PC Assetto Corsa PC Mods General Discussion

Discussion in 'Assetto Corsa' started by Thom Lee, Apr 28, 2014.

  1. Ema

    Ema

    Messages:
    37
    Here we go
    http://www.mediafire.com/file/5iaepxhy685ppfr/malagoli_fuji_speedway_gp.zip/file
    A lot of k5n files (with italian names), because i made it with my old and not so powerful pc and it was difficult to use a big file with 3dsimed...
    With that rig i also play with shadows disabled, so you can find some object with incorrect shadows settings...
     
    Jimlloyd likes this.
  2. CrisT86

    CrisT86

    Messages:
    151
  3. ArchKyuubeey

    ArchKyuubeey

    Messages:
    78
    Hey,


    Firstly, about the "real data doesn't always translate into simulators" misunderstanding.

    There are some small, not that consequential limitations in most commercial physics sims. For example in AC, the knee speed of a damper is not adjustable via setup. These kind of scenarios will require some interpretation and often a minor sacrifice to be made. There's numerous small things like this in the suspension and systems. They all add up, but it doesn't mean that you don't use real, reliable data as the starting point. You always do.

    Then there are more serious shortcomings, such as the incorrect internal force suspension calculations in vanilla AC. You must adjust the "real data, correct suspension"'s antisquat and antilift amounts (Mainly for driven axles, extremely inconsequential if there is no drive force) closer towards 0% in vanilla AC, otherwise they will be far too high. Then you must adjust generally the wheel points to maintain curves, as per the real data. It's doable.

    The end result in RWD cars for example is generally a tiny bit too much antisquat for a given antilift amount, which in practice will mean most RWD cars for example will lift the rear too much on braking (Perhaps 10mm in a bad scenario) and consequently have a too soft rear on braking (Extremely inconsequential) but when done properly still have adequate stiffness and deflection on the squat side, where more forces and load is being directed during acceleration.

    This can make some suspension types very difficult to replicate, often with minor or major sacrifices. In other times, if the layout is a typical strut or DWB setup with low anti-percents, it's not a very consequential adjustment. Examples of extremely majorly hindered cars which require a lot of expertise would be the S13-generation rear suspension cars, and an example of extremely minorly hindered cars which are easier to produce would be the NSX. Semi-trailing arms, swing arms and the like become more tricky, but they are doable. If you make the adequate sacrifices, the difference is minor albeit larger than most DWB/ML/STRUT.

    Tires of course are not perfect in vanilla AC nor are they in CSP. In no sim are they. AC probably has the best tires of the sims currently with CSP, with probably the highest telemetry correlation. Tire data inherently is difficult to deal with as it is rare for factors to be decoupled from eachother in the tires, for example the surface heating will instantaneously affect your load curves and camber curves! On top of that the common sandpaper measurement method is quite bad and not realistic, and it is very difficult to datalog a spinning, flexing tire anyway. Most of the time tire data is not very usable (Have you ever looked at a raw plot?) or just not available. However the more savvy physics guys have more data available that you can make conclusions from and generally have a better understanding, but here is where you can argue that "data doesn't translate into the simulator" because the data can be indicating multiple things depending how you interpret it. Suspension has almost no such scenario.

    Then some more general stuff:

    It's completely down to the OEM and designer how the car will behave. There are no rules, although there are common philosophies and laws that manufacturers must follow. Despite that, a sports compact from company A will be significantly different to a sports compact from company B, or even one suspension package for the same car from the same company can be extremely different. For example NSX vs NSX-R.

    OEMs use various simplifications of car behavior expressed numerically, such as understeer factor or stability index. You can make an argument that common designs fall into some criteria, and it's often correct. Such as economy sedans having a higher understeer factor, roll-understeer and often lower ride and roll frequencies compared to sports compacts. However this is no rule. People believe it to be a rule, and ignore that Honda will make potentially a radically different consumer car than what Audi will make for example.

    Hence the misconception that "no production cars lift inside rear tire" when referring to FWD sportscars like the DC2 Integra Type R. This is however incorrect. The ITR lifts the inside tire mildly (As you can see on tens of Best Motoring videos...) due to the stiffer rear stabilizer, and adequate body-roll for the tire to effectively lift off the ground. Were one to limit body-roll via stiffer corner springs, or stiffen the front stabilizer (Be it via stiffer bushings, or via a stabilizer size or geometry change), both common changes to DC2's, it's likely the inside tire would perhaps unload; but not effectively lift. A similar situation is in the Peugeot 106, where tuned cars seemingly lift the inside rear tire less than stock cars, despite the tire being unloaded for equal or longer distributions of time during cornering.

