insisted to me that gyroscopic effects from rotating parts aren't important to the behavior of a vehicle, since they weren't covered in his classic vehicle dynamics books.
    I was very disappointed by his stance regarding this, since anyone who knows a thing or two regarding proper rigid body dynamics, knows how important this phenomenon is. The importance of this in consumer sims was very clear since the GPL days, especially when cars get airborne.
    After that day, I became very skeptical regarding anything physics related from Kunos.

    That said, Stefano went over the line, he's probably under a lot of stress and I think deep down Kunos is not a bad person and I normally like his sense of humor.
     
    Last edited: Jan 10, 2017
    • Like Like x 5
  19. fsr

    fsr

    Thanks. That makes perfect sense, as setting both to 1 also seems to have a similar effect. Ok, so my problem seems to be that over crests and in compressions, the grip behaves reversed. Meaning I gain grip over crests and lose it in compressions. This is with the stock Kunos street tyres on a car with no downforce. I can feel it in the FFB too. It's not getting light over Schwedenkreuz or heavy in Fuchsrohre, but the other way around. And in the carousel I struggle to stay in it, whereas in real life the grip you get from the load provided by the banking makes it more or less impossible to even get out of it if you tried. It's the same with several cars. But with the LS_EXPX=0, it seems to react more naturally. Feel free to ignore my question. I don't want to bother anyone anymore.
     
    • Agree Agree x 2
    • Love Love x 1
  20. fsr

    fsr

    Kinda explain bicycles, doesn't it :D

    You are the first one I've ever seen point this out. I've asked this question many times. In drifting, it's very crucial. If you are going sideways through a turn at 80kmh while your rear wheels are doing 200, it sure helps keep the car stable!
     
    • Agree Agree x 4
    • Like Like x 2
Thread Status:
Not open for further replies.