RACER (free physics sim)

discussed in 555 posts, 2000-2009 · overall tone positive

Posts per month

Timeline

2000-08-29 Ruud van Gaal released the first alpha of his free Racer car simulation, developed with help from physics discussions on the group. thread thread
2001-03-29 Ruud van Gaal released Racer v0.4, his free car simulator, with a Pacejka tyre model, force feedback and source code. post
2003-07-30 Version 0.5 of Racer, the free car simulation project, was released for Windows, Linux and Mac OS X. post post

Key threads

Tuning a homemade sim's tyre model for controllable oversteer 2001-08-31 · 67 posts
Jonas Echterhoff, developing a car sim with Pacejka tyre formulae, asked why his 1966 Dodge Charger spun irrecoverably once it oversteered. Ruud van Gaal, Todd Wasson, Matthew Jessick and others suggested flattening the post-peak Pacejka curve, checking integrator step rates and numerical stability, adding load sensitivity, and using analogue steering input. There was a side discussion of friction coefficients above 1, and Echterhoff reported improvement after reducing front grip and relying more on throttle control.
Modelling drivetrain, rolling resistance and tyre forces 2001-07-17 · 58 posts
Mike Stanley, writing a vehicle simulator, asked how engine torque reaches the wheels. Todd Wasson, Gregor Veble, Ruud van Gaal and Doug Milliken discussed drivetrain inertia, rolling resistance, tyre load and slip ratio definitions. The thread moved on to how locked wheels should behave within the friction circle, with Matt Jessick arguing a locked tyre's force should oppose the velocity.
Hobbyist car physics: tyre slip ratios and low-speed instability 2000-09-05 · 38 posts
Continuing Ruud van Gaal's physics thread, Matt Jessick of Motorsims, Todd Wasson, Gregor Veble and Doug Arnao explained slip ratio definitions, pneumatic trail and tyre load sensitivity. Monster Games' Dave Pollatsek described fixing Viper Racing's hill-sliding bug in NASCAR Heat by modelling sidewall shear springs. Later posts dealt with numerical instability and wheel-lock reversal at low speeds, and F1 2000's high friction values.
Modelling a differential in a home-made car sim 2002-08-11 · 32 posts
Alex Smith asked how to model an open differential and drivetrain in his own sim. Ruud van Gaal, Stefano Casillo of netKar and Jonny Hodgson explained torque splitting, clutch coupling, inertia and a gear-network approach, debating how an open diff behaves when one wheel loses grip. Smith got a partly working model with remaining issues.
Pacejka tyre models and combined slip in sim physics 2001-04-06 · 27 posts
Ruud van Gaal asked how to model sliding and combined longitudinal and lateral slip with the Pacejka formula in Racer. Gregor Veble, Neil Everett, Matthew Jessick and Dave Pollatsek offered approaches, and Doug Milliken discussed non-dimensional tyre data from Race Car Vehicle Dynamics, noting tyre coefficients are often trade secrets. Chris West pointed to a tyre-fitting tool.
Sign conventions for camber in the Pacejka tyre model 2003-02-14 · 22 posts
Ruud van Gaal asked how to handle the sign of camber in the Magic Formula tyre model on left and right wheels. Todd Wasson, Stephen Clibbery and Doug Milliken advised using inclination angle relative to the car's lateral axis and flipping conicity and ply-steer terms on one side. Dave Pollatsek, a developer of Viper Racing, joked about the mess of unit and axis conventions in his team's codebase.
How to detect wheel-to-track collisions in a car sim 2003-02-04 · 21 posts
Nick Bull asked how hobby sim developers handle wheel contact with track polygons. Ruud van Gaal (Racer), Ashley McConnell, Stefano Casillo and Sebastien Tixier described using quadtrees or OBB trees, rays cast from the wheel along the car's axis, caching the last hit triangle, and high physics rates to avoid falling through the track. Tixier noted a similar cached-grid approach was used on V-Rally 2.
Modelling anti-dive and anti-squat in Racer 2002-05-24 · 21 posts
Ruud van Gaal described adding simple anti-pitch and anti-roll parameters to his free sim Racer as an alternative to full suspension linkage modelling. Doug Milliken, co-author of Race Car Vehicle Dynamics, and J. Todd Wasson explained jacking forces, inboard versus outboard brakes and compliance effects, and corrected some of Ruud's ideas about a 'yaw centre'.
Car physics Q&A on dampers, integrators and turbos 2001-11-28 · 21 posts
Gregor Veble invited physics questions, leading to discussion of separate bump and rebound damping, nonlinear damper curves and upgrading from Euler to midpoint, RK4 or Adams–Moulton integrators. Ruud van Gaal said Racer now used Gregor's implicit integration and asked how to model turbochargers. Jonny Hodgson, Todd Wasson, Grant Reeve and Papyrus's Randy Cassidy explained boost lag, intercooling and blow-off behaviour.
How camber thrust works in tyre models 2002-09-11 · 20 posts
Jim Seamus asked how camber thrust works while debugging handling in Racer. Haqsau argued camber thrust is an artefact of tyre-test conditions rather than a separate force and that traditional tyre models are inadequate, while Doug Milliken explained the conventional view and Ruud van Gaal and others discussed motorcycle cornering and anti-roll bar effects. Participants weighed the conflicting explanations for implementation purposes.
