Dirt Racing Improvements

After reading about the "real road" in RF2 I started to think. This can be the best thing we have ever seen in online racing, I only hope it can be customized for dirt racing as well. Dirt at times can be completely the opposite of tarmac. In most cases on dirt the line that most of the cars are driving on becomes much slicker than the rest of the track, then after some time depending on the track conditions and the weather it can then start to take "rubber" and become the preferred line. Another thing with this is we are doing sprint type races so this will all need to happen in a quick amount of time. I really hope this can be flexible enough to make these things happen, if so online racing will never be the same...

 

If ISI incorporates these improvements, we are going to experience "Next-Gen" dirt oval sim racing. I've played many open-sourced games in my online-career, and have seen many gamming communities. RF1's Dirt Oval community has by far the most aspiration and dedication i've seen in developers. I personally believe this community is capable of some major accomplishments in the Sim-Racing World.

 

I would like to be able to get the tracks to have light reflections like this. Also if there are any guru's that can tell it can be done now, please advise me how. See Picture.

 

I agree with everything stated here especially the thing Bill said about the light reflections. Would be nice to see the track change overtime as the race went on. Also would be cool to see a tear-off feature of some sort when running in-car.

 

Also agree with the posts in this thread. Some things mentioned here I see as being a lot more benificial than others, but all would make for better racing/driving.

One thing not yet mentioned I would love to see is 'cockpit adjustable shock/s' like lots of real life dirt oval cars have.

Very pleased to see this thread being used so well by all the dirt communities showing unity.

 

Agree with all said. ISI, a great % of all RF1 racers are livin on the dirt. Oval that is.

 

That would be great if RBR community can move over to rFactor 2 at last. Too much expecting and waiting over 6 years.

 

Is there a way to better quantify this information? How many laps it takes, what type of cars, how does air temp affect this, what % more or less grip is there, ....
btw, thanks for all the great comments...

 

Sprint Cars -

Big Block Modifieds -

UMP Modifieds -

Latemodels -

Gjon, there are some video clips of some of our favorite dirt oval racing we do in rFactor 1. Once Bill see's this post, he will be able to better answer your questions in much better detail then I can. But this will give you a small preview of what we do.

Thanks,
Shawn Blankenhorn

 

Basic Dirt Racing Info

Hot Laps = 3-5 laps, surface = wet/tacky (1.00) Racing groove wet/tacky (1.00)
Heats = 10-12 laps, surface = tacky (0.94) Race groove dry (0.90)
B-main = 10-12 laps, surface = dry (0.88) Race groove dry (0.84)
A-main = 20-40 laps, surface = dry/ slick (0.84) Race groove rubber (0.90)

Depending on the weather a track could dry out after heats during high temps or stay tacky all night during low temps.


We could do part of this in RF1 if invisible objects were truly invisible by switching tracks, hot laps= track1, heats= track 2 and so on, this only changes the texture and not grips.

I hope this help.

 

Alot of the dirt racing stuff will carry over into kart racing too, karts run tire pressure anywhere from 2PSI to 20PSI depending on track conditions. a lot of people kart race in rFactor. the kind of stuff us kart racers do varies a LOT!

Restrictor plate karts-

Unrestricted karts-

Big block limited karts-

Open engine karts-

this is just helping build onto ShawnB's explanation of how the tracks change as dirt races progress.

 

I agree with everything said 100%.
Also make the car setup like a real car. For Late Models. No trackbars and crap. Make it where you can change bar lenghts and make the setups dirt based. Not asphalt.

 

Previous posts have pretty much outlined the tribulations that dirt modders have with rFactor. Looking to rF2 I would like to comment and elaborate on the previous posts and add my own 2c.

Physics - The biggest impediment to realistic dirt physics is having to use double A wishbone suspension on the front. For whatever reason, double A suspension develops a lot more grip than the solid rear axle suspension. The result is high grip front suspension and low grip rear suspension. The modders then use overly stiff springs on the front to balance the front\rear grip which results in unrealistic roll rates for the cars and from there it gets worse. So solid front axle suspension is pretty important. The latest incarnation of the KDSR Midgets has an approximated solid front axle but it is still a fudge.

I believe it is possible to model a watts linkage or a Jacobs Ladder + birdcage suspension on the rear of the car. I don't see any point unless we can also accurately model the front suspension.

rF2 is coming with Spring Tension tables which take care of the torsion arm modelling. Anyone know what -kx is for a torsion arm? I will have to dig out the old physics textbooks lol!

