Car testing

This article presents in-game methods for obtaining accurate performance stats of Stunts cars. While several relevant stats can be predicted from the car parameters or obtained by processing repldump output, carrying out tests within the game is often faster than such analytical methods — not to mention more fun!

Top speed

Two different, equally important notions of top speed apply to Stunts cars. The flat track top speed is the speed the car can achieve by accelerating on a long enough straight, with no jumps or obstacles. It corresponds to the top speed of real cars, and is limited not only by the torque curve and gear ratios, but also by aerodynamic drag. However, jump boosting and powergear, in game versions having such effects, allow Stunts cars to overcome the limit imposed by drag and reach their real top speed.

Flat track top speed

The easiest way to get a car to its flat track top speed usually is accelerating for a few minutes on a long, fully banked track, taking care to skid as little as possible on the banked corners. That is best done by driving in a counterclockwise direction, to take advantage of left corner bias in game versions that have it. Once it becomes clear there is no more speed to be gained, the test can be ended by releasing the accelerator (to avoid gaining speed by jump boosting) and crashing the car against the outside wall of the banking; the impact speed from the evaluation screen can thus be used as the test result. A few cars have an overdrive final gear which doesn't actually bring extra speed on a flat track; if that looks plausible for the car being tested, it is a good idea to check the flat track top speed at the gear below the top one.

Flex-PG and anti-PG

The banked track method doesn't work well for flexible powergear and anti-powergear cars, as, once they get close enough to their flat track top speed, the slightest vertical movement on the banked road will trigger flex-PG or anti-PG, compromising the test result. An alternative method for such cars is using a very large square of fast grass as a completely flat "track". 24 x 24 tiles should suffice for most cars, assuming there is enough fast grass area inside of the corners of the square. To minimise skidding, it is recommended to drive smoothly on a circular trajectory, using as much of the width of the "track" as possible.

Real top speed

The real top speed is quite a bit easier to find in a test: just accelerate through a long stretch of track with many jumps, without releasing the accelerator while on air, and crash or sink the car at the end. To make landing impacts more predictable, it is a good idea to use longer jump distances in the latter parts of the test track. The only needed precautions are releasing the accelerator while climbing ramps (but not in mid-air!) with anti-PG cars, and keeping an eye for landing slowdowns on cars susceptible to them (such as the Nissan Skyline).

Acceleration

Tests such as 0-60 mph and 0-100 mph can be done efficiently with any car on a long flooded straight which begins at the start/finish tile. For instance, to find the time until 100 mph, accelerate from the start until you sense (for instance, by looking at the speedometer) that 100 mph has been reached. At that point, pause the replay, note the time, then continue without accelerating and throw the car into the water. Then, look at the reported top speed on the evaluation screen, and:

• If it is below 100 mph, rewind to the frame at which you had paused, continue the replay while accelerating, and try again a little later;

• If it is above 100 mph, rewind the tape to a little before the frame at which you had paused it, and try again;

• If it is exactly 100 mph, rewind and retry frame by frame until it goes below 100 mph. The next frame after that, which is the first one at 100 mph, corresponds to the time to 100 mph.

Acceleration tests are usually done with auto gears, to avoid the shifting technique of the tester to influence the result.

Dragstrip testing

As far as driving procedures go, dragstrip tests (that is, finding the time to cover a specific distance out of a standstill) are simpler to carry out than top speed or acceleration ones: just accelerate from the start (usually on auto gears, as in an acceleration test) on a long enough straight and check how long it takes to reach the desired distance. The trickiest part is setting a good enough visual marker for measuring the distance from the start. Given that a tile amounts to 204.8 feet (see the relevant section of the Car Model Physics article), tiles are a rather coarse unit of measurement for reproducing standard distances from real world tests, such as a quarter mile or a half mile. For measuring such distances accurately, we need to, at least in principle, account for:

• The distance covered by the car within the start/finish tile at the beginning of the test, which is exactly 722 graphical points (measured by the center of the car);

• The distance driven on the final tile, which determines the exact position of the visual endpoint marker to be used; and

• How the length of the car (often around 100 graphical points) affects the measurement, which depends on the part of the car (nose, center, and so forth) used to check that the endpoint has been reached.

When it comes to picking visual distance markers, using the "yellow" centre line ticks is arguably the approach that is easiest to generalise, as it allows a precision of about half a car length while not complicating the test by requiring obstacles to be added. To highlight what the end tile is, we can use a dirt road piece, whose presence won't interfere with the test. A slight disadvantage of using centre line ticks is that their alignment depends on the direction of the road; that, however, is something we can control while setting up the test.

The table below shows the best approximations for some common distances using the centre line ticks. A straight road going towards north (or east) is assumed. Note that tile counting does not include the start tile:

Distance	End tile (after the s/f)	Marker (northwards or eastwards)
Quarter mile	6th	Centre of the car at the end of the 5th tick
Half mile	13th	Centre of the car at the beginning of the 2th tick
Kilometer	16th	Tail of the car at the end of the 2nd tick
Mile	26th	Centre of the car at the end of the 1st tick

(For a southward or westward straight, change "at the end of" to "at the beginning of" and vice-versa in the "Marker" column.)

In case you want to set up a test for your favourite distance not included above, here are the tick coordinates read off Stressed, again assuming a northward straight (the entry and exit edges of the tiles thus being at -512 and +512 respectively):

Tick	Beginning	End
1st	-512	-427
2nd	-334	-249
3rd	-168	-83
4th	0	+85
5th	+168	+263
6th	+334	+419

Specifically for the half mile test, the nose of the car reaching the first block of a slalom on the 13th tile offers a very good approximation as well. The time at the frame immediately before the crash usually matches the result of the "2nd tick" half mile test described above.