Drivetrain effective inertia in car physics models 2001-02-02 · 66 posts
Ruud van Gaal asked how to account for engine and drivetrain rotational inertias using Gillespie's equations. Matthew Jessick, J. Todd Wasson, Bob Norton and Gregor Veble discussed effective inertia, drivetrain efficiency and gear ratios, with Wasson correcting his own formula. Months later the thread continued with van Gaal's article on effective inertia, track splines, combined slip, third-spring suspensions and the DOF file format of his Racer project.
Racer v0.5.0 beta 5 released and its physics debated 2002-08-10 · 66 posts
After Ruud van Gaal released Racer v0.5.0 beta 5, Ashley McConnell praised it while Kevin Webb called the default car's physics poor and struggled with his TSW wheel. Regulars defended the free one-man project's Pacejka tyre model and suggested better cars and a camber bug workaround. Van Gaal explained the network connection attempts came from its client/server design.
Should the sim view turn toward the car's direction? 2002-08-22 · 62 posts
Ian George suggested the driver's view should follow the car's direction of travel rather than stay fixed forward. Posters noted CART Precision Racing tried it, described an F1 2002 trick tying the view to wheel rotation, and discussed TrackIR head tracking, while Tom Pabst argued real drivers rarely turn their heads. Ruud van Gaal added speed-sensitive look-ahead to the next Racer beta.
Racer development, multiplayer, AI and the FILSCA white paper 2001-08-20 · 58 posts
Thom J asked about Racer's future; Ruud van Gaal said multiplayer would come before AI, and posters discussed handling collisions and latency online and AI that learns to drive. Christoph Schirmer posted a FILSCA white paper on the future of sim racing proposing a unifying body for leagues, which drew mixed reactions about funding and feasibility. Doug Milliken pointed to an ADAMS AI driver that takes hours of CPU time per lap.
Is Racing Legends still worth waiting for? 2003-04-28 · 43 posts
Trick Dacy said he had lost faith in the West brothers' Racing Legends while independent sims such as Live for Speed shipped working demos. Ashley McConnell defended the Wests, explaining they had to start from scratch after Empire kept the World Sports Cars code, while others mocked years of promises. Ruud van Gaal also discussed Racer support and TrackIR head tracking.
Suspension natural frequency formula and sim coding 2001-11-01 · 37 posts
Racer author Ruud van Gaal asked about the 3.13 factor in a suspension natural-frequency formula, and Petri Blomqvist explained that it comes from converting weight to mass using gravity in inch units. The thread drifted into jokes about units and Pi, then into practical sim-development topics such as integration errors, damping, tyre lateral force, profiling with variable telemetry and plotting data with Gnuplot. Doug Milliken and Gregor Veble also chipped in.
First impressions and bugs in Racer 0.47 2001-07-31 · 36 posts
Users tried Racer 0.47; some found it hard to drive, while Pat Dotson rated its feel alongside GPL and N4, particularly in catching oversteer. Sebastien Tixier, a PS2 game developer, reported skid sound and suspension bounce issues. Ruud van Gaal discussed fixes, differentials, controller configuration and replay storage, and Gregor Veble suggested a Pacejka grip multiplier.
Low-speed instability with Pacejka on free-rolling wheels 2001-12-20 · 35 posts
J. Perkins asked why his non-driven wheels oscillated when using the Pacejka tyre formula. Ruud van Gaal, Gregor Veble and Doug Milliken explained it as a numerical instability at low speed and recommended higher physics rates, relaxation length, sub-stepping the wheel and low-speed special cases. A side discussion corrected a wheel moment-of-inertia unit conversion, and Ruud shared Racer's implicit integration formula for suspension.
Ruud van Gaal releases Racer v0.39 sim 2001-02-27 · 32 posts
Ruud van Gaal announced Racer v0.39, his free car simulation, with working timelines and physics updates. Testers reported cars bouncing off the track, and discussion turned to integration rates, tyre models, triangle collision optimisation and energy conservation, with Chris West suggesting caching the last polygon hit.
Modelling suspension geometry and roll centres 2001-08-08 · 27 posts
Colin Reed proposed modelling the wheel hub path from suspension geometry as a precomputed 2D curve. Todd Wasson and Gregor Veble agreed it is adequate for games, while others planned full constraint systems. The discussion covered simulation frequency, unsprung mass, roll centre heights and real F1 roll centre data shared by Todd Wasson.
Racer v0.4.8 free car sim released 2001-10-04 · 26 posts
Ruud van Gaal announced version 0.4.8 of his free Racer sim with improved physics, a new car with a cockpit view and basic multiplayer. Gregor Veble and Sebastien Tixier discussed suspension load modelling and ray-to-Bezier-patch intersection. Van Gaal quickly posted a patched executable fixing missing sound on some cards, and helped with controller axis settings.
Engine RPM and acceleration bugs in a first car sim 2001-07-28 · 25 posts