Chassis torsion is not modeled that I can tell. This prevents the car lifting it's front left wheel which is characteristic of many dirt cars exiting a corner. Personally I don't give a hoot, but many players would love to see it, so why not put it in. In Midget car racing, chassis torsion\stiffness is certainly the difference between a good\bad car in the slick and it is adjustable on some chassis'

Wheel offset (spoke positioning within the rim) is not able to be modelled. I think rFactor just takes the centre of mass of the wheel and designates that as the centre of the wheel. I don't fully understand how it works or what it is intended to achieve, but it is a factor that dirt racers take into consideration.

Tire Modelling comes down to modders knowledge of the tires. I believe the tools are all there, in rF1, to make accurate models of dirt tires. The problem is where do we get the knowledge and hard numbers from? It would be interesting to know where the venom dirt tire curves came from. Extensive Internet searching revealed nothing pertaining to dirt tire slip curves. I have found a couple of "expert" opinions, in instruction books, on tire slip curves, but there seems to be little public (internet) knowledge about how dirt tires generate their grip.

Tire temps themselves are hard to quantify. Most dirt classes prohibit on board data recording preventing on board pyrometers. The few pit crews that do use thermometers have to wait for the track gate to open, the car to exit (often having to wait for an available push car) and then drive through a big puddle on the way back to the pits. Dirt car crew gauge tire temps on the condition of the tire after the race. To much blistering = too hot. Not enough scuffing = too cold. Tire temp is then controlled by siping(?) and or tire pressures.

With regards to making the dirt look better graphically, I think that a large part of the issue here is modellers ability. I know that I struggle to know what to change when texturing a car. I imagine track building is just as frustrating. There is a lot of original work that gives fantastic looking results. Eastern Creek Laser has some awesome dirt textures in it. Shame it is a tarmac track lol! I think the solution here is better modder knowledge, both car and track.

Dynamic car \ track. Probably the biggest impediment to dirt racing. In car adjustments to shocks\wings being the most obvious. Dynamic track changes during a race also. Previous posts have covered this well, just want to add my vote. In car suspension adjustments to cope with changing track conditions would be an interesting (complicated) experience.

Race time D,C,B,A Feature race structure would be cool. Don't forget to include double A, double B etc all the way down to double L(?) mains ala the last Chili Bowl. Please don't limit it to D.

 

Doh - forgot the one that generates the most flak.

AI that don't freak out as soon as they start skidding!

 

many dirt classes prohibit mirrors so the only way to know where your opponent is (behind you) is by sound. Left\Right stereo is too limiting and a modern game release needs to support the various speaker configurations around 2.1, 5.1, 7.1 dts etc etc.

 

a dirt track is going to pretty much work opposite of asphalt....the more a particular line is ran...the quicker it is going to dry out and lose grip... instead of marbles forming...a cushion will start to form which is usually a good place to find grip ( as long as you dont jump it)... some tracks will dry out then start to take rubber and that line would gain grip

im sure someone else will post a more detailed explanation than this

 

I only run dirt mods on rfactor currently. I would love to see the improvements mentioned in this thread incorporated into rfactor 2.

 

Gjon,

Just to clear up a few things here for you with all these suggestions for rF2 and ISI.

Track Conditions in Dirt:
Dirt tracks vary all over the world. And one main thing that may help you guys out, its not actually dirt tracks, rather its Clay...

Ok, tracks change quickly in dirt racing in one night. Tracks start out freshly watered by water trucks. With the track being watered down and rutt free, they are very tacky and fast. As racing progresses just a few laps, the track loses water, dries out, and gets slick. Now there are many factors in how quickly they "Slick Over". Ambient Temperature, Wind, and Racing Lines.

Wind will cause a track to dry out and get very slick and dusty in a very short period of time. (1-2 Heat Races) (Heat Races are generally 8-12 Laps depending on Venue and Sanction).

Ambient Temperature will cause a track to either stay tacky or slick over as well. The cooler it is out, the tackier the track will be. If its hot out, tracks will tend to Slick and dry out much faster. Now you add wind into the equation on a hot night, and it will dry the tracks out even faster.

Now as far as grip goes on a Clay (Dirt) track... Tacky tracks will have high grip, low temperature, and less tire wear and heating of tires. Dry Slick tracks will have low grip, higher temperature, more tire wear, and higher tire temps.

As cars race in one groove, that groove will DRY out, lose grip and push the "cushion" up the track. If cars continue to race this groove, it will "rubber-up" and gain grip, but wear tires faster.

