Question on consistency...

[kcbrown]

When starting out, you want to focus on consistency first. Don’t worry so much about speed because it will come later of its own accord. Hit your braking points with the same pressure every time. Hit your turn-in and your apex points. Track-out will be a derivative of the last two and can be used to judge whether you need to adjust your earlier points or not.

I don't understand how this can possibly be accomplished. The proper braking point is a function of your speed and braking strength. Vary either, and your braking point has to change, else you'll either be too slow through the corner (and you'll have to control the resulting line strictly through steering, which I suppose might be what you want regardless), or too fast (and thus it's off the track you go!).

So how do you discover the proper braking point when it is a function of so many factors? And how can you practice braking at the proper point, much less getting the rest of the line right, if you can't even know where it is?


It also raises the question of how to get the rest of the line right. I know you can just control the steering to drive it properly, but won't those steering inputs be very different than what they'd be if you're really getting most of what the car has to give? Which is to say, if you're driving the car at a reasonable fraction of its potential, won't the line be largely the result of the car drifting as opposed to being steered?

 

[SoundGuyDave]

I don't understand how this can possibly be accomplished. The proper braking point is a function of your speed and braking strength. Vary either, and your braking point has to change, else you'll either be too slow through the corner (and you'll have to control the resulting line strictly through steering, which I suppose might be what you want regardless), or too fast (and thus it's off the track you go!).

I will admit that it sounds "off" on the face of it, however, it does actually work... Yes, braking point is absolutely a function of speed into the braking zone, however if you can establish a solid, functional pace and start eliminating variables (braking pressure for one), you'll wind up with (hopefully) a series of laps all within a fairly narrow time window. The point here is to ignore absolute velocity in favor of getting the basic mechanics right. Once you get consistency in the cardinal points of the corner (braking, turn-in, apex, track-out), you can then approach increasing velocity by shifting one variable at a time. Frequently, novice drivers tend to over-drive some corners, putter about others, and then wonder why braking points are all over the map. Taking a logical, linear approach helps to focus on the important bits:

1. The Line. It all starts here. You could have impeccable technique, but if you can't find or stay on the line, you're slow. Period.

2. Braking points and pressures. If you pedal pressure is inconsistent, then your delta-V will be as well, yielding a moving target for braking points.

3. Turn-in point and wheel smoothness.

4. Apex point and power application through corner exit.

5. Track-out point (derivative of turn-in and apex) lets you judge the success of your previous point selection.

6. Go back to 1, lather-rinse-repeat.

So how do you discover the proper braking point when it is a function of so many factors? And how can you practice braking at the proper point, much less getting the rest of the line right, if you can't even know where it is?

It also raises the question of how to get the rest of the line right. I know you can just control the steering to drive it properly, but won't those steering inputs be very different than what they'd be if you're really getting most of what the car has to give? Which is to say, if you're driving the car at a reasonable fraction of its potential, won't the line be largely the result of the car drifting as opposed to being steered?

Your questions are all valid, however moot if you approach learning a corner in a systemic process.

Pick a brake point (SWAG if you wish), and brake until you have reached what you deem (SWAG again) to be a "reasonable" cornering speed, then apply maintenance throttle. At your turn-in point, smoothly roll the wheel over so that you hit your apex cleanly, and start smoothly rolling into the throttle so that you're at WOT right about the time you hit track-out. At track-out use your car placement to judge the "validity" of turn-in point, apex point, and cornering velocity.

For sake of argument, let's assume that the corner was pretty nice, good turn-in and apex points, leading to a good track-out point. However, you finished your braking and got on maintenance throttle about 100' short of your turn-in point. Next lap by on that corner, shift your braking point closer to the corner by 80-85' (longer straight means more entry velocity), and re-assess. Let's say that now the braking point feels pretty good, and you're still hitting your turn-in and apex points, with good result on exit, and most importantly, there is little or no "maintenance throttle" time between braking and turn-in. At this point, let's assume that you have consistency going for you: same brake point, same brake pressure, same turn-in, apex, and track-out points. Now, you are hitting the line, and doing so the same way every time. You have consistency, you are in control, but you're just not "fast." What you HAVE developed, though, is a solid set of mechanical techniques (smoothness onto and off the brake pedal, a single smooth but quick steering-wheel input, smooth application on-throttle) and a "perfect" line through the corner, at lower speeds where smaller errors get lost in the noise, instead of impacting the line itself.

Now, move your braking point in closer again, by, say 10' or so. Since you will no longer be braking for as long (shorter distance, same pedal pressure), naturally, your entry speed will go up. Again, hit your turn-in and apex points, and take a look at your track-out point. If the car still smoothly carves right to apex and on to a picture-perfect track-out, then next time by move your braking point in by another 10' or so. At some point, you're going to start missing the apex point, going wide. When that happens, back up your braking point by FIVE feet and try again. When you have the minimum amount of braking distance, but still hit your apex and track-out points, you have developed the maximum amount of entry speed you can carry through that corner. Now that you've increased your velocity at apex, start working on corner exit. Start adding more and more throttle starting at the apex point and see how that affects the track-out point. IF the car is still under solid control, but you now start pushing past the edge of the berm and onto it (or starting to drop wheels at exit), then back up a couple of steps, and adjust your turn-in point a couple of feet closer to the corner and your apex point a little further around the bend. Yes, you will then have to re-assess your braking point (later turn-in means smaller radius at the beginning of the corner thus less speed at entry), but the later-apex line allows you to get on-throttle earlier, as the radius widens more quickly. A "pure momentum" line (geometric center apex) has the car describing a 100% consistent arc through the corner, all the way from turn-in through track out. Assuming you hit your brake point perfectly (and thus have max entry speed), the instant you turn the wheel, you have used 100% of your grip budget for cornering, with nothing left over for accelerating until you straighten out the wheel again at track-out. You are carrying the maximum possible speed through the entire corner, but you're also unable to increase that speed for a looooonnnnnnggggg time. Mustangs (and Corvettes, and Vipers, and...) actually make power, though, so for the shortest possible time through the corner, you want to run some variation on the "late apex" line. That means delaying your turn-in, turning-in more sharply (and thus at a lower speed), moving your apex point further around the corner, and then gradually straightening the wheel once you cross apex. Assuming you "nailed" the turn-in phase (maximum possible velocity between turn-in and apex with the tighter radius) and are using 100% of your grip budget, as soon as you begin to unwind the wheel, you're now describing a wider radius, which means a higher maxV, which means you can get on-throttle! At this point, you're starting your acceleration at apex, rather than delaying it all the way to track out. As you're monitoring your progress when moving apex points around, the best gauge to use is engine RPM (and thus velocity) at exit. Keep moving the turn-in and apex points later and later, continuing to increase rate of acceleration to make sure you hit your track-out point. At some point (determined by the car's ability to accelerate) you'll either come up short of GETTING all the way to track-out, leaving space between the car and the edge of the track, or you will have traded too much entry speed for acceleration space, and will have a lower exit RPM (thus speed). Now, back up your changes just a tick. Note that here, as you increase your velocity through the corner, any sort of wheel-slip, rotation, etc. will be addressed via adjustments to the points on the line.


