How does torque affect car performance?

  • Thread starter Tekku
  • 135 comments
  • 17,286 views
593
Sweden
Umeå
I remember when Jeremy Clarkson made jokes about torque on Top Gear, stating that everyone likes to use it as an indicator for power, even though they barely know what it actually is.

Currently I'm on a mission to learn more about the detail of car mechanics, to better help my understanding of car performance and tuning in GT Sport. And torque is at the top of my list.

So what is torque and how exactly does it affect my car on track? Can you give me some scenarios where there's two different outcomes depending on my level of torque?
 
https://en.wikipedia.org/wiki/Torque

Whether torque differential between cars is modeled in GT and how well, remains to be seen. Without testing facilities or a way to measure acceleration it'll be hard to prove. Of course people will pop in and say they can feel it but it's anecdotal.
 
Torque is the rotational force. Torque multiplied by rotational speed results in power.

Power is generally what you want to look at. Because for example, in a theoretical car without any sorts of losses, power would be the same throughout the whole powertrain (engine plus drivetrain). In contrary, torque varies due to changes in rotational speed (the whole point of having a transmission).

Torque isn't useless, it's important for properly designing parts that must be able to withstand the rotational forces/stresses.
 
Torque is the rotational force. Torque multiplied by rotational speed results in power.

Power is generally what you want to look at. Because for example, in a theoretical car without any sorts of losses, power would be the same throughout the whole powertrain (engine plus drivetrain). In contrary, torque varies due to changes in rotational speed (the whole point of having a transmission).

Torque isn't useless, it's important for properly designing parts that must be able to withstand the rotational forces/stresses.

But in terms of selecting cars with good power values the torque might not be the most relevant stat to look at? In what cases would it be good to look at the actual torque value?
 
Torque is the rotational force. Torque multiplied by rotational speed results in power.

Power is generally what you want to look at. Because for example, in a theoretical car without any sorts of losses, power would be the same throughout the whole powertrain (engine plus drivetrain). In contrary, torque varies due to changes in rotational speed (the whole point of having a transmission).

Torque isn't useless, it's important for properly designing parts that must be able to withstand the rotational forces/stresses.

Power is not constant, it's the result of a combination of torque and rotational speed. If you have 1,000 ft-lbs of torque but 0 RPM, you have zero hp. That same 1,000 ft-lbs at 5,252 RPM, however, is 1,000 hp.

Or were you talking about the effects of gearing? Gearing multiplies the input torque at the reciprocal of the the rotational speed multiplier (that is, double the speed and the torque is halved, halve the speed and the torque doubles). Because horsepower is a product of torque and rotational speed, The reciprocal relationship of the torque and rotational speed multipliers results in constant horsepower at both the input and output (neglecting frictional loses, of course).

The old saying that torque is what moves the car is true, but what this saying ignores is that the car moves due to torque *at the wheels* not torque at the flywheel. For this reason, it's better to have an engine that makes power at higher RPM because it allows a car to use lower gearing longer. Because of inherent trade offs in internal combustion engine design, this usually means an engine with less torque at the flywheel and the increased RPM making up the extra power.

To answer Tekku's question about what is important to look at, it's the torque curve (including redline and power peak values) along with the overall gearing ratios. You need to look at all of those bits of information and how they interact with each other to get an idea of how the car will feel.
 
Power is not constant, it's the result of a combination of torque and rotational speed. If you have 1,000 ft-lbs of torque but 0 RPM, you have zero hp. That same 1,000 ft-lbs at 5,252 RPM, however, is 1,000 hp.

Or were you talking about the effects of gearing? Gearing multiplies the input torque at the reciprocal of the the rotational speed multiplier (that is, double the speed and the torque is halved, halve the speed and the torque doubles). Because horsepower is a product of torque and rotational speed, The reciprocal relationship of the torque and rotational speed multipliers results in constant horsepower at both the input and output (neglecting frictional loses, of course).

The old saying that torque is what moves the car is true, but what this saying ignores is that the car moves due to torque *at the wheels* not torque at the flywheel. For this reason, it's better to have an engine that makes power at higher RPM because it allows a car to use lower gearing longer. Because of inherent trade offs in internal combustion engine design, this usually means an engine with less torque at the flywheel and the increased RPM making up the extra power.

