[WIP] Jaguar XJ13 1967

Discussion in 'Vehicles' started by mantasisg, Nov 26, 2018.

  1. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Great, big thanks :) Will be interestig to catch the next letters :D

    I must have somewhat useful feel lol I have settled on 17 before and the highest RawSTR I saw was above 16, so I suppose 17 is good value. It is very hard to drive on the limit and look at the value changing so fast and so much real time.

    Will most definitelly have to revisit everything I do with physics, as most of my geometry is Howston. Now I have [BODY] and [JOINT&HINGE] objects set somewhat approx. I also have done whole lot of changing in .TBC and a tiny bit in .TGM. A lot of changing in .HDV very little to suspension though, hardly anything to aero, mostly adjusted tiny bit of lift. The car though already feels awesome and I almost want to share lol

    I will release a link here for public testing soon. I think.

    I have a good time in DevMode with Goodwood circuit, I had highest strength mid turn at T1 aka Madgwick curve. P.S. found out that when you do donuts you have nice "god rays" in the smoke caused by shadows of trees haha
     
    Corti, jimagn, Bozak and 1 other person like this.
  2. Emery

    Emery Registered

    Joined:
    Oct 24, 2010
    Messages:
    2,341
    Likes Received:
    966
    @mantasisg @Bozak
    Two notes worth mentioning if you use that spreadsheet:
    1) For purposes of rF & rF2, "Center of gravity coordinates" should be 0,0,0 rather than the actual center of gravity. Individual parts, as you would expect, should use their CG coordinates.
    2) The inertia for parts is calculated with point masses. Heavy oblong masses (body, V12 engine) should probably have their individual inertias calculated via the box method and the result put in columns G, H, I for their respective parts.
     
  3. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    I have interesting time today as I ported the car to the main game, just for test. And everything seems to be same as in devmode, or better no real troubles occured. Of course car is just in it's beginning of porting to rF2 so it actually has tons of stuff to work on.

    It is great now that I could test it in main game, I had blast driving it at Oulton, Mores, Zandvoort, Sebring. Now I can see that I probably have engine braking a bit too hard in trascks like oulton and mores, it pulls too hard. I like the feeling of car being stressed under hard coasting, I think thats how it should be, but of course it shouldn't act like a handbrake everytime, but it should be twitching IMO if you are really coasting on the limit.

    One interesting thing is that in Devmode car is reflective in Joesville, but not in my Goodwood track. Now that I checked in game I found it interesting that it wasn't reflective in Oulton, but it was reflective in Mores, Zandvoort, Sebring.

    Few screenshots:
    [​IMG] [​IMG] [​IMG]
     
  4. Bjørn

    Bjørn Registered

    Joined:
    Oct 4, 2010
    Messages:
    1,543
    Likes Received:
    450
    Have you ticked on "Reflection mapper", as in is set to true?
    It can also be a problem with the track, if the geometry isn't set to reflect (on for example a carbody)...
     
    KittX likes this.
  5. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    pls Help pls

    Wanted to play a bit with the physics adjustments today. Haven't opened devmode for quite a long time. And now I can't load the game, tracks loads completely, and just as the game should appear app turns off. I have noticed a line which notes about new build, a build 356. But I don't see any new build to download or install.

    Sorry for no news, thats just my nature of using threads as a tools to help me progress.
     
    philmcqueen likes this.
  6. Bjørn

    Bjørn Registered

    Joined:
    Oct 4, 2010
    Messages:
    1,543
    Likes Received:
    450
    Have you by any chance packed up physics in mas file?

    If there are duplicates of physics, it'll crash the car in devmode. For example, if you have physics files losely around in your devmode folder, but you also packed a mas file with the physics files in its proximity, it'll crash upon load. That's something I often do :D
     
    mantasisg likes this.
  7. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    No it is not that. Files are all good. I also tried launching Skip Barber with Joesville, and same thing happens. I feel like my devmode didn't update properly, or something like that.
     