Remember the eventual goal: Maximum exit speed out of the corner and into the next. To get there, you want absolute minimum braking time, maximum cornering velocity at turn-in, and maximum rate of acceleration at exit. This is absolutely an iterative process! Also remember that as you increase your exit velocity, that will also increase your entry velocity into the following braking zone... Now you know why you should only "work" one corner at a time...

The thing to remember about all that, though, is that for you to build your speed, you need to be consistent in hitting your points, so that you can properly judge EXACTLY what the results were from changing ONE variable at a time. Start slow, find the points that make the corner "work," then start refining, one point at a time, beginning with the braking point. If you're inconsistent in ANY of this, you wind up with invalid data. First lap is too slow at entry, so you move your braking point (but inconsistently increase your braking pressure as well!), and you're STILL too slow! So you move your braking point again (and slack off on the pressure) and suddenly you shoot past apex due to corner-entry understeer. So you move your braking point back again, (and spike the pressure) and you're again too slow. It's impossible, right? This is why consistency is simply so critical.

Advanced drivers can draw on their previous experiences to get "ballparked" pretty quickly, but the novice really has no frame of reference to be able to judge what "proper" entry speed would be. If you can approach learning a particular corner in a scientific perspective, changing only one variable at a time, and assessing the results, you will be head-and-shoulders above the results gained from just trying to throw a car into a corner and see what happens. Yes, the "other guy" will be faster out of the gate, but in HPDE-1/Novice groups, that really isn't what it's all about. In short, walk before you run. Get "the line" down pat at lower speeds, giving you more time to think, and get your techniques solid. Then start picking up the pace. Will the points on the line change? Yes. However, turn-in, apex, and track-out are not going to change radically. We're talking about (in some cases) a foot or two at the most. Without starting from a base of consistency, how can you possibly start making 1' incremental adjustments at speeds approaching 100mph?

Just to put perspective on things: I had the individual lap timings broken out by driver for the 2013 24hrs of Daytona. Scott Pruett had a lap variance of less than 0.3 seconds from fast lap to average, discounting any lap that was 1.5 seconds or more off pace (filtering out laps with traffic). Think about that for a minute. His BEST lap was only 0.3 seconds faster than his AVERAGE traffic-free lap. Over four hours of driving! On a pretty long track! THAT is consistency, and also one of the big reasons that his team won the overall...

Does that make sense?

 

[kcbrown]

Makes a lot of sense, actually. I like that approach very much. Fits well with the way I think.

However, I'm deeply skeptical that my innate ability will be anywhere near what's necessary to come close to the kind of consistency you're talking about. My base reaction time is around 250 milliseconds (this is the time between when I see a color change and when I click on the mouse to acknowledge it, so anything that requires actual judgment will, of course, take quite a bit longer). That means that, at speed (say, 60 mph), I can expect the variation in my braking point to be around, I'd guess, 25 feet. That's "uncontrollable" variation. There's literally nothing I can do about it.

That won't affect just my braking point. It'll also affect the variation in my turn-in point.


There's also the issue with my memory, which essentially isn't there at all. I'm figuring it'll take me something like 50 laps just to learn what the track looks like enough that I'll know with confidence what the next corner's going to look like. This is based on experience with Gran Turismo which, admittedly, isn't likely to be terribly instructive in terms of how the car will behave or anything, but is likely to be instructive in terms of teaching me the best I'm going to be able to do at things such as the speed with which I really learn a track. And I'm going to be demanding that an instructor be with me all the way until I know the track, the safe braking points, etc., inside and out. I expect I'll soon be banned from NASA events just for being too "needy"!

Good thing I'm not signing up to be a professional driver!

Fortunately, I'm just in this for the fun of it. As long as I have fun with it and remain safe, I'll be happy. I'll let other people worry about getting every last tenth of a second out of their lap times.

I'm wondering if the approach should be modified slightly. Rather than work on finding the proper braking point and (to the likely rather minimal degree possible) braking consistency first, might it be better to work on the speed through the turn first, once the line's been consistently nailed? Which is to say, make the order something like this:

1. Get to the point where I can nail the line, time after time, at significantly lower speeds than the "proper" one.
2. Gradually increase my speed through the turn, but not worry about the braking point (except to ensure that my braking point is sufficiently far out that I can easily reach the turn-in speed I want), until I get reasonably close to what the combination of me and the car can do through the turn.
3. Gradually work on getting the braking closer to the right place so that when I come off the brakes, I'm at the right speed and turn-in location to make the turn at the proper speed.

Is there any disadvantage to doing that than working on the braking first? The reason I ask is that the braking point you wind up with is going to depend on the speed you're going and the target speed for the turn, but until you've figured out the latter, you can't know the former.


Oh, and another question: how will I "know" if the track-out point I hit is a good one?



Finally, is there any way I can use a "driving simulator" such as Gran Turismo to help me with this whole process, so that I'll be starting off in a better position than I might be otherwise? I've got a force feedback steering wheel and pedals, so at least it "feels" somewhat realistic. I can use all the advantages I can get.


