The online racing simulator
Where's the shift light?
(103 posts, started )
Not really. Not enough.

My race car. Peak power at 6250 rpm, peak torque at 5000rpm. When should I shift?
Quote from tristancliffe :Not really. Not enough.

My race car. Peak power at 6250 rpm, peak torque at 5000rpm. When should I shift?

Optimal shift point is above peak power point. How much above depends on the gearing and width of the power band. As I said you can't exactly tell but while actually in the car your senses help you. Sound of the engine, g forces .. i think plenty of those were mentioned above. Asked "How does a car feel on the edge?" Michael Schumacher replied "It feels optimum." (might be not in those exact words) they were talking about cornening but I don't think this makes much difference.

The point is to drive the car by feeling otherwise you'll endup looking at and thinking about too many wrong things. I mean do you really need a 100 meter sign to know where to brake?
Exactly. It isn't quite enough information. Which is what we're saying. If you knew the gearing and the torque or power curves you could work it out to the nearest rpm in theory (of course it's never like that in reality!).

You do drive by feel. But most people, talented drivers or not, often mistake noise and revs for acceleration. You FEEL quicker because the engine is revving really hard. A shift light is used by drivers to encourage them to shift earlier (and more optimally) than they otherwise would.

And yes, I do like to have 100m boards for, say, braking references. I'm a talentless waste of space in comparison to Schumacher (even Ralf), and I need all the help I can get!
Quote from AndroidXP :1) The ideal shiftpoint is the RPM value where changing to the next gear will deliver more torque than the current gear does.

I don't get it. Why all this talk about torque when it comes to shiftpoints? To accelerate as fast as possible you want as much power as possible, not torque? And a powercurve is (more or less) linear, so the optimal shiftpoint is always at the peak of the powercurve. If you shift before this point, there will be power in the powerband you'll never use, and after the shift you will end up at a lower point on the curve than if you shifted later. Whats the point of that?

Enlighten me please.
Good point there about noise and revs. Car with a loud exhaust will always be (feel) faster!

I also agree you can't tell the exact shift point but it doesn't all go wrong if you miss it by a couple hundred revs.

I check on your progress in F3 from time to time. You really aren't as bad as you make it out to be. I respect you a lot for what you do. Really.

Back on the visual reference for braking points and such topic. When approaching a corner I look at it and there's this point where I feel I can't go any further and I must brake to slow down in time and keep myself on the racing line at the entry. I know people who look at tyre walls, signs, lines ... you name it and are still way faster than me. I guess it's a matter of habit. I personally like the way I do it.
Quote from tristancliffe :In fact, peak power and peak torque are probably the most useless bits of information ever published.

It's possible to make a reasonable estimation of the torque curve from peak torque onwards from that data. Not quite reasonable enough to make decent predictions about shift points though.

In the absense of a torque curve, I'd much rather have those four figures than nothing at all.
Quote from nikka :I don't get it. Why all this talk about torque when it comes to shiftpoints? To accelerate as fast as possible you want as much power as possible, not torque? And a powercurve is (more or less) linear, so the optimal shiftpoint is always at the peak of the powercurve. If you shift before this point, there will be power in the powerband you'll never use, and after the shift you will end up at a lower point on the curve than if you shifted later. Whats the point of that?

Enlighten me please.

Nope, torque is what drives the wheels (and thus your car) forwards... Power defines how long you are driven forward (before you have to change gear)...
To be honest, you can use either. You shift at the RPM that gives more power AND torque in the next gear. Plus you can easily get the torque curve from the power curve and vice versa. As long as you know a few (say 10 minimum) data points of either curve you can work out the ideal shift point +-a couple of hundred rpm.
WHEELtorque.
Yes. Technically at the wheels for both. With torque that is gear multiplication and losses, with power it's just losses (power isn't multiplied by gear ratios)
I see shift lights are a bit misunderstood and maybe overrated here, some of the people who posted have probably never pushed a real life engine and don't know what it feels like, or their RL experience is limited to 1 or 2 cars.

First of all, yes you can judge when to shift either with the tachometer or by feel/sound. No you won't get it right the first time you drive a car, but as you learn it you learn where to shift. Basically not even a dyno is needed, giving full throttle from low revs up to the limiter in 2nd or 3rd gear (or even 1st depending on the power) will give you a good idea of where the engine starts pulling and where it stops. So basically, without being too accurate initially you have to keep the engine inside that RPM area. Usually you need to shift very near peak power, but not much past it especially when the engine is peaky as the power drops dramatically after peak power RPM. Then you need the next gear to be "in the power", meaning the RPM should be after peak torque. That's the very general rule. The point that each gear is different is not really valid, cause when you setup a race car, what you strive to achieve with the gearing except the correct top speed for the track, is even spacing between the gears, exactly to not have this problem. In road cars the gears are also usually pretty much evenly spaced, except maybe the top gear in some cars. As for not being able to calculate ideal shift points by feel, that is partly true, a mathematical analysis and calculation will be more accurate, but is really not needed, it's overkill as Tristan said.

