Comparison of racing simulators

The following tables compare general and technical information for a number of racing simulators. Please see the individual products' articles for further information. This article is not all-inclusive or necessarily up-to-date.

In each table, the simulators list follow the same order; the ordering is by sim engine and evolution:

• NASCAR 2003 (all developed with David Kaemmer)
• Racer
• F1 Challenge 99-02, GTR, GT Legends, GTR 2, rFactor, RACE (using evolutions of the ISI engine)
• netKar Namie, netKar Pro (Using Stefano Casillo engine)
• Live for Speed (by Scawen Roberts, Eric Bailey and Victor van Vlaardingen - Not to be confused with the Need For Speed series)
• Richard Burns Rally, Driver's Republic (1st beta) (different engines, but both physics engines are developed by lead physics designer Eero Piitulainen)
• Virtual Grand Prix 2
• X Motor Racing
• TORCS The Open Racing Car Simulator
• Grand Prix Legends (all developed with David Kaemmer)
• Grand Prix 4 (developed by Geoff Crammond and the MicroProse label)

Physics

Tyre simulation - part 1

Note 2: "plus some integration for dynamic behaviour for the quick spinning on the Y axis of the tire (quick steer)".

Note 3: similar to models like "string model", "spring model", "brush model", "slip model".

Note 4: There's one data file per manufacturer, but the files do not differ. Therefore at least one of them is not based on real data.

Note 5: 100% customizable tire model that allows define any tyre model (Pacejka, Own system or based on tyre manufactures)

Tyre simulation - part 2

Note 1: On non-tarmac stages, there's higher grip after previous cars have swept the racing line.

Note 2: On non-tarmac stages, soil accumulation on sliding tires increase their grip.

Note 3: 48 sampling points, arranged as 16 along the circumference by 3 across the width of the tire.

Note 4: 16 sampling points along the circumference.

Note 5: True 3D Tyre simulation.

Suspensions

Note 1: Instead of dynamically modeled suspension components, Racer uses parameters that define linear camber and toe change per unit of suspension deflection. This allows a rough approximation of the camber and toe effects that a dynamically modeled suspension has. Roll centers are static, and their placement is defined by a parameter.

Note 2: Suspension geometry in LFS can move only in X-Y plane, so trailing arm is approximated by moving the wheel up and down along the Y-axis.

Note 3: As of the beta version, suspension geometry in DR is completely customizable in the number, position and characteristics of any suspension rod, spring, damper and axis.

Drivetrain

Note 1: No proper RWD support; the RWD cars available for testing are made by switching the front and the back end, therefore handbrake affecting to the front wheels.

Note 2: This diff behaviour can be achieved through correct setup of the active differential mappings.

Note 3: No proper 4WD support. Cars can be set to 4WD, but power distribution is always 50%/50% (front/rear), and there are no means to adjust centre diff locking. This gives the 4WD cars a tendency to understeer.

Note 4: Limited Slip Differentials (LSD), are divided in four categories: Viscous Coupling (Syncro), Clutches, EDL and Torque Biasing Differential. The Clutches category include Passive Clutch, Progressive/Locking Clutch (Positraction or Salisbury type), Simple Hydraulic Clutch and Electronic Clutch.

Note 5: Their settings cannot be changed while driving, unlike in real life WRC.

Note 6: Power distribution is customizable. Racer is hardcoded so that only one differential per car is allowed. Therefore in a 4WD car, all four wheels are driven, but only two of the wheels can have a "real" differential. Also, Racer does not implement a center differential for 4WD.

Note 7: Transmission layout, as of 1st beta version, is completely customizable in DR. Only one type of differential is provided, probably an open one, but not verifiable.

Road surfaces and covers that affect handling

Note 1: The physics of the interaction between tire and wet asphalt is fully implemented, but in version 1.0.1 is not used yet.
Note 2: RBR initially has a total of 132 different surfaces (be it road surfaces or elements such as different types of rocks, water... even several types of bushes and poles!).
Note 3: There are 3rd party tracks featuring non-conventional surface types, such as snow, dirt (Blackwood for GPL), rain (Nurburgring in rain), and probably more. Also, there is a track surface type (surface type 8) that was supposedly going to be used as water, however it wasn't finished and doesn't work properly. It usually results in the car sinking into it, then being launched.
Note 4: In every surface type, there are parameters such as SoilThickness and SoilDensity in RBR and WetGrip/DryGrip in ISI-based sims that can be adjusted.
Note 5: Racer surface types have several parameters that can be customized on a per track basis. Specifically, these are grip, rolling resistance and viscosity.
Note 6: Ice is only used for internal testing, not present in the official tracks yet.
Note 7: rFactor engine can be adapted to model any kind of surface editing a terrain data file, but at the moment there is no specific support to model the interaction of different tire models with different surfaces.
Note 8:The event version of rFactor, that taken to demonstrations with BMW and Intel supports full wet weather, wet lines and rain drop particles. ISI have stated that this feature will reach the public version of rFactor at some point.
Note 9: A 3rd party track (Watkins Glen Wet) has modified grip values to resemble wet-weather racing.