I'm going to the NASA event at Sonoma Raceway on March 15th and 16th, so I'll soon get to experience this for myself. I'm a bit apprehensive, as it's been 20 years since I've been on the track, but because my intention is to take it slow and easy (and listen to everything the instructor has to say, though he may have to say some things multiple times so that I'll remember them), I think at least I'll be safe. The last notable thing that I can remember in the last track event I participated in was that I went right past one of the sharp left-handers at Thunder Hill and into the mud (it had rained a few days prior) because I knew that attempting to make the corner would have resulted in Bad Things happening. And that happened because I didn't remember the nature of that corner or its braking point. So you can see why I'm apprehensive about my memory, as it will have a major effect on safety.

 

[SoundGuyDave]

Makes a lot of sense, actually. I like that approach very much. Fits well with the way I think.

However, I'm deeply skeptical that my innate ability will be anywhere near what's necessary to come close to the kind of consistency you're talking about. My base reaction time is around 250 milliseconds (this is the time between when I see a color change and when I click on the mouse to acknowledge it, so anything that requires actual judgment will, of course, take quite a bit longer). That means that, at speed (say, 60 mph), I can expect the variation in my braking point to be around, I'd guess, 25 feet. That's "uncontrollable" variation. There's literally nothing I can do about it.

That won't affect just my braking point. It'll also affect the variation in my turn-in point.

Actually, I'll disagree with you here... The reaction-time measurement is based on a randomly-generated stimulus, meaning your decision loop is something like: See the light, process the information, make the decision, act. This would be akin to something sudden occurring on track, like my interaction with a coyote last year. When talking about something like a brake point, or a turn-in point, there is the benefit of foresight, and anticipation. This is why I stress permanent landmarks with my students. You're closing in on the corner at 100MPH, and your PLAN is to hit the brakes at the "3" marker in order to get to your desired entry speed. Your decision loop starts seconds before you get to the action point, since the corner didn't just suddenly appear. You have the benefit of anticipation and timing in your favor. As you come up on the "3" marker, which is your braking point, you have already cut well inside of your decision loop, eliminating the process and decision portions. You KNOW you're going to brake, and you KNOW that you will brake at a specific point, which is coming up in 3...2...1...BRAKE! Your 250mS reaction-time loop can and will be shaved to something around a 10-20mS error factor. At speed, that's well within the 10' window that you want. Think of it like tapping your foot to music. If you wait to hear the beat, then trigger your toe to tap, you'll be behind. But, because you can anticipate the rhythm, you can rapidly fall into sync within a VERY narrow margin of error. If you have a non-movable landmark (I normally use pavement patches, braking markers, joints in the berm, etc.) for each of your cardinal points, you'll simply drop into the rhythm of things. After all, the brass ring is to see, process and act without conscious thought, allowing you to focus your mental faculties on strategy. That, however, requires a goodly amount of seat time to achieve!

There's also the issue with my memory, which essentially isn't there at all. I'm figuring it'll take me something like 50 laps just to learn what the track looks like enough that I'll know with confidence what the next corner's going to look like. This is based on experience with Gran Turismo which, admittedly, isn't likely to be terribly instructive in terms of how the car will behave or anything, but is likely to be instructive in terms of teaching me the best I'm going to be able to do at things such as the speed with which I really learn a track. And I'm going to be demanding that an instructor be with me all the way until I know the track, the safe braking points, etc., inside and out. I expect I'll soon be banned from NASA events just for being too "needy"!

LOL!! I've yet to see somebody run out of NASA for being too needy, or wanting instruction! Unfortunately, I can't help you with simple memory tricks, or anything like that. Yes, you NEED to learn the track to be able to drive it consistently. Best bet is to start working now. Find a good YouTube video of the track you're going to drive, preferably from a Mustang, or Corvette, and not a Miata or a WRX. Something with reasonable horsepower, weight, and rear-wheel drive, and prefereably a camera angle from inside the car, behind the driver to give you a better example of what you will see when you're on track. Download the vid (or book-mark it!), and then find a nice, clean lap, and start dissecting it. For each corner, identify the brake point (landmark!), turn-in point (landmark!) and apex point (landmark) and WRITE THEM DOWN. Next, run through the lap again, looking at where the driver tracked-out. In most cases, the proper track-out point is going to be against the outside edge of the track, right next to or on top of a berm. Look at where the car wound up relative to the length of that berm. Then WRITE THEM DOWN. If you can, loop that lap, and just let it soak into your consciousness, to the point where you start anticipating the driver's actions. Don't be afraid to follow along with your hands, either! Mimic turning a steering wheel, moving your foot back and forth between brake and gas, reaching over for a shift (then getting your hand back up on the wheel!) You're training your memory, locking in trial cardinal points, and also working on muscle memory at the same time. Spend at least 10-15 minutes per day at this, and 20 minutes per day in the three days preceding the track event.

Good thing I'm not signing up to be a professional driver!

Fortunately, I'm just in this for the fun of it. As long as I have fun with it and remain safe, I'll be happy. I'll let other people worry about getting every last tenth of a second out of their lap times.

Fantastic attitude! Keep that firmly in mind when you hit the track, and while you may be slower than some for the first session or two, you'll wind up being faster than MOST by the end of the day or weekend!

I'm wondering if the approach should be modified slightly. Rather than work on finding the proper braking point and (to the likely rather minimal degree possible) braking consistency first, might it be better to work on the speed through the turn first, once the line's been consistently nailed? Which is to say, make the order something like this:

1. Get to the point where I can nail the line, time after time, at significantly lower speeds than the "proper" one.
2. Gradually increase my speed through the turn, but not worry about the braking point (except to ensure that my braking point is sufficiently far out that I can easily reach the turn-in speed I want), until I get reasonably close to what the combination of me and the car can do through the turn.
3. Gradually work on getting the braking closer to the right place so that when I come off the brakes, I'm at the right speed and turn-in location to make the turn at the proper speed.

Is there any disadvantage to doing that than working on the braking first? The reason I ask is that the braking point you wind up with is going to depend on the speed you're going and the target speed for the turn, but until you've figured out the latter, you can't know the former.

