Once the car's basic parameters are laid down, the next step is to balance it so that it's neither understeering nor oversteering too much. We want the car to be stable in corner entry, reasonably neutral in midcorner, and to deliver the power progressively as we leave the corner.
The primary parameters for setting the car's overall balance are the anti-roll bar settings. Camber, toe, and spring rates also can impact the car's balance, and the damper settings (which we'll get to in a moment) can also impact the car's balance in transient conditions and over bumpy surfaces. The brake bias affects the balance in the entry phase of the corner if the driver is trail braking.
At this point, we'll focus on the anti-roll bars and brake bias. The rule of thumb for the bars is stiffer at the front and/or softer at the rear increases understeer, while softer at the front and/or stiffer at the rear increases oversteer.
When adjusting the bars, keep in mind that changing the overall roll resistance (for example, softening the front bar while leaving the rear the same) will impact the tire temperatures and thus may require a camber change. Also, going softer overall may allow the suspension to bottom where it didn't with higher overall roll resistance, while going stiffer overall may make the car more nervous over bumps.
For this reason, GRE incorporates a feature which assists the user by optionally keeping the overall roll resistance the same whenever the bars are adjusted.
For the brakes, more bias to the front increases understeer, to the rear increases oversteer and makes the car less stable in corner entry. The trick is to find the most rearward brake bias the driver can handle (given his/her style of corner entry) without destabilizing the car, since this will shorten braking distances.
Anti-roll bar settings are impacted by relative tire width. I believe GPL's cars all have the same width tires at the front, and the same width at the rear. However, the GP Lotus has wider rear rims, which makes its rear tires more efficient. The GP Lotus seems to need higher rear roll stiffness relative to the front than the other cars to avoid having too much understeer.
Overall chassis width vs. center of gravity height has an impact on the amount that the chassis rolls. Wide chassis like the Ferrari and Lotus need a little less overall roll resistance, while the narrow Brabham and the BRM, with its second crankshaft way up high, need stiffer bars.
Brake bias settings are a function of the weight distribution of the chassis. A chassis with more weight on the rear wheels at rest will have more weight on them under braking, so it can tolerate more rearward brake bias.
Brake bias is also a function of the length of the chassis. In a shorter chassis, the more weight is transferred to the front wheels under braking, so the brake bias has to be set more to the front to avoid locking the rear wheels.
Both anti-roll bar settings and brake bias are impacted by the ride height. As you raise the ride height, more weight is transferred, so bias needs to move toward the front and the bars need to get stiffer overall.
For most of the cars other than the GP Lotus and BRM, I tend to start the bars at about 150 to 180 at the front and around 110 to 130 at the rear. The GP Lotus starts out with nearly equal values front and rear, in the 140 to 160 range. (Note that the Lotus Trainers seem to lack the GP Lotus' wide rear rims, so they use values similar to the other Trainers.) The BRM gets higher overall values because of its high CG.
For brake bias, I'll start out around 58 and move it back until the car gets unstable under braking. Circuits like Rouen and Mexico, which have corners where heavy braking while turning is required, will wind up with a click or two more front bias to help me keep from losing it in these corners.
I adjust the bars, keeping the overall roll stiffness the same, until the car is comfortable for me in the corners. If the outside edges of the left side tires are more than a degree or two higher than the inside edges, I'll go a little stiffer overall.
Because of the right front tire's tendency to overheat, I tend to run the rear bar fairly stiff. Before GRE became available, with its extended ranges, I always ran the rear bar at 200 at all but the flattest of the ovals. Now I can experiment with even higher values. I run the front bar much softer than the rear, especially on the more steeply banked tracks.
Overall roll resistance starts at road course values for the shallow ovals and goes up for the steeper ones.
Brake bias is set the same way as on road courses. On the fast ovals, it's easy to neglect brake bias, figuring it isn't important since you don't brake in the corners. However, you do need the brakes when things start to go wrong ahead of you, so it's worth it to get a decent balance on the brake bias.
The dampers can be used for tuning the car's transient responses.
Under lift throttle, braking and turn-in conditions, going stiffer in bump at the front will tend to promote more understeer, as will going softer in droop, or rebound, at the rear.
Going softer in bump at the front and/or stiffer in rebound at the rear will tend to make turn-in crisper; it will make the car less stable under lift throttle, braking and turn-in.
The dampers at the outside front and inside rear corners will have more effect on deceleration, since that is where the greatest suspension movement will be.
Under acceleration, going stiffer in bump at the rear and/or softer in rebound at the front will make the car make the car tend to be more reactive to throttle; it will tend a bit more towards power oversteer, especially if the driver gets on the throttle abruptly. Going softer in bump at the rear and/or stiffer in rebound at the front will soften the car's responses to throttle application.
The dampers at the inside front and outside rear corners will have more effect during acceleration.
Within reason you can trade off anti-roll bar stiffness for damper stiffness. For example, you can stiffen the rear dampers in bump and rebound, and soften the rear anti-roll bar. This will make the car more responsive to brake and throttle applications. It will also react more to bumps at the rear.
An advantage of this is that if the car gets into a big slide, it will be more likely to recover itself. On the downside, the car will tend to understeer in long corners. The net effect will be that the car will be more forgiving but may be a bit slower, especially on circuits with long corners.
See the Damper section of the Basic Settings chapter for more details on the impact of changes in damper settings on the car's overall behavior. See the Damper Tables for a handy reference to which corners have most effect on transient responses.