Andreas Ekstrand asked why his simple car sim had odd engine speed behaviour from a standstill. Ruud van Gaal, Matt Jessick and others suggested checking torque calculations and units, and noted that without a clutch model the engine effectively stalls at low rpm. Months later the thread revived with Uwe Schuerkamp urging developers to pool efforts on Ruud's Racer, which Ruud said remained a personal project open to user-made cars and tracks.
Modelling steering caster and aligning torque 2001-07-23 · 25 posts
Ruud van Gaal asked how to model caster angle mathematically for his Racer sim. Gregor Veble explained that caster adds a lever arm affecting self-aligning torque and steering feedback, without changing tyre forces. Steve Smith claimed GPL models caster return, while Doug Arnao said GPL does not and that N4 was Dave Kaemmer's first model with caster.
Weight transfer, spring rates and kart handling physics 2002-04-30 · 23 posts
Pat Dotson asked why softer springs increase grip and how cross weight affects oversteer, with reference to his racing kart. Haqsau, Todd Wasson, Racer author Ruud van Gaal and Doug Milliken discussed tyre load sensitivity, roll stiffness distribution, CG height and kart behaviour. They concluded that drivers lean out in karts mainly to unload the inside rear tyre on a solid axle.
Ruud van Gaal releases an alpha of his Racer car sim 2000-08-29 · 23 posts
Ruud van Gaal released a very early Win32 build of his hobby car physics program, later known as Racer, needing OpenGL in a window. Testers reported crashes that turned out to require the separate source archive, compatibility with various cards, missing sound and an 11th-gear bug, and encouraged him to continue.
Rolling resistance and torque split in a home-made sim 2001-07-22 · 22 posts
Nick G asked for rolling resistance formulas and how engine torque is distributed to driven wheels. Todd Wasson, Ruud van Gaal and others said rolling resistance is roughly coefficient times load and largely speed independent, and that an open differential sends equal torque to each wheel. Matt Jessick and Doug Milliken pointed to SAE papers, and posters corrected a claim that friction coefficients cannot exceed 1.
Modelling anti-dive and brake bias in Racer 2001-03-07 · 21 posts
Ruud van Gaal asked about anti-dive and anti-squat modelling because his Racer Ferrari spun under braking. Matthew Jessick, Steve Smith and Doug Milliken explained that anti-geometries reduce pitch but not weight transfer, and suggested the real issue was brake bias, recommending far more front bias. Tester feedback on Racer bugs and tyre grip values was also discussed.
Wheel rates versus spring rates in Racer suspension 2002-09-13 · 21 posts
Ruud van Gaal asked whether wheel rates are always lower than spring rates, as reports said his Racer sim handled soft springs poorly. Respondents confirmed wheel rates are normally lower, with pushrod suspensions a possible exception, and discussed bump stops and numerical stability of integrators. Doug Milliken noted that many road cars engage progressive bump stops under modest load.
Racer v0.4.7 free car sim released 2001-07-30 · 21 posts
Ruud van Gaal announced version 0.4.7 of his free car simulator Racer, uploaded to High Gear as a Windows-only build. Gregor Veble explained the new combined-slip tyre method, which normalises slip ratio and slip angle into one slip vector, crediting Brian Beckman's earlier work. The rest of the thread was friendly banter about girlfriends and feature wish lists.
Modelling AI steering with spring-mass-damper in Racer 2002-09-19 · 21 posts
Ruud van Gaal asked how to make Racer's AI steering realistic using a spring-mass-damper model. Suggestions included limiting steering rate to human-like values and modelling steering torque, and Matthew Jessick wrote at length recommending crosstrack-error path following over heading control. Others discussed typical steering ratios.
Getting started writing your own car physics sim 2001-11-11 · 21 posts
A newcomer asked for advice on writing an action-oriented car simulator using an existing 3D engine. Replies recommended Milliken's Race Car Vehicle Dynamics (Doug Milliken gave the SAE ordering link), Ruud van Gaal's Racer resources and a good rigid-body system. The discussion moved to explicit versus implicit integration, stability versus accuracy, and stiff springs and low-speed tyre behaviour, with contributions from Mike Stanley, Gregor Veble and Todd Wasson.
Feral's F1 Championship Season for Mac 2002-11-13 · 20 posts
Larry announced the Mac version of EA's F1 Championship Season 2000, published by Feral, with a 45 MB demo. Mac users discussed the system requirements, frame rates, USB wheels via USB Overdrive, and Racer as another Mac driving sim.
Baraff's integration methods for stiff car physics 2001-10-15 · 19 posts
Ruud van Gaal discussed a time-reversed integration technique from David Baraff's papers that may stabilise stiff springs in his Racer sim. Gregor Veble compared it to predictor-corrector methods, Sebastien Tixier recommended fourth-order Runge-Kutta, and Matthew Jessick of Motorsims warned that implicit methods gain stability at the cost of the dynamics being modelled. Side discussion covered code structure and graphics performance.
Racer's Brno track, derailed by a DGF spat 2001-08-19 · 42 posts
Pat Dotson praised Racer with the Porsche GT3 on the Brno track. David G Fisher replied 'You're an ass', which he later said was copied from Dotson's own post in another thread. The resulting row of insults drowned out the Racer discussion.

