Discussion in 'Third Party Content' started by woochoo, Jan 5, 2015.
Wonderful, fantastic Work! Thanks so much !
I had a request for a non-steam version of the mod. This is not an update. It's the same as the workshop version.
Non-Steam v2.0 rfcmp: https://drive.google.com/open?id=1pOvkVZ2SqXB9Dlt5PPnP9R_RvHmt-U5Z
As for the Lotus 47/ Europa. My wife and I almost bought one before we had a family, but realized how impractical it would be. Looking over at other cars' tires out the side windows was entertaining. Now I finally have a 2003 MR2 Spyder in Solar yellow. It's as close as I could get to that Europa.
Thats a car that I would pick-up for a second car as well if I could afford to buy it and then take care of it properly. I think MR2 is fantastic car. Also Miata of course, but I would give priority to MR2.
It is awesome how both of those are so similar to 60s Lotus cars. Miata is like Elan, and MR2 is like Europa.
A little update on a cars. I am in sort of R&D/Education phase, I am figuring out how to deal with different dimensions of wheels, different and very different from currently used ones. Eventually it also results in having to understand chassis.ini better and there is some tricky things. For tires it seems I will have to get advanced up to using ttool maybe - which is scary to me, especially while it is said everywhere that it takes multiple hours to generate the lookup table. Not really motivating to even try if there is at least some doubt about something, and there is of course. Technically I need a lookup table for different dimensions, eventually also need correct masses and inertia values.
I have to admit, I had not even tried to run a race with these cars yet, just hot-lapping. But once I did, I found at least a couple of items to look at.
I accidentally pressed the wrong key for a screenshot and got the traction control menu. Those should be disabled shouldn't they?
In the setups, I went full tanks, 3 stops. The first two stops were 159L maxed and 159L also full tank. The 3rd stop went up to 169L for an extra lap or two.(le Grand Circuit)
My AI controlled co-driver actually came in properly for fuel and tires, without my having to assume control. You guys must have correctly configured the rcd files, thanks!
I just assume that you drive Chappel Auto XF. I jsut looked up into its physics, and there is indeed little typo for it's 3rd fuel stop setting. I corrected it, thanks for telling.
Not sure about TC, never payed any attention to it, must be same values as for Howston cars. Any chance that TC that you have activated is not physics related ? Personally I have unassigned all F row help buttons, because it happens way to often that they gets activated unintentionally, and the worst part is that then you might drive with those on for weeks not realizing it.
I did take try to get fuel consumption at LGC working ok. Glad that you've had a good result!
I think the TC would be the standard in-game feature. I haven't looked to see if i could set all the values to 0 or whatever, but it's a game accessibility feature rather than sim-related, so it should probably stay with default availability and functioning. You just have to press the right screenshot key
Thanks. I'll double check all that next time I boot up.
Hi. For the past days, perhaps it was a couple of weeks already, I got myself very much involved into improving physics for various cars and improving my abilities as physics maker little bit by little bit. Eventually I focused on cars from this pack, making them be more like their real life versions.
TL;DR - I am having good time working on these cars physics, and I am happy with increasingly better results. And I want to tell you about it.
I started with Prosch GT19 and did most of work with it so far, ending with a car that drives a lot more like actual 1967 Porsche 911s. I made a fix for rear semi trailing arms geometry shifting their rotational axis to more correct angle. Also fixed some big time errors in geometry file, that seems like no one actually noticed yet, you are doing bad job. Additionally found that model can take actually correct short wheelbase, so I was able to reduce it by 9cm. Big changes to aero and tires/wheels. Initially I was right to think that this car should have a lot of lift at rear end, not so right to assume that at the front end it would be nearly neutral. By reading this article I found surprising aerodynamic figures: https://www.caranddriver.com/review...-targa-911s-carrera-archived-comparison-test/, these are 1974 cars, apparently front bumper has something that reminds tiny splitter, but I assumed that 911S from 1967 would have very similar figures. So eventually testing with DAMPlugin showed that I had to use actually twice as much lift at rear end, and very similar amount of that on front end. At the end it is very balanced car - equally bad at both ends, but it makes it great. This car has very reactive steering, if needed can use micro scandinavian flick or sudden turn in while still on brakes. Practically as soon as tires wear a bit it is mostly either understeer, oversteer or all wheels sliding - then you know that you are fast enough through a turn with sixties 911s. First and last - the tires. I used redapg tools to generate different sized wheels, I don't know for sure what it does, perhaps just changes nods locations, but it seems to be ok for not too much different tires, at first the tire didn't feel very well, but I discovered that I just had to redo realtime section of TGM to properly fit new size of tire, and it is pretty awesome now, at least in practice. This narrow tire is rather forgiving and predictable to suit better natural behavior of 911s, its planted enough but isn't too much resistant to slide and is not too sudden to grip back up, though not too lazy either. If you read the article above, you'll see that if you slide with Prosch don't lift off throttle abruptly.