I would suggest working braking points first, but with good reason. If you start working on entry speed (and exit speed follows blindingly quickly, trust me!) you WILL start carrying more and more speed into the following corners. By having an established braking point you can start seeing the impact your speed has on braking distances, and you won't be surprised into a "four-off" at the end of braking. Also, start getting used to the concept of using your braking to determine your entry speed. If you brake however and wherever, you'll have a much tougher time getting consistent entry speed to start working with. Also, it'll make your instructor happy to know that you can whoa the car up properly. That will then allow him to focus on getting you hustling a little sooner without wanting to leave "a little in the bank for the wife and kids," if you follow.

Oh, and another question: how will I "know" if the track-out point I hit is a good one?

THAT is an excellent question! We've pretty much figured out that the combination of turn-in point, apex point, and rate of acceleration post-apex will determine your track-out point. Look for the following:

1) Your car winds up right at the edge of the track at corner-exit. Not 6' in, and not with two in the dirt; try for 18" or less. I generally try to get my students to just "kiss" the berm with the edges of their tires. Feel that it's there, if you follow. That way you know you used the WHOLE track!
2) You were able to smoothly unwind the wheel starting at apex, and were pointing dead-straight down the track when you got to the edge at exit. Do NOT drive the car there! Let the car flow there naturally.
3) You were able to get to WOT somewhere between apex and the track out point.

In my experience, there is generally a berm at exit of most corners, if for no other reason than to protect the track edge from being beaten to death by the tires. GENERALLY SPEAKING you'll want to wind up parallel to the edge of the track, right next to the edge of the berm well past the half-way point, perhaps 2/3 of it's length away from the corner would be a good initial target. Of course, YMMV, batteries not included, depends on the track and corne in question, etc.

If you do the above, you hit a perfect exit. If however:

1) The car comes up too far away at exit (more than 18" off the edge), then you either turned in too late, or were a complete wuss with the throttle!
2) The car comes up "short" on the berm, and you still have steering lock in when you get there, then you turned in too early or missed your apex completely. This is another indicator where consistency is so critical!
3) The car winds up on TOP of the berm with the outside wheels. This is a variation on #2, where you either turned in just a tick too soon, missed your apex a bit, or got just a little greedy with the throttle on the way out.
4) The car seems positioned properly, but you dropped the outside rear tire (only!) on the berm and chuffed up a small dust cloud on your way out of the corner. Nailed it!!

Finally, is there any way I can use a "driving simulator" such as Gran Turismo to help me with this whole process, so that I'll be starting off in a better position than I might be otherwise? I've got a force feedback steering wheel and pedals, so at least it "feels" somewhat realistic. I can use all the advantages I can get.

Simulators are great for "learning" a new track, but they're absolutely nothing like the "real thing." You can experiment with different lines through corners with them to see what seems to let the car flow (if the software physics package is goo), and if the graphics are done well enough, you can pick out some "prospective" landmarks. It will all change once you put rubber on tarmac, though! In iRacing (good sim!) my lap times at Road America are nearly 7 seconds higher than I've done in the real world. The sims get some things right, and others, not so much... Honestly, I've been pretty happy with doing the video thing, assuming you can find the right one to use as a guide.

This one's not ideal, but it might give you a start:

http://www.youtube.com/watch?v=yLezvppt_EM

Note on the first lap how he misses apex on one corner, and winds up well out on the rumbles at exit... Also note how he's experimenting with a couple of different lines through the hairpin. ALSO note that he's pretty much hitting apex everywhere, as well!

I'm going to the NASA event at Sonoma Raceway on March 15th and 16th, so I'll soon get to experience this for myself. I'm a bit apprehensive, as it's been 20 years since I've been on the track, but because my intention is to take it slow and easy (and listen to everything the instructor has to say, though he may have to say some things multiple times so that I'll remember them), I think at least I'll be safe. The last notable thing that I can remember in the last track event I participated in was that I went right past one of the sharp left-handers at Thunder Hill and into the mud (it had rained a few days prior) because I knew that attempting to make the corner would have resulted in Bad Things happening. And that happened because I didn't remember the nature of that corner or its braking point. So you can see why I'm apprehensive about my memory, as it will have a major effect on safety.

I'll be waiting (with bated breath!) to hear a report on how things went, what you learned, how much fun was had, what worked and what didn't, etc.

Biggest thing to beware of getting (back) on track for the first few times is your personal fatigue factor. Although you are driving in a seated position for relatively short stints (I mean really, 20 min? How hard can it be!), the mental drain is immense! Stay hydrated, keep up on proteins and carbs, and if you feel like you're getting tired, or confused about something you thought you pretty much had down before, PARK THE CAR. Also, listen to your instructor (duh!), he/she will have MUCH better specific information about the track than I could give you half-a-continent away. When you do your "meet and greet" on the first morning, chat about their instructional plan, and then roll with that. Trust me when I say none of us instructors WANT our students to go slow (outside of instructional scenarios, of course) none of us WANT our students to suck....

Have fun, be safe!

 

[claudermilk]

Fantastic thread! I will file this great info away for the next track day I do. I was kind of doing some of it, but this helps to organize what to do in my mind.

Fantastic attitude! Keep that firmly in mind when you hit the track, and while you may be slower than some for the first session or two, you'll wind up being faster than MOST by the end of the day or weekend!

I actually found this is what happened with my first day. I was kind of slow & giving point-bys in the morning. By the afternoon, no more of that. I ended up being one of the faster cars out that as far as I could tell.

 

[SoundGuyDave]

Fantastic thread! I will file this great info away for the next track day I do. I was kind of doing some of it, but this helps to organize what to do in my mind.

Most of this is a re-statement from that sticky I wrote on track driving techniques, so be sure to take a peek at that as well, if you already haven't!

I actually found this is what happened with my first day. I was kind of slow & giving point-bys in the morning. By the afternoon, no more of that. I ended up being one of the faster cars out that as far as I could tell.

Exactly. You need to have a foundation to build on! In the early sessions, unless you approach things systematically, the "fast guys" are almost invariably the ones that have the bigger scrotal sacks. By the end of the day, though, the people that started by building that foundation have usually eclipsed them!