To answer Tekku's question about what is important to look at, it's the torque curve (including redline and power peak values) along with the overall gearing ratios. You need to look at all of those bits of information and how they interact with each other to get an idea of how the car will feel.
This highlights the weakpoint of all of these discussions. That is, the intermingling of real life principles with game physics. We have no idea how PD models torque or HP or what they take into consideration and can't assume anything at all. I've never found any advantage, in any GT game, of taking the torque curve into consideration when tuning.
 
This highlights the weakpoint of all of these discussions. That is, the intermingling of real life principles with game physics. We have no idea how PD models torque or HP or what they take into consideration and can't assume anything at all. I've never found any advantage, in any GT game, of taking the torque curve into consideration when tuning.

I thought I could feel it sightly in GT6, but it may have simply been illusory. It didn't seem to make much difference in tuning, but I did think I could feel a difference in when a car seemed to pull the strongest and whether I could really put my foot down coming out of a corner. Less of a lap time thing than a feel thing.

Of course, without a difference in lap times the chances of it all being in my head are just that much greater. :)
 
I thought I could feel it sightly in GT6, but it may have simply been illusory. It didn't seem to make much difference in tuning, but I did think I could feel a difference in when a car seemed to pull the strongest and whether I could really put my foot down coming out of a corner. Less of a lap time thing than a feel thing.

Of course, without a difference in lap times the chances of it all being in my head are just that much greater. :)
GT5 had a wonderful system for testing acceleration properties of each car. I did a lot of testing and experimenting with gearing options and always had my best results by using the peak of the power curve. Why they took out that simple but incredibly useful feature is beyond me. I highly doubt anything has changed in this regard but I don't know any way of testing it accurately now.
 
I like using a bicycle to describe torque and horsepower. When you ride a bike, you have to push on the pedals with your feet. If you keep doing that over time you build speed. The force you exert to push down the bicycle pedal is torque. The momentum that's gained from pushing the pedal multiple times over a period of time is power. In other words, torque is the strength of your legs. Power is you pedaling multiple times.

The same concepts applies to car engines. The only difference is that a car engine is far more complex than a human pedaling a bike. A human pedaling a bike has peak torque at 0 RPM, and that's because a human's leg strength is always constant. A naturally aspirated car engine on the hand generally produce peak torque at 3000 RPM's. This is because the engine's fuel and air quantity, mixture, and efficiency, compared to its friction is at its best, and that's when it produces the most strength. As the RPM's start going higher, the engine's strength begins being applied multiple times over a given period of time, and that's how horsepower is derived. Keep in mind that 3000 RPM's is an arbitrary number and was only used for explanation sake. Peak torque is dependent on an engine's design and varies for every car.

So how does this apply to Gran Turismo? Well depending on a car's torque figures and at what RPM it achieves that peak torque, the car will react differently upon acceleration. For example, a Subaru BRZ has peak torque of 155 ft-lbs at 6500 RPM's. A BMW M4 has a peak torque of 411 ft-lbs at 1200-5500 RPM. The Subaru BRZ will struggle accelerating, especially in higher gears, unless it's revved out to at least 6500 RPM. A BMW M4, on the other hand, will accelerate with ease, even in higher gears, starting at 1200 RPM. It's also important to keep peak torque in mind when cornering. The car with the most torque at a given RPM is more susceptible to spin its tires if given too much throttle. For example, if both the BRZ and M4 are taking a corner in 3rd gear at 3000 RPM at full throttle, the BMW M4 will spin out and BRZ won't, because there's far more power being applied to the M4's rear wheels than there is for the BRZ.

I hope that helps. That was the best way I could explain it.
 
Last edited:
I thought I could feel it sightly in GT6, but it may have simply been illusory. It didn't seem to make much difference in tuning, but I did think I could feel a difference in when a car seemed to pull the strongest and whether I could really put my foot down coming out of a corner. Less of a lap time thing than a feel thing.

Of course, without a difference in lap times the chances of it all being in my head are just that much greater. :)
I remember driving the Aston Martin DP-100 in GT6, and I noticed that shifting at lower RPMs (I think around 5000, not too sure) would help the car reach its top speed quicker than shifting at high RPMs - by a significant margin too, if I remember correctly. Thinking about it now, I'm not sure if that was because that was the range where torque was at its maximum, or the range where power was at its maximum.
 