Cushion
Now for the "cushion" LOL.. The cushion is the loose clay that gets pushed up the track. As the cushion gets pushed up, cars will slowly move there back tires into this loose cushion to flatten and harden the loose clay and also move the racing line up higher to create more places to race. Now if a car tries to put there entire car in the fresh cushion the car will lose all grip, push, and more then likely crash into the wall.

I hope that part helps ISI out in the Dynamic Track options that you may give us Dirt Guys.

Now as far as gameplay goes. I don't think everything being asked here is needed fully. As far as heat and feature structure goes we really don't need to be able to pre-program in 6 Mains. We do it now in rF1 manually. But being able to change the track dynamics between each event would be great.

Cautions/Restarts - One popular thing in Dirt Oval Racing is what's called the Delaware Restart. What is Delaware Restart? This style restart gives the leader of the race to start on the front-row by himself, then 2nd place, and third place are on row two. But 2nd place has the option to choose Inside or Outside.

Not to jump back to a subject and cause some confusion, but want to discuss the dynamic track options again.
If we can have the ability to not have the track conditions change on a "restartwarmup" command that would be great. That way after we run Heat 1 of 4 the track will be how it was at the end of Heat 1 for the start of Heat 2. Then Heat 3 will be how the end of Heat 2 was, and so on so fourth. Then for the start of the Feature (Main Race) we can do a "Restart-Weekend" Command, and it would be like water trucks coming out, and making a somewhat Fresh (Green) Track.

Basically have restart warmup keep track dynamics in effect of previous sessions. Then have Restart weekend have it reset track dynamics.

Would also love to see rut effects in the TDF. Dirt Cars will tear up clay tracks and cause ruts, mainly in the corners. These ruts could also change racing lines, setups, and tire choices. That is one thing that would really help us out as well, but may be somewhat difficult to pull off.

I know we are asking a lot from ISI, and especially late in the creation of rF2, so anything you can give us will be greatly appreciated, and if you can fill us in on anything you may be working on.

Also if you would like to see some Dirt Oval Racing in rF1 and how we do things, feel free to contact me.

If you have any questions, don't hesitate to contact me by email. I am more then happy to sit down and talk to you on a personal basis to help clear things up.


Thanks,
Shawn Blankenhorn
Open Header Motorsports
SimDirt Studios

 

is there anyway the rear axle could be modeled so it stays/pivots with the hubs

 

Thanks for your interest.

All the dirt oval scene has the same characteristics as far as tracks go within reason. The issue programming for this is a few things that maybe can be addressed.

We race sprint style races, so most races are at maximum 30-40 laps for a feature event. The tracks progress through out the night, that is what makes this tougher. Currently we build 2 tracks for each track we make, one we call tacky and one we call slick. We Time trial and heat race on the tacky tracks then run the A main, B main, etc on the slick tracks. Unless there is an option to keep the track the way it was left the previous race in a server it will be hard to let the track progress as each race runs, although this would be the most ideal situation for us.

A typical dirt track can slow down over 2 seconds per lap as the night goes on, but there is always an exception. Some tracks will take "rubber" after they slick off and actually get much faster at the end of the night. Some tracks are more known for this then others, so again the ideal situation would be a way to script which one would happen on a particular track. Now how fast does the track change during a typical race. This can happen pretty quick as well. In my opinion a good compromise for a 30 lap race would be to have the track slow down about 1 second in lap time, this is on a 1/2 mile track. What we did to get us real close which will help in determining a TDF grip would be take a lap time, for the sprint cars at Knoxville for example they run the track in 15 seconds, then get the percentage it would take to go 1 second slower and multiply that by the TDF value and this always got us real close.

Weather does play a big role in all of this too. If it is hot and the sun is out the track will dry out much quicker and get slicker and take rubber more easily. Out west were it rains all the time, there tracks tend to have lots of grip all night long.

I have been apart of making most of the dirt oval tracks that are ran in rfactor and we have done a lot of stuff with grip locations to make the racing more real. To get this to work correctly it will be a major improvement in realism and make the racing much more competitive then it is now. We have many real world sprint car drivers that work with our mod team to develop our sprint cars and this is always the biggest complaint we have, is the tracks are always the same.

If I did not explain this well enough please contact me as well. Also feel free to download a few tracks from the website to see how we make a slick track vs tacky track. If we just had the transition from tacky to slick based on where guys were running on the track this in itself would be a huge improvement, but of course I would like to see it go further then that if possible as well.

Also, if you can tell us. How will this work, will the grips be altered based on vertex locations on a track? Trying to figure out how you are assigning new grip locations to a track.

Thanks again,

Bill Malicoat

www.bozracing.com