  8. Bjørn

    Bjørn Registered

    Joined:
    Oct 4, 2010
    Messages:
    1,543
    Likes Received:
    450
    Integrity check... Reinstall... Back-up vehicles and locations folder and perhaps some other config files - then try deleting some of those files so they can be recreated via Steam download..?
     
  9. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Not a fan of reinstalls. Tried integrity check, downloaded something that was 0 bites. Dev mode still turns off after game loads. Still could try deleting whole moddev folder and hoping things to get fixed by redownloading via verification.
     
  10. Damian Baldi

    Damian Baldi Registered

    Joined:
    Jan 10, 2012
    Messages:
    551
    Likes Received:
    188
    Remove old plug-ins if you have installed.
     
    mantasisg likes this.
  11. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Thanks for the help guys, I have managed to miss that I actually had packed files there... so @Bjørn was spot on from first guess.

    So I am continuing on developing weird tires :D
     
    Emery, Hazi, Duncan and 1 other person like this.
  12. Bjørn

    Bjørn Registered

    Joined:
    Oct 4, 2010
    Messages:
    1,543
    Likes Received:
    450
    Told ya' ;) Good stuff!
     
  13. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Slowly getting around things... it is very interesting to explore rF2 physics.

    As for example I discovered that I had my pneumatic trail "a little too big", it was about 14cm as a reference at 300kg load with zero slip, with rear wheel just a little shorter pneumatic trail as rear wheel is wider so the contact patch has to be shorter. As a result car felt like it is aligning too hard lol, in some cases I liked the sensation, but sometimes it felt slightly strange.

    I had read that pneumatic trails vary per tires. I can't quote now, but I found somewhere figures of 1-5cm of pneumatic trails. I still have no idea how to actually know the figures, but logically narrower tire will have longer trail, tire that deforms more especially longitudinaly will also have longer trail, I don't know if compound and tread type should have any impact... I aimed for about 6cm at refference of 300kg at no slip, probably a bit on the larger side, but it is not a length of whole footprint :D and car feels good.

    Speaking of pneumatic trail, I find it interesting how rear tires also have it. I mean of course they will have it, but does it really do anything there ? I mean rear wheels are not steering, so whats the difference if reaction forces for tire goes in the center or with slight leverage. Wouldn't rear wheels care only about shearing force ? Or does MZ torque for rear wheels also important for instantanious alignment ? I did found more rear pneumatic trail might be more resistant for slides, but not totally sure about that.
     
    Chris Lesperance likes this.
  14. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Being alone with this is horrible. So I thought I would share current thing as a test version for those who would like to help me in development, in any way.

    Model implementation is super basic, it is meant to be just for physics experimentation currently, also there are some troubles with blender modeling software regarding rF2. No driver, no lights, no gauges, no rain stuff, no deformations, no damage, no dirt, no animations of anything, steering wheel is not tilted, interior looks bad, exterior looks bad, no mirrors, no lods.... and so on.

    Physics began from Ford G40 MK4 mod, which I believe started from Howston, so there is significant amount of legacy from there, but I doubt you could tell. A lot of values needs to be checked again, and will be checked again. A lot of things are based more or less on logic and/or guess based on feel and my understanding how the car should drive. Some values such as anti roll bar rates, longitudinal weight distribution I took from Aphidgod physics from AC, he does not have rF2 unfortunately. I am confused about making gears, so they might not be correct, though ratios are correct. Car hardly goes faster than 300, not sure if that is related to too much drag, or gears not good enough, car had too low topspeed even with way too much tire expansion. Speaking of aero I wanted to make steering to go light at highspeed, and I think it worked out rather well.

    The most time I was spending with tires, I have altered them heavily, they are very different than they are on Howston. I was mostly testing in my LiDAR Goodwood WIP track. It is good because there are lots of footage of GT40 being driven on the limit there, lots of laptimes and XJ13 had to be quite similar to GT40, as it had to race against it.