    The DC2 economy models such as the GS-R on the other hand do not significantly lift the inside rear tire (Except on a violent, sudden trailbraking action where the soft cornersprings and dampers will allow unloading of the inside rear tire) because the front roll rate is higher than the rear roll rate. The dominant way this is achieved is via a thinner rear ARB, which couples the wheels together less and doesn't jack up the inside rear as much. There needs to be a distinction made here.

    For a concrete example:

    Front
    25.0mm Solid, 98R = 68389.690995392 x 0.3819^2 = 9974.47298032
    24.0mm Solid, 96R, GS-R = 58086.558987938 x 0.3819^2 = 8471.78580151

    Rear
    23.0mm Solid, 98R = 152618.83365846003 x 0.5138^2 = 40289.9130498
    14.0mm Solid, GS-R, RS/LS = 20951.193829124 x 0.5216^2 = 5700.11923298
    13.0mm Solid, RS/LS = 15576.48121224 x 0.5216^2 = 4237.83966032

    At the wheel before bushings, the later Type-R stabilizer makes 9.50 (!) times rate, while the front only makes 1.18 times rate compared to the softest OEM bars*. The excessive rate will largely jack up the wheel more and couple the wheels more without significantly influencing steady state behavior. It's not optimal for total axle forces, but it's one way to safely and easily acquire a looser handling to a FWD car.

    Things like this is why I urge people to not make generalizations and to try to maintain a context.

    I hope the essay was entertaining for those who are interested.

    * Correction, softest OEM bars I calculated. 22mm OD front bar exists for the LS, that'll be a bit softer still compared to the 24mm OD front bar. However they'll be similar.
     
    Last edited by a moderator: Nov 22, 2020
  4. OooAhh Cantona

    OooAhh Cantona

    Messages:
    1,137
    Location:
    United States
    Are you guys referring to the 2018 Hockenheim that was updated by PatrickPat? I had the same initial thought: "what is different with this VheEth/Reiza version compared to PatrickPat version?"
     
    Last edited: Nov 22, 2020
    Elio92 likes this.
  5. Ema

    Ema

    Messages:
    37
    Yesss, the Dev Mode... i've completely forgotten, maybe i activate this 4 or 5 years ago on the old pc.... Thanks a lot!! :gtpflag:
     
    norms likes this.
  6. Ema

    Ema

    Messages:
    37
    I think Patrickpat is based on the old Codemasters conversion, VheEth is from Automobilista... I don't know if AMS1 tracks are laserscanned, but i think is a better base....
     
  7. CrisT86

    CrisT86

    Messages:
    151
    I was talkin about the T78 version which has around 4-5 years at least i think. That's the only version that includes all layout of the modern version of Hockeinheim
     
  8. gutbomb

    gutbomb

    Messages:
    201
    It’s actually the same texture and even the same material, it’s just that the model itself was so low poly that this is just how the light is hitting it, making it look like that. I tried a few settings to make the lighting more even on those materials but then it just made things look too dark in sunlight or too bright in shadows. I ended up just turning my grassfx setting to the highest setting and it makes the grass so dense you can’t see the underlying texture and everything looks normal. I think the mesh would need to be rebuilt to get rid of that and that’s unfortunately beyond my skill level :(

    the grass on the hay bales is intentional, it’s supposed to look like little bits of straw sticking out of the haybale. I usually have grassfx turned all the way up and it looks a little better than in your screenshots but I admit i wasn’t convinced it was the way to go. I had the idea a few months ago to start doing that to hay bales and thought it looked cool but yeah in your screenshots it looks a bit weird.
     
  9. Montie

    Montie

    Messages:
    43
    I would love to see the 2001 layout as well. Great work
     
  10. ParanoidDoge

    ParanoidDoge

    Messages:
    526
    Location:
    Scotland
  11. JakubP

    JakubP

    Messages:
    297
    .
     
    Last edited: Nov 24, 2020
  12. LiquidSkyMan

    LiquidSkyMan

    Messages:
    190
    Location:
    United States
    if U do the additional layouts I think I know someone who will do the AI ;)
     
    Paulo Ribeiro, kimiraik and VheEth like this.
  13. twobegreen

    twobegreen

    Messages:
    5
    Fantastic work. I would also be happy about the other layouts.
     
  14. maruto

    maruto

    Messages:
    76
    What a wonderful gift. Can't thank you enought.
    It is such a pleasure to finaly drive this track with that level of quality.

    ..

    I've made a billboard skin for myself, mostly Advertising Local beers and Canadian companies ...
    as a Canadian who lived most of is life in Québec,
    I'd rather see local billboards than motoroil ;)
    Those sponsors are fictionnal, St-Jovite circuit doesn't have that much billboards anyway...

    Feel free to include it in future updates if you like it.

    Thanks again!