Now, regarding the actual shift lights. Most race cars use preset RPM lights for the reasons I stated above. There are dynamic lights, but they don't work like in LFS, they don't calculate the ideal shift point, they can just be set to come on at different RPMs depending on the gear you're in. The calculation of the ideal shift point for each gear will still have to be done by you. As for beeping lights... Well. I have only ever come across two cars with those in my life, they are completely pointless, redundant and distracting. No they are not preferred by racing drivers, and no they are not common in F1. Also, shift lights IRL are not set to come on at the ideal shift point, they are set to come on just a bit before that, so by the time you react to it and reach out to shift, you are at the shift point. That's why you can see in race cars, the driver usually in high gears (where the acceleration is slower) will wait a couple of seconds after the shift light comes on before shifting, to let the RPMs build up to the shift point. The exception is when the car has a row of lights (like F1 cars) to alert you the shift point is coming up. In that case, last shift light comes on=ideal shift point.

While not all drivers (even top level ones) can always get the most out of their engines (some are better than others), any good driver after a short time in a car can judge when to change gear. More precisely, after shifting he can easily spot if he has shifted either too soon or too late. Maybe he can't get it within 100 RPM of the ideal shift point, but he can get it as close as makes no difference.

Ultimately IMO, shift lights are only really necessary on high powered cars, where sometimes you can have a hard time keeping track of what's going on around you, controlling the car and monitoring the tachometer/sound/vibrations. Especially in the lower gears where the acceleration will be very fierce and the RPMs will build up very fast while it might take a lot of concentration to put all that power down.

That's all, thanks for reading this far.
Shift lights are probably more useful in games because we can't really feel the power.

Quote from Kosmo :I As for beeping lights... Well. I have only ever come across two cars with those in my life, they are completely pointless, redundant and distracting. No they are not preferred by racing drivers, and no they are not common in F1.

I assume this is in response to me since I mentioned my beeping RX-7. (or did some else mention a beep?) I just want to clarify that it is not a shift light. It is a factory installed safety feature (also present on the RX-8) which comes on just before redline, which may happen to be the ideal shift point. I don't know.

The car has a rotary engine, which revs quickly and smoothly. The beep (actually more of a buzzer) is mostly for people who are used to driving piston engines which tend to approach redline with lots of noise and vibration. With a rotary, 2k rpm feels the same as 8k. Also the powerband is really linear, so you don't really feel the acceleration much. It's really smooth so it feels slower than it is. The power seems closely proportional to the revs, at least in my non-turbo car. Once you get used to it, the beep is not necessary (especially for racing) but it's a good feature for the street and rotary newbies.

It does have one advantage over a light: it's not visually distracting. Of course on a race car it would have to be impractically loud.
Oh I don't remember who said it. I know certain cars have this as factory equipment, I wasn't talking about those, since as you said those are there primarily to protect the engine from inexperienced drivers. Also, car companies often like to show off with gadgets so a simple light wouldn't cut it anyway.
#90 - mdmx
Hmm.. first of all the exact ultimate shift point really isn't THAT important, is it?? As long as you don't let the engine hit the rev limit, and don't change in the middle of the best torque zone, you lose almost nothing.

For example you lose more if you shift up 1sec before the next braking point, better to overrev the engine a bit and adjust gears a before next race. Not to mention shifting in the middle of the corner and losing grip for instant and forced to take too wide driving line.

As for shifting by hearing, yes, i think you definetly can learn that. But sure, it takes a bit of learning. Shifting point is a bit different on all the gears, depending the setup. But after you drive few laps, you start to learn the decent shift points, based on what you hear. Again, it's not THAT important to shift exactly on the right rpm. If you are focusing mostly to shifting, you can do much better if you start to focus on driving line and keeping the car balanced with throttle.

The thinner the good torque zone is, the more important shifting point is, obv. The more engine is revving, the harder it is to hear the amount of torque from sound. With F1 it's waaay more important than with road cars. And with rally car on gravel you can't hear the engine. Same with super high power track cars, the gearing and diffs are so loud it's hard to hear the revs.

So, personaly i really don't care about road car shift lights... Never payed too much attention to them before, on long straights i used to watch em, but it's really not that hard to shift without. And it's same for everyone, just focus on another things and forget the damn light.

And what goes to custom installed shifting lights, well, if the devs would took that route then it's very easy to install adjustable suspension to road cars as well, still they are removing the setupping possibility. Let's think them as stock road cars, and have fun with em.
Quote from Joris :Now it's time to lose the gear indicator on car's with H-pattern gearboxes.