Another factor with torque that is important to remember is that as it's a measure of work, it's not overall reduced by the drivetrain as bhp is.

But multiplied by the ratio of the gears, hence the reason that even with modest torque you can overcome tyre grip under acceleration in first gear, but barely if at all overcome the tyres rolling resistance in higher gears.
 
Another factor with torque that is important to remember is that as it's a measure of work, it's not overall reduced by the drivetrain as bhp is.

This is wrong. In order for bhp to be reduced, torque *must* be reduced. If torque were not reduced, the bhp value would remain constant. The two numbers are inextricably tied to each other.

Now, even with friction gearing can result in a net torque increase at the wheels, but this will always be less than the ideal number calculated based on the ratio without accounting for friction. For example, a 2:1 ratio, without friction, would give 2x the input torque with half the rpm and therefore would have the same bhp. In reality, with friction, a 2:1 ratio gives something like 1.85x the input torque at half the RPM and therefore slightly less horsepower. You're still technically reducing torque along the way; if you weren't, bhp could not decrease.
 
This is wrong. In order for bhp to be reduced, torque *must* be reduced. If torque were not reduced, the bhp value would remain constant. The two numbers are inextricably tied to each other.

Now, even with friction gearing can result in a net torque increase at the wheels, but this will always be less than the ideal number calculated based on the ratio without accounting for friction. For example, a 2:1 ratio, without friction, would give 2x the input torque with half the rpm and therefore would have the same bhp. In reality, with friction, a 2:1 ratio gives something like 1.85x the input torque at half the RPM and therefore slightly less horsepower. You're still technically reducing torque along the way; if you weren't, bhp could not decrease.
You missed a rather key word from my post, 'overall'

Overall, even taking into account frictional loses torque is increased by gearing (assuming that the input gear is smaller than the output one).

http://www.sci.brooklyn.cuny.edu/~kammet/gear_notes.pdf
http://mechanicalmania.blogspot.co.uk/2011/07/how-does-gear-ratio-affect-torque.html
https://en.wikipedia.org/wiki/Gear_train#Torque_ratio
https://www.nord.com/content/gear-ratio


Yes power and torque are related, but that doesn't mean that they are affected in the same way by gearing.
 
I would still argue that torque is indeed "overall" reduced because the result, at output, is less than in an ideal, frictionless, system.

I don't understand the relevance of your links as they do not offer any rebuttal to the more detailed explanatory portion of my post. Not do I understand your comment, "but that doesn't mean that they are affected in the same way by gearing," since I rather explicitly pointed out that they are not affected the same way.

I agree that in an ideal system torque is multiplied (either increased or reduced, depending on whether the multiplier is greater or less than 1) while bhp remains constant. However, in a real system friction results in some torque loss overall because you're getting less torque at the output than the multiplier would lead you to expect. Yes, there might be a net increase, but there is an overall loss because you cannot have a loss of bhp without a loss of torque.

That said, I expect that now we're mostly just quibbling over whether "overall" can be used as a synonym for "net" in this situation. I suggest not, you suggest yes, and the reality is that we're probably both right and both wrong depending on whether such linguistic choices are disclosed explicitly at the initiation of a discussion.
 
Power is not constant, it's the result of a combination of torque and rotational speed. If you have 1,000 ft-lbs of torque but 0 RPM, you have zero hp. That same 1,000 ft-lbs at 5,252 RPM, however, is 1,000 hp.
Obviously, combustion engines don't idle at 0 RPM. The start/stop feature for traffic lights doesn't count for other obvious reasons.
I was talking about the transition throughout the drivetrain. The engine outputs power (or torque at given rotational speed, whatever), which then drives the wheels by transitioning through the drivetrain.

Let me quote myself:
Torque is the rotational force. Torque multiplied by rotational speed results in power.
I thought that was clear enough.
Or were you talking about the effects of gearing? Gearing multiplies the input torque at the reciprocal of the the rotational speed multiplier (that is, double the speed and the torque is halved, halve the speed and the torque doubles). Because horsepower is a product of torque and rotational speed, The reciprocal relationship of the torque and rotational speed multipliers results in constant horsepower at both the input and output (neglecting frictional loses, of course).
Person looks at engine output, car is moved by the output at the wheels with different rotational speeds. That's why the transition matters. You can calculate the whole day long with torque and rotational speed, at the end you'll end up with the same.