    I wanted to make a tire that grips very well, does not slide all the time, but requires smooth inputs and fast corrections while being rather comfortable around the edge, but if abused going wrong fast. In general it has a lot base grip, quite low sliding friction, quite a lot load sensitive, quite extreme driving force and lateral force curves (which I have no idea how would actually look like), significant grip loss with overheating tire, also works well in the wet. Car does not understeer very strongly, usually if you start to understeer you start to oversteer after that. Car may have too much front brake bias in many cases. If you are able to set the car for a corner it can take it at quite respectable speed. Initially I wanted a bit lower base grip, but pace was way too slow in Goodwood, so I suppose base grip will drop anyway with wear so... you'll feel when tires will start giving up, I wonder if they does not give up too much/too fast now.

    It is not very serious work, just wanted to get it to drive as believable to me as I can now. Refining everything and developing further will come later. First of all, I'd like to improve my car geometry knowledge and get to understand fluid mechanics for aerodynamics lol, also get done with some simple things which confuses me such as gears.

    Here is a link, please don't share it anywhere, as the mod is in quite awful state now, it is just sketchy test version to feel how the car should/would drive.

    https://www.mediafire.com/file/y6j8...ngWXVgim36pIsXUQEhHEb9KMfhJCdLvVA-1jnBrg49RPE
     
  15. Jokeri

    Jokeri Registered

    Joined:
    May 17, 2014
    Messages:
    164
    Likes Received:
    15
    What is pneumatic trail and how did you determine its 14cm?
     
  16. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Pneumatic trail is the distance between hind the tire contact patch center point and the point where resultant force is applied as tire changes direction which is behind the center. Pneumatic trail is dynamic, it changes with load, contact patch shape and slip. You can also check wikipedia :D

    I did not determined that it is 14cm, and it definitely is not 14cm, unless maybe for huge deflated tire it could be so. At first I did not pay proper attention, so it was 14cm by mistake. I reduced it by half now, and it is probably still too large.

    I have searched for some cues about pneumatic length figures, I remember seeing some somewhere, but I also could have mistaken them with mechanical trail values (related to steering axis, caster).

    Yesterday I have tried to look up for them, but best i could find was the study that assumed initial pneumatic trail to be 1/6th of a contact patch. So I suppose it could be somewhere between 1/4-1/8 of contact patch for various tires.

    I have to look up for knowledge about contact patch shapes and parameters, but I think they vary between 10-20cm in length. But they obviously vary a lot depending on tire diameter, pressure and construction type, though as much as I looked yet radial and bias-ply tires are more different in contact patch width and less different in contact patch length.

    Tires are always very confusing.
     
  17. jimagn

    jimagn Registered

    Joined:
    Oct 6, 2010
    Messages:
    164
    Likes Received:
    37
    I've tried your mod at a couple tracks (Kyalami 76 and Rouen '70). I really enjoy what you've done so far. For my driving style the default setup gives lots of oversteer on corner entry and exits corners with lots of stability. I adjusted the setup to make it a little more predictable on entry. (My favorite rF2 car is the BT20, and I like a car to tend to oversteer.) The XJ13 is still difficult to catch when it starts to spin, but not impossible. Setup changes were very effective and predictable. The car responds well to changes.

    For a sports car from 1967 that is supposed to compete with GT40 and T70, I'm having trouble getting comparable lap times out of it. That's probably due to slow driving on my part. :D What is the weight and horsepower? I don't have your Goodwood track. What other tracks are you using for testing so I can compare?

    Keep up the work. This is definitely worth the effort.
     
    mantasisg likes this.
  18. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Thanks for reply, thats kind of notes I want to get :)

    You are probably getting a lot of oversteer on entry because of engine braking. Ever since we did the mod for AC it was very debatable. Others usually want the effect rather minimal, and I think it should be quite significant. It is hard to tell how much, but I don't want it too much, or much too little. Still adjusting that, perhaps could be a bit softer, but there has to be pull from lock under aggressive downshifts. It is managable right now, but you have to be careful with coasting torque.