All ISI based simulators can be modified to support any surface.

Aerodynamics

Note 1: The undertray angle is not adjustable and the undertray angle does not vary with the pitch of the car. The ground clearance of the car does not affect undertray downforce.

Note 2: The physics engine that Grand Prix Legends shipped with did not model downforce. Downforce was added to the model with the release of the 1969 mod for GPL. For its 1966 Mod, this was changed to 'upforce' since the wingless F1 cars of 1966 tended to generate lift at higher speeds.

Chassis

Note 1: see Car.ini fields reference

Note 2: Uses a 3x3 inertia matrix (thus allowing for combined pitch and roll to have different inertia to separate pitching then rolling) that is dynamically generated from a group of masses distributed around the vehicle body; thus adding passengers and fuel usage, automatically effect the inertia matrix in a realistic manner.

Damage

Note 2: Only wings and suspensions.
Note 3: Front wing.
Note 4: Simple graphical representation, only large body panels fall off.
Note 5: Includes damage vulnerability based on suspension stiffness with stiffer settings on uneven/bumpy surface causing damage more easily.
Note 6: Engine damage does not result from impacts - only from over-revving the engine.

The CTDP 2005 Mod for F1 Challenge '99-'02 did add Gearbox damage and failures into the game.

Force Feedback

Note 1: Optional implementation from 3rd party also available - Real Feel plugin.
Note 2: Disabled in netKar Pro version 1.02.

Artificial Intelligence

Note 1: Of a rather rudimentary sort.

Note 2: Not as shipped-most AI wrecks due to bad LP lines or errors in the .ini file, but some modding has been done in this area.

Note 3: AI Mistakes are only considered partial in rFactor because they are not convincing to anyone. As of V1.087 beta there is a *.PLR variable called "AI Mistakes".

Multiplayer

Connection quality

Note 1: More data packets are send to peers who are closer on track, network conditions permitting.
Note 2: "Yes" is good. Here, "significant" means any outright freeze of the program or a decrease or stutter of fps significant enough to impact the driving experience negatively (e.g. causing a crash or loss of control).
Note 3: Third-party plug-in needed.
Note 4: Must be enabled in .PLR settings.
Note 5: Only when cars of new joined players are loaded into garage. It is settable in multiplayer settings or .PLR file
Note 6: Hotseat mode.
Note 7: "Every time there are more than 27 racers on the track, the excess players are unable to race properly due to all kinds of strange errors."
Note 8: The dedicated host program takes up one of the peer slots.

Configurability options

Note 1: Possible by server uploading results to a third-party website.
Note 2: Possible by killing the program then the next driver connecting with the same name.
Note 3: Results (XML) log contacts and incidents and these stats can be viewed, for example, when uploading XML results file for viewing with rFactor Report (one example).
Note 4: Ability for multi-class results possible in the CTRA servers.
Note 5: Ability for Crash Ratio via yellow flags possible in the CTRA servers.

Sound

Note 1: "Recorded" if long microphone-recorded sound samples (>=0.3 secs) were used for the vehicle engine main sound; "Synthesized" otherwise.
Note 2: The user can choose any of the two systems.
Note 3: Sample based sounds available using 3rd party Car Sound Remixer utility/plugin.

Graphics

part 1

Note 1: Suspension travel is only visible in open wheeler cars (LX4, LX6, MRT, FOX, FO8, BF1). Real-time two-dimensional overlay schematic available for all cars.
Note 2: Available with HDR plugin originally for use with GTA San Andreas.

part 2

Note 1: Rain is available for some tracks as a third-party add-on. It does not affect physics.
Note 2: A configuration file hack allows players to change the driving view to the rollbar (top) view.
Note 3: Currently implemented in the demonstration build taken with BMW and Intel to demos based on the current 1.250 build, to be implemented in the public version at a later date.
Note 4: Available as 3rd-Party add-on for some mods. Then only attacthed to steering wheel

References