I got one new tire size for Galoppare GT127, and it turned out to be very good upgrade to stronger Ferrari 275 GTB way of handling. For Prosch it is 6.5in. tire, for Galoppare it is 8.1in.. For a reference before it was default Howston tire sizes: 9.1in. at front and 12in. at rear iirc. I also made a tire set for 9.1inch tires front and rear. I thought thats enough for now. Naturally, I had to rebalance cars that changed tire sets as the way front/rear worked changed, mostly I just had to make front harder, or rear softer. It didn't take much for Gallopare to become much better, just new tire size, adjusting realtime TGM section for tire size, slightly stiffer front end.
I did a tire for Martin Franklin that replicates Mini Marcos, but it turned out to be breaking physics, much too different tire size. But I did some changes to its aero and used tires of Prosch, so you don't have to drive Mini with same tires and wheels like Ford GT40. I did changes to geometry as well, although I made swing axle for now, will need to move to solid axle as I'll learn hopefully.
Shelton Forster STG35 got 9inch wheels all aboard, much more like Shelby Mustang GT350, although probably not the exact size. Car eventually also needed slight adjustment to rebalance the car. Also lowered it a bit, as it was highest car. Now it is almost too good as good as Galoparre, however it also needs solid rear axle, so it would probably bring it down a bit. Interestingly there was a project back in the day to make independent rear suspension for GT350, but Ford turned it down. Actually same people that were building those cars in sixties did a continuation project these days, and made it with independent suspension at rear, it turned out to have bump steer, I don't know if they got solved that since that test at willow springs some years ago.
A lot of cars got new wheels set, I mostly have worked on those yet, cars that uses same set as before haven't got as much attention yet, though I did reduce sharpness of widest tire a bit. Also just to mention increased wet surface cooling a lot for all tires.
Most of the fastest cars probably has too optimistic downforce, except maybe Chappel Auto - Chaparral 2F, Forster x4 - GT40MK4 and Lillid - Lola T70 representing cars, so it will need to be looked at. But Forster (Ford GT40 MK1 and MK2) and Galoppare (Ferrari 330P4) protos likely has too high downforce settings IMO
I have been increasing steering wheel lock to lock turns for many cars. For example I had it overlooked for Fantini Abiati - Abarth 1300OTC, Anton Headly - Austin Healey SLM. For some reason original Howstons 1.75 turns lock to lock slipped through for those cars, and with wide tires they were rubbish. Now they makes much better sense.
Narrower tires and larger steering wheel lock to lock also turned to be pretty awesome for Prosch prototypes, that should be like Porsche 906, 907 and 910. Less biting rear end feels way more correct, smoother and brings quite a bit of soul that belong to those cars, I think.
What about some other cars, plenty got new tires and needed some new balance, and feels greatest so far. Except maybe Cheltonvie Spear - Corvette Stingray 427, that one seems worse somehow.
I discovered that new Teretonga track is really awesome for low-medium power cars. Had pleasant testing with Ardouin Cars (Alpines). Also lightest car from the pack wooden monocoque RN Motors - Costin Nathan really needed small tires.
I might need to do some research for Lighthouse GTLE - Lotus 47 and Maitre 360M - Matra 630M. They might need different tire combination maybe, although both seems to have had really wide rear tires.
In conclusion still plenty of research, discovering, learning, tweaking, testing. But it is easy to say that it pays off when it provides increasingly better experience at driving most awesome cars in history simulated, I hope, reasonably well. If there are at least a couple of you that can join this, then it is good enough for me.
Matra and Lotus:
There is one basic issue I come at recently. I used to set cold tire pressures really low and make it default, so the grip would be what I think can be considered as normal for these cars, but it seems to cause too much deflection, so I am considering increasing tire grip accordingly and increasing tire pressures up, now most cars has ~ 100 kpi, which is pretty low, I am considerign toping them up by 40kpi, I wonder how much then I should increase friction, I am also afraid to decalibrate tires that feels good now, so not sure. I wonder how much of friction coeficient would be equal to 40kpi pressure difference at same temperature.
Another basic issue with not entirely unique tires that I generated with redapg tools, is that I can't fully control their mass and inertia, and of course everything what is in lookup content - ttool output. That of course results in some inaccuracies, and for example of min marcos - errors. But for similar tires it doesn't feel like much of an issue, skipping god knows how much ttool studying and hundred of hours generating tires with it.
Excellent work, as always. Thank you for your continued attention.
