04.02.2022 Category: Site news
Episode 5: Tire physics model and ‘feel’
This is probably the most complex part of the VRC physics as tires are kind of ‘black magic’, and round indeed, Ronnie Ton was right after all… Vehicle dynamics, aerodynamics, combustion engines can be described with formulas that have been developed over time. Tire dynamics is a totally different ‘animal’ so to say, and there are many ways to go about it.
The testing data we obtained from Todd and Doug’s extensive tire tests was essential to understand how the tire, in the first instance a foam tire, would interact with the road surface, in case of VRC asphalt for outdoor on-road tracks, carpet for indoor tracks and later dirt for off-road tracks. Of course, we didn’t have data for all tire-surface combinations nor for all kind of tires used in r/c, but what Todd had learned from the data of foam tires and rubber tires laid the basis for the tire model used in VRC.
Obviously, a tire physics model for tire research purposes is something completely different than a tire physics model to work in a sim. For the first you can use unlimited computer power to calculate results very accurately, and time plays no factor. In a sim the tire physics model needs to come up with a result every 0.002 or 0.003 of a second, in other words: (almost) real time. That requires a totally different approach, the tire physics model must be both as fast and computer resource efficient as possible and produce results which ‘feel’ right. And how do you calculate what forces are generated through the contact path of the tire when load, angles, width, and deflection are constantly changing because of the vehicle dynamics at work? It was an immense difficult task for Todd to develop such a tire physics model from scratch and no wonder Todd spent almost 2 years on the groundwork for the tire physics engine. And on top of this, Todd had no idea what sort of result we were looking for as he had zero experience with r/c cars.
Todd came on the VRC project through Lothar Pantel somewhere in 2001. In 2002 Todd and I met for the very first time during the 2002 IFMAR Worlds 1:10 sedan in Cincinnati, the first of many many to follow. I thought it was useful for Todd to finally get to see an r/c car race and watch how these cars behave. Todd flew over from Minneapolis, we got acquainted and we had extensive talks about car and tire physics. He also met Michael Salven, Serpent top driver and Chief designer, who provided essential feedback to Todd so he could start to understand what sort of ‘feel’ we were looking for. It was the first of many meetings, in Heemstede at the Serpent factory where the discussions continued with Michael and Rene Cornella and later also with Julius Kolff on 1:10 electric on-road rubber tires. As the Heemstede track was just around the corner Todd could finally lay his hands on a real r/c car and experience ‘feel’ for the very first time himself. Interestingly, Todd had learned to drive an r/c car by using the development version of VRC v3 and when he drove a real r/c car, I believe it was Michael’s car, he was able to drive it quite fast around the Heemstede track, of course keeping enough margin not to crash the car first time out. This ‘drive yourself’ experience was essential for Todd to understand what ‘feel’ means when he received feedback from Michael, Rene and Julius when they got involved in testing tires.
Rene Cornella was Pro Engineer designer at Serpent and had a very good understanding of vehicle dynamics and was most helpful in developing the chassis and tires. At one of Todd’s stays in Heemstede he stayed in Rene’s apartment for several weeks. Todd and Rene were also avid iRacing enthusiasts and were sharing tips on improving car setups all the time. Todd and Rene spoke ‘the same language’ so to say when it came to discussing Virtual RC v3 physics issues.
Of course, to test anything, we needed a running chassis, engines, tires and a body, and a track to start testing. Tony West had modelled the Heemstede track and the Serpent 950 1:8 nitro chassis plus a body, wheels and tires so we could run the car on a track and start testing. Can you imagine how excited we all were when we drove the first laps on Heemstede? It was only the start of a long and exhausting process of finetuning the chassis, the tires and the engine. And you can probably also imagine that we had a ‘chicken or egg’ problem if the car wasn’t working as expected: was it the chassis (vehicle dynamics) or was it the tires (tire model or parameters). The start of this process was not easy, countless hours were spent on testing the chassis and the tires.
I have so far skipped the development of the vehicle dynamics model which Todd had started to work on from day one. The chassis interacts with the tires, and the tires interact with the chassis and the ground surface. The chassis physics model that Todd designed was more or less a 2D representation of the suspension, suspension movements for on-road chassis are relatively small so 2D would work OK for now. The only 3rd dimension needed was to calculate the effects of the caster angle on the tire contact path so that was dealt with separately. The 1:8 and 1:10 scale on-road cars (also electric) were all modelled using a 2D suspension configuration. For the 1:12 electric chassis Todd modelled the rocking rear end and front suspension just like in the real chassis. When we started working on VRC Pro which would include off-road chassis, Todd redesigned the suspension geometry completely to a 3D model, a very complex but significant improvement on the vehicle dynamics. More on that when I get to VRC Pro development.
In those days communicating with each other wasn’t as easy as it is now with Skype, Whatsapp etc. I don’t remember how exactly we communicated in those days, for sure we didn’t have a test platform as we later developed for VRC Pro. I do remember that Todd was able to send new tires (virtual of course) over after having made several parameter changes based on discussion he had with Rene Cornella, Julius Kolf and Michael Salven, this was mainly email back then, and a bit of MSN. More or less ‘real time’!
Although I was involved in most conversations, I must admit that much of it was beyond me. I hadn’t raced for some 15-18 years so my input was less important. This was all for VRC v3 remember which was released in 2005. All this testing continued for VRC Pro (v4) until that completely new gaming platform was released end of 2011. I can’t go into too much detail on the physics itself and even all the different parameters used to finetune the suspension, shocks, arb’s, the tires, the engine and what have we.
Todd emphasized once more to me what a ‘rare species’ physics engine developers are. There are very few developers in the field of racings sims who have the knowledge and experience to build complete physics engines like Todd has done over 14 years working for VRC. And all have their own ‘secret’ approaches on doing things, and even though ‘physics are physics ’, there are many ways that lead to Rome. With the knowledge and expertise Todd has built over these years Todd is now working on racing car physics engines for the biggest titles in the racing sim industry. Something we can be proud of!
However much I regret that Todd left our team end of 2014 I can totally understand that it was time for him to move on to bigger projects. And because of the immense complexity we decided it was best to ‘close the book’ on the VRC physics to assure that we would not be able to mess with it and most likely screw things up. People who think that someone else could just continue where Todd left simply don’t understand physics engines. Even the best physics developer would need years to understand Todd’s engine.
But were not there yet, we have only just begun, and this is only about v3. I will address another very special feature of v3, the famous ‘AI cars’, in a later episode. There is so much more to reveal about Virtual RC v3, let’s switch to graphics and Tony West for now, that story is next.
Comments
Only active members can post comments
(Total posts: 1)
|
Jack B.
|
The death penalty is an extremely controversial topic you can try this out . From the legal and moral standpoints, it is one of the most barbaric forms of punishment that can be applied. It has been used in the US for centuries, but it has never been abolished.
Edited by author: 1.2.2023 8:19:41 GMT
|