Think of it this way: The "early fast" guys are running inconsistently, say at 70% of their capability, +/-35%. That means that they're somewhere between 35% efficiency (slow, butchered corner) and 105% efficiency (oops! Off the track we go!). The "late fast" guys run consistently, at 80%, +/-10%. When they excel, they drive at 90% of their ability, leaving that little bit of reserve for safety, but when they "butcher" a corner, they're STILL driving at the AVERAGE for the "early fast" guys.

More seat time means more consistency, and better execution. That takes the 80%, +/-10% and makes it 85%, +/-5%. Even more seat time just continues to push the average closer and closer to 100% and shrinks the variance at the same time. Realize, though, that the "100% of ability" is a moving target as well. More seat time makes your 100% a LOT faster than a novice's 100%!

 

[kcbrown]

Actually, I'll disagree with you here... The reaction-time measurement is based on a randomly-generated stimulus, meaning your decision loop is something like: See the light, process the information, make the decision, act. This would be akin to something sudden occurring on track, like my interaction with a coyote last year. When talking about something like a brake point, or a turn-in point, there is the benefit of foresight, and anticipation. This is why I stress permanent landmarks with my students. You're closing in on the corner at 100MPH, and your PLAN is to hit the brakes at the "3" marker in order to get to your desired entry speed. Your decision loop starts seconds before you get to the action point, since the corner didn't just suddenly appear. You have the benefit of anticipation and timing in your favor. As you come up on the "3" marker, which is your braking point, you have already cut well inside of your decision loop, eliminating the process and decision portions. You KNOW you're going to brake, and you KNOW that you will brake at a specific point, which is coming up in 3...2...1...BRAKE! Your 250mS reaction-time loop can and will be shaved to something around a 10-20mS error factor. At speed, that's well within the 10' window that you want.

Hmm...there is some validity to that, it seems. I guess it remains to be seen if it's enough.

In my experiments with a "christmas tree" simulator (here), which gives you the anticipation you're talking about, my total end to end variation over 10 runs is about 150 milliseconds, with a standard deviation of 40 milliseconds.

Think of it like tapping your foot to music. If you wait to hear the beat, then trigger your toe to tap, you'll be behind. But, because you can anticipate the rhythm, you can rapidly fall into sync within a VERY narrow margin of error. If you have a non-movable landmark (I normally use pavement patches, braking markers, joints in the berm, etc.) for each of your cardinal points, you'll simply drop into the rhythm of things.

Makes sense. I guess only experience will show what the results will be here. It may be that the "christmas tree" experiment isn't terribly applicable to this.

After all, the brass ring is to see, process and act without conscious thought, allowing you to focus your mental faculties on strategy. That, however, requires a goodly amount of seat time to achieve!

Very nearly everything I do requires conscious thought on my part. Even walking. I have some (not a lot, but some) trouble holding a decent conversation when I'm walking because of it.

But given sufficient practice, I may be able to come close to the ideal here. While I'll still need to be consciously involved, I may be able to "optimize" things so as to minimize the amount required.

Maybe it's not really any different for anyone else, and I'm just more "aware" of it than most...

LOL!! I've yet to see somebody run out of NASA for being too needy, or wanting instruction! Unfortunately, I can't help you with simple memory tricks, or anything like that. Yes, you NEED to learn the track to be able to drive it consistently. Best bet is to start working now. Find a good YouTube video of the track you're going to drive, preferably from a Mustang, or Corvette, and not a Miata or a WRX. Something with reasonable horsepower, weight, and rear-wheel drive, and prefereably a camera angle from inside the car, behind the driver to give you a better example of what you will see when you're on track. Download the vid (or book-mark it!), and then find a nice, clean lap, and start dissecting it.

How will I know a clean lap?

Regardless, this looks like excellent advice, because for me, the most important thing is going to be recognizing what's coming up.

Ideally, I'd have the track in the video game. Gran Turismo doesn't have Sonoma Raceway, though.

It does have Laguna Seca, and I've been "driving" on that quite a lot, enough that I feel I know the track "by heart". It'll be very interesting to see how it drives and feels in real life.

Fantastic attitude! Keep that firmly in mind when you hit the track, and while you may be slower than some for the first session or two, you'll wind up being faster than MOST by the end of the day or weekend!

Oddly enough, based on my experience with the video game (the best place to discover such things!), I'm not going to time my laps. Or, at least, I'm not going to bother to look at those times until long after the event is over. The reason is that if I look at my lap times, I'll wind up "competing" against myself and pushing the car too hard. That inevitably results in off track excursions in the game, and I have no reason to believe it'll be any different in real life.

I would suggest working braking points first, but with good reason. If you start working on entry speed (and exit speed follows blindingly quickly, trust me!) you WILL start carrying more and more speed into the following corners. By having an established braking point you can start seeing the impact your speed has on braking distances, and you won't be surprised into a "four-off" at the end of braking. Also, start getting used to the concept of using your braking to determine your entry speed. If you brake however and wherever, you'll have a much tougher time getting consistent entry speed to start working with. Also, it'll make your instructor happy to know that you can whoa the car up properly. That will then allow him to focus on getting you hustling a little sooner without wanting to leave "a little in the bank for the wife and kids," if you follow.

Makes sense. I just don't understand how I can do that without creating serious risk of going off track while taking the corner, because the closer the braking point is to the corner, the more speed I'll be carrying and, thus, the more correct my path through the corner will have to be in order for the car to be properly positioned on exit.

THAT is an excellent question! We've pretty much figured out that the combination of turn-in point, apex point, and rate of acceleration post-apex will determine your track-out point. Look for the following:

1) Your car winds up right at the edge of the track at corner-exit. Not 6' in, and not with two in the dirt; try for 18" or less. I generally try to get my students to just "kiss" the berm with the edges of their tires. Feel that it's there, if you follow. That way you know you used the WHOLE track!

Well, but let's say that I take a relatively shallow entry line and a relatively steep exit line. Which is to say, let's say that I apex early. I still might hit the edge of the track on exit, but the line will still be all wrong.