I participated in an event recently that used this mod at Le Grand Circuit.'67. Another driver and I chose the Chappel Auto. His setup used the shorter gear ratio set that is available. Amazingly the car never blew up while running at redline rpms for over five laps. We were seeing 330 kph. Since this is not very realistic, I would like to suggest that you make the car much more fragile when abused like that. I'd also like to see a gear set with ratios somewhere between the 2 that are currently offered. The taller gear set seems too tall: The car never reaches its potential top speed and doesn't seem to reach peak horsepower. Thanks for your consideration.
I made virtually the same comment several months back, but shortly thereafter I made a lower stiffer setup that allowed me to extract the max from that long long top gear.
Thanks for feedback. I did not know that it was possible to do 330km/h with Chaparral. I'll try to adjust damage, will do some research about gear ratios, and it seems that it is possible to lower it way too much.
Got first try success adding solid rear axle to Mustang - Shelton Forster STG35 today. Pretty happy about it, took example from Boxmaster actually, and modified it slightly. Car felt great from first go, and even pace was just as good as with independent rear suspension, although a bit more rough and not very kind to unloaded tire, that induces wheelspin rather easily, maybe I should make differential stronger, another thing to research... Edit: Ok it had something called NoSpin Detroit Locker, that was basically LSD for trucks, I'll just assume its ~95% and high preload. Another interesting detail - steering wheel turns lock to lock are 3.75, which is too much for most FFB steeering wheels, not sure if to use it all.
Next will be Mini Marcos - Martin Franklin. I wonder if there are more cars with solid rear axle in there.
Edit: Good thing I checked Mini suspension again, turns out it is not solid axle, it is trailing arm set. Another interesting detail - hidrolastic system, didn't know about that, but in this example bellow it is just solid peace of rubber. By the way, Austin Healey has solid rear axle.
Mini rear frame with trailing arm visible:
I am not sure if you'd like to hear more often from me, perhaps I should keep on brief and as short as possible, but still would like to communicate something.
So I look last post was April 7, my god, time certainly runs fast when you are busy and very interested. Can't believe almost two weeks went by. I have almost only worked on these physics this whole time, it is a joy, but also I admit definitely got too addicted. 60s is best decade of interesting cars IMO
I will add a file to this post that is supposed to be very amateurish changelog of work in progress stuff with typos and everything, there is most of stuff mentioned, although not all. And it is for now, there will be more work, before mod will be actually updated, which I am not sure when actually, but it is not like very soon. I think. probably.
During past days I mostly focused on Mustang that is Shelton Forster car, I added solid rear axle to it rather early before, but it really wasn't very awesome, it wasn't broken worked alright, but it wasn't good to drive, was pretty difficult to survive any slide. I did quite a lot of tweaks, including getting better at rear solid axle, and suddenly it switched from horribly unforgiving to too forgiving car, so I reverted some aero changed that I definitely made too easy before. I have basis rear solid axle from Boxmaster, I left it as 3link geometry with panhard rod, although with many changes of course. I was about to look if I could make it two link without panhard rod, but at some place I found info that Shelby switched default Mustang rear axle which IIRC was from Ford Falcon, to rear axle of Ford Galaxie. I googled it and it appeared to be three link with panhard rod, so it was awesome, and it works well in sim, although I am not sure if I placed panhard rod really well, cause there is ~3cm/an inch of lateral movement. I also think that bigger amount of differential pump is good for this car, but I am not certain how it exactly works, and it would be awesome if someone could explain to know if it actually is a thing for this car.
Here is Ford Galaxie rear axle, honestly I am not entirelly sure if they took it with middle link and tracking bar, but works well.
I also was adjusting rear pushrod a bit for more of characteristic mustang rear end roll, I overdone it a bit, and then it became Stadium Super Truck:
Of course I fixed it. Also in screenshot it does not appear like solid rear axle, because replays use .pm file which I have not worked on, I only worked on chassis.ini file.
One interesting stupid little story. It sometimes happens somehow that cars feel totally perfect and then other time for some reason - not so much. It happened one evening that I disliked some behaviour of few cars, then made changes in files, but forgot to update .mas file before updating package (I skip devmode and go stright to game usually). So next morning cars are absolutely perfect (same cars, same track, same conditions), and I love the changes, till I see in setup that some values that should have been different were same, and then realised that it was massive placebo. However, I made a conclusion that being very tired at night, and not in best mood might have great impact on perception of cars handling, something to think about for all simracers ? Other thing I though about that due some unknown rF2 might run slightly differently sometimes, but it seems highly unlikely, also I have not estimated input lag in any way, and can't tell if there was any. Very likely being very tired and not happy - more input lag, perhaps literally haha.