    Yes thats idea about slides, it has to be possible to catch, but not easy. Not huge angles shouldn't be difficult, but big angles, or very strong slides will very likely make it spin. For these tires I have pushed sliding severity higher, but also made sliding to appear later, it kinda matches how I perceive footage with similar cars on the limit, they do slip a lot and often, they likes to drive with significant opposite lock, but they don't feel like drift cars. I think I might slightly reduce rear tires grip drop within overheating.

    If you don't oversteer at exits, you might not be driving fast enough there.

    I have left rather free setup choices now. For example differential setting may be removed in the future, but now they are necessary. I suppose you have pumped up diff coast setting, I have not tried it on high setting yet. Originally it should be low and not adjustable afaik.

    Car is ~500 bhp and weights 1000-1100, don't remember exactly. It also may have too little drivetrain power loss now. It is probably 1-3% too slow now, still. But might also just be me that is 1-3% too slow. Car was supposed to beat small block GT40. Big block GT40 or T70 should beat this car at pure pace. As for example from Goodwood, fastest ever small block GT40 laptimes were ~1'20 (high), iiric, and T70 ~1'17, iiric. So I adjust the car to be 1'21'5-1'22 in my control. I would like to give a link for testing Goodwood, but I didn't got it to work in main game yet. Tire grip has HUUUUUGE impact on pace, I can not even touch the base grip, and simply increase kinetic friction (for slides) + minimizing grip loss due to overheating and car would become instantly much faster, I could make it no brainer at all, you could just slide through all track still having grip as the car slides it would seem like you are over the limit, but you'd be under control the whole time.

    I also test in Nordschleife, it should do 7'35-7'40 there, I did 7'42 so far, felt very intense and I was happy about that. Can't wait for legit official Nordschleife. Originally similar cars did ~8'20 in Nords, but then the track used to be about 40s longer. I also like to test at 60s or 1991 Le Mans, Oulton Park, Sebring and more tracks... Recently had a lot of fun at Mid-Ohio, no data for pace, but it is good to try handling everywhere and anywhere. I do also usually test in wet conditions too, pure wet, or drying wet, because same tires has to work in the wet. I did not know there is classic Kyalami for rF2, I know that T1 of Kyalami is easy to enter sideways lol
     
    Last edited: Feb 23, 2019
    jimagn and pkelly like this.
  19. jimagn

    jimagn Registered

    Joined:
    Oct 6, 2010
    Messages:
    164
    Likes Received:
    37
    I ran the 1967 Howston Group 4 Mark III at Kyalami and Rouen. It is significantly faster than the XJ13, but has a big block engine. I thought since the year was correct, it might be a good comparison. Not. Maybe the '68 Howston Group 4 would be a better comparison.

    I like the oversteer on entry, though I did increase the coast side of the diff a little. I'll continue to test at Oulton, Sebring, LeMans 60s, and maybe Nordschleife. I'll see if I can provoke more oversteer on exit. I also decreased the tire pressures as low as they can go which I always do with the current tire model to try to get more even tire temps across the contact patch. I think I also decreased the rear bar a little.
     
  20. mantasisg

    mantasisg Registered

    Joined:
    Aug 17, 2016
    Messages:
    413
    Likes Received:
    637
    Tracks with a lot of straights where top speed tops out for a long time will be slower than they should be because top speed should be higher, I guess it is too low because I can't get gears right, maybe.

    I should try lowest pressures too. I don't believe that lowest pressures gives best speed, they should increase rolling resistance, increase wear, make tires more prone to overheat, increase contact pressure on edges and decrease in the middle (which should be killer in the wet), make tires less responsive...

    Yesterday I tested plenty of cars in newly released Paul Ricard, I really like the track, only rain things looks not to be implemented. I liked the track with various cars, and it was awesome, but I didn't like it with my Jaguar, then tested in Oulton and it was fine. I wonder if it is something related with track layout only, or somethign about surface. Though I sometimes suspect that rF2 does perform physics/FFB slightly differently, but that might as well be just me crazy.
     

Share This Page