The old saying that torque is what moves the car is true, but what this saying ignores is that the car moves due to torque *at the wheels* not torque at the flywheel. For this reason, it's better to have an engine that makes power at higher RPM because it allows a car to use lower gearing longer. Because of inherent trade offs in internal combustion engine design, this usually means an engine with less torque at the flywheel and the increased RPM making up the extra power.
Because you came up with a 0 RPM example, I'll give you this one: Take your 1000 ft-lbs and apply it to a wheel with an infinite radius, what will the acceleration of the car be?
Might sound ridiculous, but still an interesting and useful question.

To answer Tekku's question about what is important to look at, it's the torque curve (including redline and power peak values) along with the overall gearing ratios. You need to look at all of those bits of information and how they interact with each other to get an idea of how the car will feel.
The characteristics of the torque curve apply to the power curve too. There's no way an engine can magically have more torque at a given rotational speed without a proportional increase in power. A curve is just the output over a range of different rotational speeds.
 
Last edited:
I thought that was clear enough.

It was not, else I would not have asked.

Person looks at engine output, car is moved by the output at the wheels with different rotational speeds. That's why the transition matters. You can calculate the whole day long with torque and rotational speed, at the end you'll end up with the same.

In other words, you agree with me.

Because you came up with a 0 RPM example, I'll give you this one: Take your 1000 ft-lbs and apply it to a wheel with an infinite radius, what will the acceleration of the car be?
Might sound ridiculous, but still an interesting and useful question.

The answer depends on how that 1,000 ft-lbs is applied to the infinite radius wheel. If the infinite-radius wheel is driven by an infinite-radius gear, the wheel gets 1,000 ft-lbs and the car accelerates fairly rapidly. ;)

But I suspect that your thought exercise contained the unstated premise that the torque was applied via a drive mechanism consisting either of finite-radius gears or concentric with infinite-radius wheel. In which case your effective multiplier is 1 over infinity resulting in, for all practical purposes, zero acceleration.

The characteristics of the torque curve apply to the power curve too. There's no way an engine can magically have more torque at a given rotational speed without a proportional increase in power. A curve is just the output over a range of different rotational speeds.

Yes, I have already said as much. However, you can't just look at the peak value, or any value at a single given rotational speed, to understand how the car will accelerate. You have to look at the entire curve, along with gearing, to know how the car will operate. For example, let's say that car A has a flat torque curve from 2,600 RPM through 5,200 RPM. That's going to result in a very different dynamic for how the car feels than a car with an engine having the same amount of torque but instead at a sharp peak available only between 4,800 and 5,000 RPM even if both cars are otherwise perfectly identical (gearing, weight, aero, etc.). You would not be able to tell anything meaningful about the differences in how the two cars felt simply by comparing only the torque (or horsepower) at 4,900 RPM. You would need to look at the entire curve to be able to have any insight into the differences in how the cars would feel.

So, I say again, to answer the OP's question about what's important to look at, you need to look at the torque curve (including redline and power) and gearing, and understand how they all interact with each other, to get an idea of how the car will feel.
 
I remember clarcson on top gear sls amg test said that its just too much torque if you push it too far it kills you.
But in game sls amg drives like kitten never bites or tryes to kill you. Then again m4 its almosf undriveable. So something must be wrong there.
Try same sls amg in assetto corsa and its looses traction on straight road.
 
Gears don't multiply torque. (except on machines calibrated for a 1-1 gearing ratio)
Your vehicle makes effectively the same power and "torque"(which horsepower is) in every gear.

Gearing changes your rpm at a given speed, in a given gear. A car that has 500hp at 5,000rpm, has 500hp in 1st gear, and also has 500hp in 4th. Outside of some very small potential frictional/weight differences, of course.

Cars accelerate faster with quick gears, because they reach peak power more quickly.
If you drag race two cars, with different gear sets, you can usually make a movie out of it, because the quick gearing car pulls ahead, then shifts as the longer geared car hits peak power, and takes it's turn pulling ahead. No, it doesn't work that easily in every scenario, but it does in many.