2) You were able to smoothly unwind the wheel starting at apex, and were pointing dead-straight down the track when you got to the edge at exit. Do NOT drive the car there! Let the car flow there naturally.

I don't understand how I can possibly do that unless I've nailed the entry speed. Anything less than that requires that I actively steer the car to the exit point, right?

3) You were able to get to WOT somewhere between apex and the track out point.

Hmm...

Depends on the car, though, right? But I think I get the idea, that being that, basically, if you did things right, your curve towards the track out point will be gradual, enough that getting on the throttle won't upset the rear.

But that implies that if you didn't do it right, then the only way you'd really know that is through reluctance to get onto the throttle.

So that raises the question of just how much of a curved line you can be on and still safely get on full throttle. I'd guess that learning that takes experience. How do you learn that safely?

If you do the above, you hit a perfect exit. If however:

1) The car comes up too far away at exit (more than 18" off the edge), then you either turned in too late, or were a complete wuss with the throttle!

I expect to be the latter, if only because I want to maintain full control of the car at essentially all costs. I don't expect to truly be at full throttle until the car is going perfectly straight.

Simulators are great for "learning" a new track, but they're absolutely nothing like the "real thing." You can experiment with different lines through corners with them to see what seems to let the car flow (if the software physics package is goo), and if the graphics are done well enough, you can pick out some "prospective" landmarks. It will all change once you put rubber on tarmac, though! In iRacing (good sim!) my lap times at Road America are nearly 7 seconds higher than I've done in the real world. The sims get some things right, and others, not so much... Honestly, I've been pretty happy with doing the video thing, assuming you can find the right one to use as a guide.

Gran Turismo 6 seems to do a pretty decent job as far as the physics go, with perhaps the only questionable thing I've found so far being that easing up on the throttle makes it possible to correct an oversteer condition. However, that might actually be realistic for the cars I've been "driving" (which includes, of all things, a Boss 302! Fun.

But it also includes cars such as the Lotus Elise, Acura NSX, etc.). People who have driven their cars on the track in real life seem to think that GT6 does a pretty good job of making the simulated car "behave" like the real world car. I'm sure there are some subtle differences, but I'd hope the basics are the same.

I'm not at all planning on using GT6 to guide my expectations of how my car will handle on the track. I'll be approaching my car as if it's one that I don't know a thing about, because I don't! But I'm hoping that GT6 is at least useful for experiencing the concepts that you've been talking about here (the line through the corner, etc.). So far, what I've experienced with it seems to be consistent with what you've been saying. Somehow, I can "feel" when I nailed the corner, because things just "flow".

However, I seem to be worthless at being consistent in the game, even when I know the track intimately. My average lap time variation at "Laguna Seca" with the same car set up the same way, and this is when I'm trying to be consistent, is over a second per lap. I chalk that up to a combination of my reaction time variation and an inability to be precise with my car control, but it may be that there's something else going on. Whatever it is, I'm hoping that an instructor will be able to figure it out and tell me how to correct it. That presumes that the same thing will happen in real life, something that I can't actually presume at all, because there's so much more sensory input in real life.

This one's not ideal, but it might give you a start:

http://www.youtube.com/watch?v=yLezvppt_EM

Note on the first lap how he misses apex on one corner, and winds up well out on the rumbles at exit...

Yeah. It's as if he could have shifted his entire line a few feet over but otherwise kept it the same, thus nailing the apex and not going onto the berm.

Also note how he's experimenting with a couple of different lines through the hairpin.

I don't see that at all. You're talking about the first corner he takes (the first time around, being immediately after he took the point-by), right? His line looks the same to me.

ALSO note that he's pretty much hitting apex everywhere, as well!

I did notice that. If he's near the limits of adhesion, then it's impressive.

I'll be waiting (with bated breath!) to hear a report on how things went, what you learned, how much fun was had, what worked and what didn't, etc.

I'm really looking forward to the experience!

Biggest thing to beware of getting (back) on track for the first few times is your personal fatigue factor. Although you are driving in a seated position for relatively short stints (I mean really, 20 min? How hard can it be!), the mental drain is immense!

I imagine it is, especially when the track is new, the car is relatively unfamiliar, and the concepts are "new" (there's a huge difference between conceptual understanding and seeing how to put it into practice).

Stay hydrated, keep up on proteins and carbs, and if you feel like you're getting tired, or confused about something you thought you pretty much had down before, PARK THE CAR.

Great advice! Though for these two days, I'll be with an instructor the entire time. If I'm not getting something, I'll ask about it.

Also, listen to your instructor (duh!), he/she will have MUCH better specific information about the track than I could give you half-a-continent away.

Heh. I plan to! Regardless, I think the advice you've given so far will prove to be invaluable, and I'm looking forward to trying it out!

 

[Rabee]

Dave, great info thank you.
I read your sticky about driving techniques a while ago and it is amazingly helpful!

 

[claudermilk]

I'm pretty sure I read Dave's sticky before. I'm going to have to pull it up again now that it's not as much theoretical as before for me.

kc--I've spent some time on GT6 (and too much on GT5, 4, 3). It does do a decent job of presenting the tracks & the physics. I think part of why I naturally fell into a more-or-less correct line my first day out was all the time practicing finding them in the game. Oh, and IIRC, GT3 or 4 has Sonoma (as Infineon). I loved/hated it and am sad to still not see it back.

 

[DILYSI Dave]

Keep it floored until you see jesus, then throw out the anchor. Duh.

 

[SoundGuyDave]

Hmm...there is some validity to that, it seems. I guess it remains to be seen if it's enough.

In my experiments with a "christmas tree" simulator (here), which gives you the anticipation you're talking about, my total end to end variation over 10 runs is about 150 milliseconds, with a standard deviation of 40 milliseconds.


Makes sense. I guess only experience will show what the results will be here. It may be that the "christmas tree" experiment isn't terribly applicable to this.

This still boils down to a matter of control and prediction. With the "Christmas tree" simulator, you're still only looking at a 0.500 or 0.400 pace, but on the road course, you can generally see your braking point solidly and clearly for AT LEAST a second, usually more.