    Screenshot_wec_gte_chevrolet_corvette_c8r_rt_mont_tremblant_22-10-120-15-0-35.jpg Screenshot_wec_gte_chevrolet_corvette_c8r_rt_mont_tremblant_22-10-120-15-0-54.jpg Screenshot_wec_gte_chevrolet_corvette_c8r_rt_mont_tremblant_22-10-120-15-1-7.jpg Screenshot_wec_gte_chevrolet_corvette_c8r_rt_mont_tremblant_22-10-120-15-1-40.jpg Screenshot_wec_gte_chevrolet_corvette_c8r_rt_mont_tremblant_22-10-120-15-2-27.jpg Screenshot_wec_gte_chevrolet_corvette_c8r_rt_mont_tremblant_22-10-120-15-2-58.jpg Screenshot_wec_gte_chevrolet_corvette_c8r_rt_mont_tremblant_22-10-120-15-3-12.jpg
     

    Attached Files:

    Last edited: Nov 22, 2020
    KevinK2, Papifix, soulkobk and 21 others like this.
  15. VheEth

    VheEth

    Messages:
    195
    It must be a person we know in common :D
     
  16. LiquidSkyMan

    LiquidSkyMan

    Messages:
    190
    Location:
    United States
    To quote the old man in the "Surprise visit" part of Clockwork Orange -

    "Who on earth could that be?" :lol:

    EDIT: So PM me when ur ready, boss!
     
    Last edited: Nov 22, 2020
    xezez and VheEth like this.
  17. ArchKyuubeey

    ArchKyuubeey

    Messages:
    78
    Alright, I get your point. If you gather more experience you'll understand why such discrepancies can happen. More often than not it's a perception issue, but in this case there can be an empiric reason for it.

    The common reason would be wearing bushings and bumpstops. If your car truly is stock with old bushings, it's possible that the rear stabilizer is becoming decoupled and can indeed not be influencing the wheelrate! A similar situation can happen when bumpstops rot, robbing them of their rate ultimately.

    In the case that those aren't the reason, and your bushings are perfectly working, there's only one empiric thing I can think of, the fact the rear spring is a two-rate type in the ITR.

    However I'm 99.9% sure the spring is effectively a linear rate because it should work like a tender spring. The load of the car will compress it into the linear rate section.

    I've compared the DC2 to video and it lifts the tire up essentially at a closer time on the same corners on Tsukuba and in slalom testing than what I can tell by eye, and the handling is not divergent from the real car. Hence I'm inclined to believe it's a perception issue for you. Your idea of what is right ("Every production car keeps the tire on the ground") is not congruent with reality.

    In fact when I started the DC2 physics some years ago, before I had a good grasp of how to calculate rollbars and the required data to do it, I struggled to figure out why my DC2 *doesn't* lift the tire up like IRL. :lol:
     
  18. ales100i

    ales100i (Banned)

    Messages:
    698
    Just try it. To me Codemasters conversions can look good, but are oversimplified. All the curbs are the same size and shape and can be driven over the same way. It's far from that IRL. Some curbs are flat, stepped, very high. And this changes the driving lines completely. The next thing is road camber. All Codemasters tracks have zero camber, which again totally changes how you drive through a corner. And lastly, Codemasters tracks have zero degradation of track mesh, which happens everywhere with time. Before tight corners after long straights bumps develop, in some corners dips develop that can help you take the corner more agresivly or can destabilise the car. All these change the racing line, brake points completely and for me add to the immersion. I love studying real life onboard fotage and try to replicate the driving style.
     
  19. LiquidSkyMan

    LiquidSkyMan

    Messages:
    190
    Location:
    United States
    I was wondering if someone could tell me what causes this error and if there's a fix. It seems to only happen in a race not Practice or Hotlap mode. Also it's random, happens on different tracks but not 100% of the time sshot-027.jpg
    I appreciate any help and thanks in advance
    using latest CSP available and latest SOL
     
  20. JakubP

    JakubP

    Messages:
    297
    .
     
    Last edited: Nov 24, 2020
  21. browser

    browser

    Messages:
    76
    Nice...

    I am totally confused with all the Lambo’s what the latest model is that would compete in WEC GT PRO class. Which one should I use?
     
    jeffnghongda likes this.
  22. djtopa

    djtopa

    Messages:
    68
    I just hope somebody will someday get the DC2 70% right in feel like they do IRL.All those till this day dont come even close...and I tried all I can find and thats a lot lol NHF to all who tried and thanks.
     
    SiR80, JakubP and jeffnghongda like this.
  23. VheEth

    VheEth

    Messages:
    195
    Guys, just want the national layout of Hockenheim? Should I also add the two short layouts? Let me know.
     