Just for example, a car might reach 130mph in 4th gear @ 6,000rpm(max). 5th gear is too long to accelerate with, 130mph top speed.
Shorten the gearing so 4th reaches 120mph @6000rpm, and it will accelerate to the end of 4th gear more quickly, but won't reach 130mph as quickly, if at all. Shorten it even more, 4th only reaches 110mph, now the car can pick up speed in 5th gear. Now the car accelerates faster constantly, and reaches 130mph quicker than the original, but also has a top speed of 137mph.
On the opposite end of the spectrum, lengthen the gears. Now 4th gear reaches 137mph, but the acceleration is slowed.

Why? The amount of time spent at maximum horsepower.


Ya know what never, ever happens? Changing a gear ratio and suddenly having 510hp on your 500hp car.


If anyone disagrees, feel free to link the horsepower-adding gear set for sale.
 
Last edited:
As others have said, you need to look at power output over the whole (useable) rev range.

The engine in my daily driver would beat the old F1 NA engine from 1k rpm to 4k rpm because it has more power at that rev range.

More power at a given rpm = better acceleration at a given rpm.
 
I always see torque as a form of good acceleration.

I once drove a Ford Focus Mk1 TDdi of only 90 BHP company car but it had a really nice flat torque vs rpm curve for a very wide rmp range, meaning regardless of how high or low I was in rpm, if I stepped on the gas then the car was immediately responsive with acceleration (due to having a good amount of torque for that rpm) and it felt really good. It makes the car feel agile at a wide range of rpm for a certain gear you are in.

I have also driven similar TDi cars of 110-115 BHP, that is more power but if the torque vs rpm curve is like a very high and steep "mountain" then the car feels good if you stay in that narrow rpm range, but if you are out of that optimal range then the car feels slow and unresponsive.

Both cars were manual stick shifters and the 90 BHP car was a joy to drive, when I was not in the optimal rpm range to accelerate (for example coming out of a corner or after a traffic manoeuvre), if at any given situation I stepped on the gas pedal I felt immediate reaction and acceleration. Those 110 BPH cars with their less optimal torque curve accelerate faster in their optimal rpm range but once you are a bit below that optimal rpm and hit the gas, it feels very slow in response.

Not sure how all that translates to GT:Sport but that's my real life experience of it.

This for example is a great torque curve, nice flat curve over a wide range of rpm. Basically anything between 1000-6000 will give a fast response in regards of torque. But also looking at the BHP curve, this is an engine that you want to drive in high revs (5000-6500 rpm-ish) for racing since anything lower will have lesser BHP (but still good torque).
images


And this for example is a worse torque curve, the optimal range only being between 1500-3500 rpm. Add in the BHP as well and this is a car that should be driven to race in the range of 2800-3500 rpm.

twin-power-turbo-6-cyl-diesel-torque-curve-en.jpg.resource.1373896855626.jpg
 
And this for example is a worse torque curve, the optimal range only being between 1500-3500 rpm. Add in the BHP as well and this is a car that should be driven to race in the range of 2800-3500 rpm.

twin-power-turbo-6-cyl-diesel-torque-curve-en.jpg.resource.1373896855626.jpg
Completely, 100% factually incorrect.
You would take that all the way to redline.

@ 3,000rpm, your bottom graph makes a measly 555,000 lb-ft of torque per minute.
@ 5,000rpm, it is making 1,155,440 lb-ft per minute.

Unless you think 555,000 is a greater force than 1,155,440, you take it to 1,155,440.
 
Completely, 100% factually incorrect.
You would take that all the way to redline.

@ 3,000rpm, your bottom graph makes a measly 555,000 lb-ft of torque per minute.
@ 5,000rpm, it is making 1,155,440 lb-ft per minute.

Unless you think 555,000 is a greater force than 1,155,440, you take it to 1,155,440.

How high you take it in revs all depends on the gearbox as well. At 5000 rpm there is a big torque drop-off already, if you shift a gear up and you end back in the optimal rpm range it's better to shift.

Not sure what you mean with torque per minute.
 
Yes but numbers in torque in every car in real life gives different results. Because every car its just different.
And now we have those different racing games where you can find how different same cars handling there.
 
Back