Very nearly everything I do requires conscious thought on my part. Even walking. I have some (not a lot, but some) trouble holding a decent conversation when I'm walking because of it.

But given sufficient practice, I may be able to come close to the ideal here. While I'll still need to be consciously involved, I may be able to "optimize" things so as to minimize the amount required.

Maybe it's not really any different for anyone else, and I'm just more "aware" of it than most...

This is entirely possible. Unfortunately, I'm a bit at a loss on how to help or give advice without "being there." Only thing I can say is to work with your instructor. Possibly a tempest in a teapot, possibly not, no way to tell from here.

How will I know a clean lap?

Regardless, this looks like excellent advice, because for me, the most important thing is going to be recognizing what's coming up.

Ideally, I'd have the track in the video game. Gran Turismo doesn't have Sonoma Raceway, though.

It does have Laguna Seca, and I've been "driving" on that quite a lot, enough that I feel I know the track "by heart". It'll be very interesting to see how it drives and feels in real life.

Ah, sorry. Vernacular. A "clean lap" is one without traffic (necessitating a variation from the ideal line) and one where a minimum number of errors are made. As a serious suggestion, starting the week before your event, stop doing any track-related activities (in your case Gran Tourismo) that are not specific to the one you're going to visit. The LAST thing you need is that moment of confusion when you equate T5 at one track with T5 at the one you're driving. Focus strictly on your "prep regimen" for Sonoma.

Oddly enough, based on my experience with the video game (the best place to discover such things!), I'm not going to time my laps. Or, at least, I'm not going to bother to look at those times until long after the event is over. The reason is that if I look at my lap times, I'll wind up "competing" against myself and pushing the car too hard. That inevitably results in off track excursions in the game, and I have no reason to believe it'll be any different in real life.

Excellent! In the HPDE-1/Novice group, the LAST thing you really should be focusing on is raw lap times. Worry more about technique execution, line consistency, and situational awareness. Even if you're running at 50% of capability, there's still lots to learn!

Makes sense. I just don't understand how I can do that without creating serious risk of going off track while taking the corner, because the closer the braking point is to the corner, the more speed I'll be carrying and, thus, the more correct my path through the corner will have to be in order for the car to be properly positioned on exit.

Remember when I said "start slow?" Here's where it pays off. As an instructor, I'm less worried about you quickly finding the ideal braking point than I am you finding a consistent braking point. If your initial point is 50' too soon, and you're 10mph too slow at entry, that is something that we can work with for a start. That lets us focus on finding the ideal turn-in and apex points, without worrying about carrying too much speed into the corner, pushing you past apex. Was it too late of a turn in? Too little steering input? Too much entry speed? This eliminates the variables. Don't push your braking point to start, just get one that's consistent. It will move during the course of the day or weekend. As will your turn-in and apex points, as you refine your skills and capabilities. You need to "rough in" a line before you refine, if that makes any sense. Let your on-scene instructor guide you, though. What I recommend works for me, and the vast majority of my students, but not all people are the same. I do have reasons for what order I focus on, and they're all safety-related.

Well, but let's say that I take a relatively shallow entry line and a relatively steep exit line. Which is to say, let's say that I apex early. I still might hit the edge of the track on exit, but the line will still be all wrong.

You are absolutely correct! A late apex line has a few hallmarks, and first and foremost is a tight initial radius to get from turn-in to apex, then a gradual increasing-radius to get from apex to track-out. With the scenario you described, a typical "early turn-in" or "early-apex," you WILL hit the edge of the track, but you'll be FAR from parallel (or near-parallel). In fact, you'll either be absolutely unable to unwind the wheel (and add power), or you will arrive at the edge of the track with significant wheel input still dialed-in. If you aren't virtually straight on the wheel when you arrive at the edge of the track, you need to turn-in later (assuming you hit your apex).

Short form: Missed apex= too much entry speed or improper steering input.
Hit apex, short on track-out= early turn-in
Hit apex, track-out not all the way to the edge=late turn-in

In practice (particularly if you're not at 100% speed!) it's fairly easy to find an arc where the turn-in, apex, and track-out points all connect. Then your job is to refine those points as your speed increases.

I don't understand how I can possibly do that unless I've nailed the entry speed. Anything less than that requires that I actively steer the car to the exit point, right?

Again, start at a lower speed, where you have complete control of the car, and minimize the error factor. Plan on leaving 50-100' too much braking room, so that your entry speed is positively under control. Once you have the basic line down, start picking up the speed by shortening your braking zones in progressive small increments.

Hmm...

Depends on the car, though, right? But I think I get the idea, that being that, basically, if you did things right, your curve towards the track out point will be gradual, enough that getting on the throttle won't upset the rear.

Exactly. When starting out, go easy, and just work with your instructor on the basics of the line. As you pick up pace, you'll start to feel out how much throttle you can add, and when.

But that implies that if you didn't do it right, then the only way you'd really know that is through reluctance to get onto the throttle.

Yes. In a traditional late-apex 90*-ish corner, you have a tight radius from turn-in to apex, using all of your grip budget to get the car turned. As you hit apex, you start to unwind the wheel, which increases the radius of the corner, allowing you to carry more speed, progressively. As you unwind the wheel (reducing the % of grip budget spent on turning), you should be able to start adding more throttle (increasing the % of grip budget spent on accelerating).

So that raises the question of just how much of a curved line you can be on and still safely get on full throttle. I'd guess that learning that takes experience. How do you learn that safely?

How much throttle to add is a function of the car's total grip budget, modified by track conditions. If you're on a slick track, or cresting a hill, it reduces the net available grip. Basically, if the car starts to understeer or oversteer, you're exceeding your grip budget. How to get close to your grip budget safely? Simple: small incremental changes. This presupposes consistency in the first place, though! Tire break-away is generally not sudden (with stock-ish suspension), but is gradual. Moving from 95% (perfect carving) to 101% (starting to slide) will not automatically put you into an unrecoverable spin. The tail end may start to slide out a little bit, in which case you start your oversteer recovery drill. If the car is heavily modified (very stiff whatever!) the "safe range" of overshoot on grip budget is reduced, and you can experience "snap-oversteer," where the tires simply stop gripping and just slide. This is extremely difficult to catch (even for pros!), and one of the main reasons that we generally recommend learning the basics of driving techniques and car-control skills prior to modifying the car for more than durability.