Busiest threads

First impressions and bugs in Racer 0.47 (19 posts about it)
Drivetrain effective inertia in car physics models (17 posts about it)
Racer development, multiplayer, AI and the FILSCA white paper (15 posts about it)
Ruud van Gaal releases Racer v0.39 sim (14 posts about it)
Racer v0.4.5beta released (12 posts about it)
Modelling anti-dive and brake bias in Racer (12 posts about it)
Frame rate, physics rates and how GPL runs at 288Hz (11 posts about it)
Hobbyist physics coders discuss tyre relaxation length models (11 posts about it)
Pacejka tyre models and combined slip in sim physics (11 posts about it)
Hobbyist car physics: tyre slip ratios and low-speed instability (10 posts about it)
Racer v0.5.0 beta 5 released and its physics debated (9 posts about it)
Racer v0.4.6 released (9 posts about it)
Is Racing Legends still worth waiting for? (8 posts about it)
Racer v0.4.6.1 released (patch) (8 posts about it)
Racer v0.4 released (7 posts about it)

Sentiment by year

Each post that discussed the sim was rated from −2 (very negative) to +2 (very positive).

Year	Posts	Positive	Negative	Tone
2000	18	3	1	mixed (+0.11)
2001	333	67	4	mixed (+0.20)
2002	123	35	3	positive (+0.29)
2003	49	16	2	positive (+0.29)
2004	20	7	0	positive (+0.45)
2005	4	1	0	positive (+0.50)
2006	5	2	0	positive (+0.40)
2009	3	0	0	mixed (+0.00)