It's really not that hard to put a sensor on the gate that tells you what gear it's in. The 370Z does this as do many other road cars, and many grassroots club racers do this.

Besides, those of us with paddle or sequential shifters like the gear indicators!


Quote from mdmx :Hmm.. first of all the exact ultimate shift point really isn't THAT important, is it?? As long as you don't let the engine hit the rev limit, and don't change in the middle of the best torque zone, you lose almost nothing.

Why go to all the trouble of learning to corner, learning to brake and learning to pass if you're not going to bother learning to shift? Faster is faster. Finding the optimum shift point is easier than heel-toe-downshifting - just run some numbers and try to maximize horsepower across the ratio. A good shift is worth a quarter car length in a drag race down the straight - multiple optimum shifts could get you a very easy pass...
Real Manly men don't need a shift light
Real manly men don't even need to shift
Quote from Kosmo :Real manly men don't even need to shift

+1 hahaha

Well, some road cars have it; the vast majority, not.

I think it's ok to take it out of them, but shift light needs a bit more attention on race cars... The only things that you really HAVE to see all the time in a cockpit is the tachometer, the gear indicator and the shift light...
Quote from Kosmo :Real manly men don't even need to shift

Not with their hands, anyway. :hide:
#96 - mdmx
Quote from MadCat360 :
Why go to all the trouble of learning to corner, learning to brake and learning to pass if you're not going to bother learning to shift? Faster is faster.

Those things affects much much more to your lap times than perfecting the shifts, first things first, right? And why don't you learn to shift then, instead of complaining about shift lights?


Quote from MadCat360 :
A good shift is worth a quarter car length in a drag race down the straight - multiple optimum shifts could get you a very easy pass...

Quarter car length compared to.. what.. Perfect shift vs good shift? I don't think so.

With intelligent shift lights, like in LFS, which knows the exact optimal shifting point on every gear, nobody gains nothing. Without shift lights, the ones who actually bothers to learn how to shift will gain that small advantage. Which is very good thing.

Most stock road cars doesn't have factory installed shift lights. Of course you can install one, you can also install adjustable suspension and bigger turbo.
If you really want to know the perfect shiftpoints for an individual set you can log pulls through each gear at the drag strip, then exporting the telemetry data and calculating your instantaneous acceleration. I had done something similar a few years back on my car each time I did mods that effected the power band.

In most cases though, you will only save tenths of a second on shorter tracks, maybe a second on long tracks that require a lot of shifts as mentioned previously.
Quote from mdmx :Those things affects much much more to your lap times than perfecting the shifts, first things first, right? And why don't you learn to shift then, instead of complaining about shift lights?

Quarter car length compared to.. what.. Perfect shift vs good shift? I don't think so.

With intelligent shift lights, like in LFS, which knows the exact optimal shifting point on every gear, nobody gains nothing. Without shift lights, the ones who actually bothers to learn how to shift will gain that small advantage. Which is very good thing.

I'm not complaining about shift lights. I was commenting on the idea that an imperfect shift doesn't matter - it does. First things first, yes, but you're not cornering while you're going down the straight are you? A quarter car length is easily the difference between an average shift and a perfect shift, once you factor in momentum losses and speed at the end of the straight.

If two identical cars are going down a half mile straight, having just exited a 2nd gear 40 MPH corner at exactly the same speed and directly alongside one another, perfect shifting versus good shifting will win the drag race before the braking zone.

The benefit for total lap time for perfecting your shifts is small, but what else can you do on the straight other than draft?
Motor racing isn't about finding large advantages. It's about sewing together lots of tiny ones - like shifting better (0.05 seconds per lap), lower drag wheel bearings (0.05 seconds per lap), being lighter (0.1 - 1.0 seconds per lap), increasing downforce (or decreasing lift) without increasing drag (0.01 - 3.0 seconds per lap)....

There is no one big BEST thing to do in racing. But if you aren't doing your shifts well then you are going to be slower than if you are. Surely that is all the motivation one needs to learn about it, and to do it better?
Quote from tristancliffe :Motor racing isn't about finding large advantages. It's about sewing together lots of tiny ones - like shifting better (0.05 seconds per lap), lower drag wheel bearings (0.05 seconds per lap), being lighter (0.1 - 1.0 seconds per lap), increasing downforce (or decreasing lift) without increasing drag (0.01 - 3.0 seconds per lap)....

There is no one big BEST thing to do in racing. But if you aren't doing your shifts well then you are going to be slower than if you are. Surely that is all the motivation one needs to learn about it, and to do it better?

Hold on, first you tell us the advantage of perfect shifting is negligible, now you say that every bit - however small it might be - is important... What's it gotta be now?

Where's the shift light?
(103 posts, started )
FGED GREDG RDFGDR GSFDG