  24. ArchKyuubeey

    ArchKyuubeey

    Messages:
    78
    Just for you, here's a gallery of images of DC2's lifting the inside rear. Of course the drivers are very skilled and quickly step on the gas, so it never really lifts up very high. They're not like torsion beam cars where it'll stay up even if you floor it.



    If you want, you can be my quest and measure for me stock DC2 and ITR rollbars. I'll calculate out the rates and you'll see that it ends up basically the same. My rear geometry isn't necessarily perfect I suppose so that could be one source; but the RC height and other relevant factors won't influence this thing. Only the handling. Stiffer cornersprings will *lessen* the effect, and stiffer rear ARB from the point of being completely jacked up will actually lessen it as well in most situations. The ideal situation for rear lift is with low ride rates, and high amount of rear coupling which is what the ITR and 106 have, and cars with torsion beams have the coupling inherently.

    Sadly FWD cars never really feel right in any sim because the FFB is so different. No matter what I do, it'll never feel as similar as RWD cars do. I can just rest well knowing that there isn't, and won't be for a long time, a more accurate DC2 suspension in a sim when it comes to actual behavior. If it feels right kinesthetically or looks right visually is another thing. Such perceptions will vary from person to person. It's one reason why driver feedback has limited value.
     
  25. ParanoidDoge

    ParanoidDoge

    Messages:
    526
    Location:
    Scotland
    The more the merrier mate. The short layouts are fun to drive when you use the Aprilia Griiip / any wee cars.
     
    RenaultDJRTP, Papifix, VheEth and 3 others like this.
  26. Shokeus

    Shokeus

    Messages:
    151
    Location:
    France
    Porsche 911 (993) Cup 3.8 RSR 1997 , V0.5 : by Shokeus Act80's

    I did the best to stick to reality but it is difficult to find the right information on this model , 3D level I did not succeed in integrating properly the gas hatch in the engine hood and I did not succeed in doing the template/AO for the bodyshell , I still have the same problem and if I use blender I have holes in the body , if a charitable soul would like to help me it would be nice , waiting for fun to you (I like it a lot !)

    Link : https://mega.nz/file/sNYi1JwT#FYNAAY3QoR-Sd3CWnVG7iSqbjn99AHKvBhjIdMzPdNI

    [​IMG]
     
  27. TaistoTurri

    TaistoTurri

    Messages:
    280
    Location:
    Finland
    The more the merrier when it comes to layouts. That's my point of view. There's no such thing as useless layout for me. :)

    So my vote for short layouts is yes.
     
    axel krueger, Papifix and VheEth like this.
  28. DaWallace

    DaWallace

    Messages:
    36
    Guerilla Mods GT4 Pack 0.7 Update

    sin_gt4.jpg


    Changelog:

    • added Sin R1 GT4
    • added Alpine A110 GT4
    • adjusted BoP
    • reworked some car sounds

    known bugs/issues:

    • Dash of Sin R1 GT4 is placeholder, not fully functional
    • Dash of Alpine A110 GT4 is placeholder, not fully functional
    • Sin R1 GT4 wipers not working
    • Alpine A110 GT4 wipers not working
    • some mirrors not working correctly
    • sound of Sin R1 GT4 is a placeholder (new sound is work in progress)
    • some interior are placeholder and not finished yet

    requirements:



    Pack download:
    http://bit.ly/GUERILLA_Mods_GT4_BETA


    opimized skins & templates download:
    https://bit.ly/35RxZCF


     
  29. safi hellie

    safi hellie

    Messages:
    1,126
    Are you the guy who 'updated' KUNOS car physics?
    That is some serious ego.

    Out of interest if its OK for you to update KUNOS's very own creations..... why do you or any of your cohorts object to someone updating your 'stuff'.

    Is it just pure ego?
     
    Last edited: Nov 22, 2020
  30. RMi_wood

    RMi_wood

    Messages:
    616
    Location:
    Finland
    Physics take time, and sometimes as much as you want real specs, it cant always happen.
    With GTR2 I spent about 300hr on physics for the Porsche 917K, and I never drove one.
    It did however FEEL and LOOK right. Every positive change one way, resulted in 2 bad traits picked up.
    In the end the specs were very close and the results achieved.

    In trying my hand at AC, getting "real" physics is doable, but there will need to be tweaks to get it to feel right.
    Suspensions aside, so much is achieved getting the dyno, gearbox and tires close to spec...
    this is the easiest to replicate and where 99.9% of mods (and Kunos) get it wrong.

    btw: my first mod ever was a 298 "hi-po" and Borg Warner gearbox for the Simbin Mustang in GTL
    (there was 0% basis in reality for the game supplied physics)

    Instantly that car FELT better and more "historic" than the cars Simbin had access to (and that some actually owned)
     
    Camill08, racealot, JOCA and 6 others like this.