Barring a really bad setup, most setup parameters don't have a huge impact on overall lap times. As Dave Kaemmer says, most setups in GPL mostly impact driver comfort.
However, gearing in GPL is critical, particularly at tracks with a long, uphill straight. Spa, Kyalami, and Elkhart Lake are examples of circuits where gearing can have a significant impact on lap times.
For
this reason, GRE provides a graphical aid (example shown at right)
for developing gear ratios. This graph has dots showing the speed
the car will be going at redline in each gear, and has bands covering
90% power and 90% torque. Once the top gear is known, this graph
helps the engineer to quickly select the proper gear spacing.
When developing gear ratios for a given circuit, the top gear dictates the rest of the gearbox setttings, so you need to get fairly close to the right top gear first.
In Pro Damage mode, most of the engines will not tolerate running at or above redline for sustained periods. Even in Intermediate and Novice mode, abusing the engine will lead to random blowups.
Therefore, it's important to use a top gear tall enough that the car does not quite reach redline on the longest straight. This is particularly important on fast ovals and at road courses with long straights, such as Spa, Monza, and Elkhart Lake. At such tracks, the engine may go several hundred RPM higher in the draft than when the car is running alone.
I have a suspicion that the Trainer engines may run out of steam even before their redlines. They have relatively fat power curves, and it's possible that their breathing is so restricted that their power drops off before reaching redline. If that's true, it may be beneficial to gear them so they never reach redline, and also to short shift when going up through the gears. With the Advanced Trainer, even when I think I've got a good top gear, I find that often I can go taller and the car will go faster.
On the track, ideally the engine will have just enough power to induce wheelspin in the lower gears at full throttle. I think many people run the lower gears too short. They like the responsiveness of the car in the lower gears, and it's fun controlling wheelspin with the throttle.
However, power above that which can spin the wheels is wasted. It's much better to gear the car so that this power is used at higher speeds. Therefore, I try to pick the tallest bottom gears the car will pull in the lower gears without bogging. I figure if the rear tires are just on the edge of starting to spin under full power coming out of slow corners, I've got the bottom gears just about right.
However, the available ratios for the bottom two gears sometimes will not permit us to choose as tall a gear as we'd like. The Cosworth, Repco, and both Trainer engines would benefit from taller first and second gears on long circuits, but unfortunately we have to work with what we've got.
Also, some engines have very little torque in the low end of the rev range. The Murasama and especially the BRM have this problem. Be careful not to gear these so tall that they bog down coming off the slow corners.
Note that the Repco engine has such a broad torque band that it's beneficial to use very tall bottom gears. However, with a very tall first gear, sometimes the Brabham won't start rolling even at full throttle from a dead stop, if the engine is idling when you open the throttle. In this case, a quick tap of the clutch will get the revs up and the car will light up the tires and take off.
The obvious approach to gear ratio spacing is to space the ratios evenly so that the redline dots make a nice straight line from first to 5th. This approach, however, would be wrong.
The car accelerates most quickly in the lower gears, so it spends relatively little time in these gears. The more powerful cars have an excess of power in the lower gears as well. Any power above that which induces wheelspin.
On the other hand, the car spends a lot of time in 4th and 5th gears on long straights. It's accelerating slowly because most of its power is consumed by overcoming wind resistence.
It's important to gear the car so that most of the power and torque are used where they are needed, which is at higher speeds. Therefore, the most efficient gearing will show a parabolic curve of the redline dots, with the high gear dots spaced more closely together and the low gear dots spaced farther apart. The broader an engine's torque band, the closer together its top gear dots should be.
Spacing the lower gears out and making the upper gears close together will also mean less shifting in the slower bits, which will also save a little time.
Again there's an exception. The BRM has such a narrow power band that it's not always practical to space the gears as you would for the other engines. If the low gears are spaced too widely, the engine bogs down after upshifts. On circuits with long straights, optimal acceleration seems to come from spacing the BRM's gears so that the redline dots are just about in a straight line.
Sometimes it's beneficial to set a certain gear a little taller or shorter than the setting dictated by a smooth parabolic curve, in order to have just the right gear for a critical section of the circuit. For example, Kyalami has several short straights between second gear corners. Some engines, with wide power bands such as the Ferrari, can be geared so that no upshift to third is necessary.
Once you've optimized the lower gears and the gear spacing, you may find that the car is accelerating more quickly on the long straights, so you may need to raise the top gear because now the car is reaching redline in top gear. This, of course, may require you to readjust all the lower gears to keep the optimal spacing.
Keep in mind that many of the setup parameters have an impact on the car's behavior that may require adjustments to other parameters.
Once you've finalized your spring rates and relative front/rear roll stiffness, you may want to take a look at tire temperatures again, with an eye to making a small camber change or change in the overall roll stiffness to get the inside and outside edge tire temps on the left side tire to be even. You may also want to see if the car will tolerate a bit more rear brake bias, especially if you've done things to increase the car's stability under braking and turn-in.
Once you're happy with the car, you might try lowering the ride height a quarter of an inch. If it still doesn't bottom the suspension anywhere, leave it there. But check the tire temps again; now you may be able to reduce the overall roll stiffness and go a click farther to the rear on brake bias.
The whole process will be an iterative one. As you hone in on a good setup, you'll probably find yourself making smaller and smaller changes until finally you've got a setup that's both comfortable and quick.
Your stopwatch and the seat of your pants will tell the tale ... and that delicious feeling in your gut when you realize you've put together a real sweetheart of a setup.