I expect to be the latter, if only because I want to maintain full control of the car at essentially all costs. I don't expect to truly be at full throttle until the car is going perfectly straight.

Not a bad plan, at all, however as long as you can be smooth and progressive on your throttle application (do NOT treat it like a light-switch!), you may find that there is simply no issue with starting to get on-throttle right at or after apex. No sudden moves, though!

Gran Turismo 6 seems to do a pretty decent job as far as the physics go, with perhaps the only questionable thing I've found so far being that easing up on the throttle makes it possible to correct an oversteer condition. However, that might actually be realistic for the cars I've been "driving" (which includes, of all things, a Boss 302! Fun.

But it also includes cars such as the Lotus Elise, Acura NSX, etc.). People who have driven their cars on the track in real life seem to think that GT6 does a pretty good job of making the simulated car "behave" like the real world car. I'm sure there are some subtle differences, but I'd hope the basics are the same.

Not familiar with that particular game, but yes, one would hope it would be fairly close to real-world.

I'm not at all planning on using GT6 to guide my expectations of how my car will handle on the track. I'll be approaching my car as if it's one that I don't know a thing about, because I don't! But I'm hoping that GT6 is at least useful for experiencing the concepts that you've been talking about here (the line through the corner, etc.). So far, what I've experienced with it seems to be consistent with what you've been saying. Somehow, I can "feel" when I nailed the corner, because things just "flow".

However, I seem to be worthless at being consistent in the game, even when I know the track intimately. My average lap time variation at "Laguna Seca" with the same car set up the same way, and this is when I'm trying to be consistent, is over a second per lap. I chalk that up to a combination of my reaction time variation and an inability to be precise with my car control, but it may be that there's something else going on. Whatever it is, I'm hoping that an instructor will be able to figure it out and tell me how to correct it. That presumes that the same thing will happen in real life, something that I can't actually presume at all, because there's so much more sensory input in real life.

You nailed the main problem with anything other than a fully-instrumented simulator. There are just sooo many other sensory inputs in the real world. As obvious as extreme peripheral vision cues (that give you a sense of relative speed by the rapidity of objects moving from focal vision through the end of peripheral visioin), or as subtle as the wind noise coming through the lowered side window. Don't kid yourself, your body and brain are processing all this!

Yeah. It's as if he could have shifted his entire line a few feet over but otherwise kept it the same, thus nailing the apex and not going onto the berm.

Yup. It looked to me like he was understeering a touch at 1:15 or so, and that pushed him past apex. I would bet that he was just asking a touch too much on colder tires, if I had to guess.

I don't see that at all. You're talking about the first corner he takes (the first time around, being immediately after he took the point-by), right? His line looks the same to me.

Sorry, I guess that would be the second hairpin. I'm not sure why he takes different lines in what (at least appears to be) are such similar corners. Starting around 1:45, you can see he took a wide line around the first half, preserving entry speed, dove down to apex for the second half, and then scooted out nicely. This is a classic line for a basic tight 180* corner, or one with a decreasing radius. Then, at 2:25, he runs a very traditional double-apex line, typical of what you might use for a pair of corners in quick succession in the same direction. Not having driven the track, I'm not sure what the advantage would be to changing up game plans like that.

I did notice that. If he's near the limits of adhesion, then it's impressive.

Hitting your marks! That's the key to it. Whether he's near the limit of adhesion or not, he hit his marks, and that's what the game is all about, and how consistency is built.

I'm really looking forward to the experience!


I imagine it is, especially when the track is new, the car is relatively unfamiliar, and the concepts are "new" (there's a huge difference between conceptual understanding and seeing how to put it into practice).

Great advice! Though for these two days, I'll be with an instructor the entire time. If I'm not getting something, I'll ask about it.


Heh. I plan to! Regardless, I think the advice you've given so far will prove to be invaluable, and I'm looking forward to trying it out!

Glad to be of service in whatever small way I can! Have fun, and let all of us know how it went...

 

[5iv]

Lots of great advice in this. Sub'd.

 

[Random Asshole]

Keep it floored until you see jesus, then throw out the anchor. Duh.

 

[SoundGuyDave]

One of the greatest movies, EVAH!!!

 

[Random Asshole]

Dave can describe a full lap of any track he's driven with his eyes closed while piss drunk, on fire, jumping out of a helicopter.

Me? I'm not worth two squirts of piss unless I'm sitting in your car. I'm no pro racer.. But, to my merit, I have turned a lap or two faster than Dave in a go kart with at least 40lbs of beer gut ballast. So take this for what it's worth.

Sounds like there may be some extraordinary personal pressure being applied here. Going fast consistently needs to feel like taking a Sunday stroll. This may be too basic of an approach than the OP is looking for, but maybe it will help somebody.

1. Start with finding the right entry speed to a corner. Approach the corner at something like 6/10ths and slowly ramp it up until things start getting shitty, then dial back. Take note of things like the position of the car before you turn the wheel, such as engine note/response at the point you began powering through when you feel like you nailed a corner.

2. Once you've found the correct entry speed, then start working on your braking marker. Stay on the gas a little longer and/or brake a little later until your entry speed is near or equal to the result of your braking point. Find a braking point that gives you the entry speed that keeps you on the "go pedal" once you get on it. Remember, once you are done with the brakes, you are done until the next corner.

3. Stay focused on your marks. Point to point to point. Don't bank on absolute maximum corner entry speed on every corner all the time. Leave a bit on the table and just focus on where the car is going next. Once you have that that down. You will naturally push harder without even realizing it.

4. Practice. Do it so many times that you don't look at your speedometer and can judge braking points just by depth perception, engine note, the force, etc.


Sounds kinda, "yeah, duh", I guess. But like I said, I'm not Dave.
And don't push yourself and overthink things. This is supposed to be fun, remember?