Another car that required quite a bit of tinkering was Austin Healey - "Anton Headley" car. It was mostly because of solid rear axle too. I think I did solved it, at least the big part of it. But it is known that the "small part" is always the most work. It is fun little car, and should be playful. I think with solid rear axle it probably will never feel as awesome as I had it with independent rear suspension at one point, but it is very fun anyway.
Also had good progress with Abarth the "Abbiati" car. Turns out there is way more torque in this car than we thought, it should actually be capable of 246 km/h and thats fast, it is actually slightly faster car than Porsche 911S - "Prosch GT19", and both of them are rear engine. Speaking of Prosch GT19, had some good improvement there as well, some better steering feel and front end comfort, sometimes I feel like it is really representing Porsche 911s as well as I hoped it will, but sometimes feels like there is still something, but not completely sure what - it is very fun car it feels mad when it is much over the edge, and it feels like thats where it belongs, but sometimes it gets slightly rough and way less rewarding perhaps when in too big angle and with too much overheated rear tires (although it seems pretty much normal to have its rear tires overheated by some amount, when putting multiple fastest laps constantly sideways everywhere), but perhaps it is how it should be. As long as it is in check it is brilliant. Perhaps it requires a cooldown lap every few super oversteer laps. But before you say it - it doesn't feel quite right to make tires cooling down even better, although perhaps could be just so slightly better cooling. Anyone interested in testing ?
I haven't been focusing on fastest cars yet. But I did go through engine specs of GT40 MK1 the Forster 41 in game, also found some info which turned out that currently it was almost hundred killos too light, also it had too much torque. So less power and more weight. How WE are going to spin tires out of Goodwood chicane now ? Well probably they have slightly more HP in modern historics, better oil and some better other replacables...,however they run less wide tires in Goodwood. So I added a set of tires that rear and front tires would be same for this mod, although I might make fronts still more narrower. This has two benefits - now it can spin tires rather well out of something like Goodwood chicane, and it also serves as rain wheels set, since wide tires are known not to be good for wet track, and narrower tires should be better.
I also turned my attention back to aerodynamics of these cars. I think I had too much downforce to many of those. I think Lola (Lilid) and Chaparral (Chappele) are one of those with fairly good settings. But several other cars probably had just about zero downforce, slight lift or slight downforce, lets talk about something sub plus or minus 50kg at "speed". I found little bit of info that Lola T70 breadvan body type could achieve 200kg of downforce, but with expense of a lot of drag, probably with this wing type, by the way it makes similar shape to Porsche 917K, and also similar downforce figure actually:
It is also mentioned (https://www.conceptcarz.com/vehicle/z11341/lola-t70-mkii.aspx) that GT40 had 300kg ! of lift. I suppose it was so in 1964, no surprise it wasn't so great at first. I just assume that in 1965 (GT40 MK1) had much closer to 0 of aerodynamic lift, but probably still no downforce, although not sure.
Here is great video talking with Jim Hall, according to him all aero then was about reducing lift, not many actually thought about flipping it all the way over to making downforce:
I've stood at the very spot Jim Hall was standing. at the Petroleum Institute Museum. I took hundreds of photos and video, but lost them all. Moving 12 times in 11 years will fo that.
Sorry to hear about the media loss dave. I was there also, thanks to the patience and forebearance of my wife and the good friends with whom we were traveling. I can share some photos if you want. Not hundreds. And no video. I was too busy being there to think much about photos. Too bad the museum is out in the middle of nowhere. Not conveniently located. But worth it.
Love your posts, manatsisg. So much info. I'm glad you have an intermittent obsession with these cars.
I drove 6 hrs each way all by myself. I was very very tired on the way back. Odessa Texas is a long long way from anywhere. I would love to see your images!
Early Minis had the rubber suspension, it changed to Hydrolastic around 1967 I think, the same system was used on all BMC/BL cars at the time. No one used Hydrolastic for racing, it was a simple job to take it out and put the rubber system back in. Your picture is of a racing set up but it's essentially the same as the road car version. Quite a few race Minis did replace it with a beam axle which improved the handling and was a lot lighter.
Early Sprites had 1/4 elliptic rear springs that were only mounted at the front end of the spring. Later they switched to 1/2 elliptic mounted at both ends which improved the handling. The front suspension top arm was a lever arm shock absorber. Again, fitting a top wishbone and coilovers improved the handling.
Wow Dave, you prompted me to look up where you live (and also Paris, Texas for movie fans)
Yep, that is some drive. Just as well you have sim practice!
At the time we were in Killeen TX. Right near the center of the state. During that drive I saw my one and only Roadrunner bird. plenty of dead roadkill skunks and around Killeen, Condor/Vultures by the hundreds. Now I'm up in the Northern half of Idaho. Just across the border from Spokane Wa. (pronounced SPO-can) right on the shoulders of the Rocky Mtns.
Separate names with a comma.