> when automatic transmissions are consistently
> outperforming humans at fuel efficiency
> when automatic transmissions are consistently
> outperforming humans at fuel efficiency
In a measured test, with a skilled driver, who does the right thing for the entirety of the test, it is most likely that he or she will be more efficient with a manual transmission than with an automatic.
However, in real life, for real humans, over a longer period of time, the drivers with modern automatic transmissions will be more fuel efficient than drivers with similar cars and manual transmissions, because the automatic transmission is a reliable machine that will do the same thing, every time. It doesn't care if you had your morning coffee or not, if some asshole cut you off, or if you got distracted or bored or spaced out and forgot to change your gear to the right one.
And, disregarding the disruption the EVs will be, technology constantly improves, while human skill is fairly constant, so it was just inevitable that automatic transmissions would one day overtake manual transmissions in terms of speed and efficiency. Arguing otherwise is just irrational nostalgia.
I've experienced quite a few humans who, when driving an automatic, seem to select a speed and throttle position that keeps the transmission right at the top of a gear range, with the engine turning much faster than necessary, instead of going slightly faster and letting it upshift. I doubt that's fuel efficient at all.
Even with an automatic the driver has influence on which gear is used, and a driver who naturally doesn't need to know about gearing may unintentionally be telling the transmission, via other inputs such as throttle position and speed, to select too low a gear than what a manual driver would choose.
(For instance, one of my cars has a manual gearbox with an electrohydraulic actuator system and an automatic clutch, operated by flappy paddles. My insurance company says that it is an automatic, while the equivalent of the DMV (correctly) states that it is a manual.)
I drive a manual and I know exactly why it is not fuel efficient: the idiot behind the wheel :-)
My own anecdote is that I'm a better driver in a manual. I'm a lazy driver, and lazy in a manual means avoiding stopping/gear changes by anticipating green lights and so on. Being lazy in an automatic results in more stop and go for me because it's no more effort. I could be as efficient in an automatic, if I put in the effort (which I would probably do if I knew my numbers were going into a study).
Other modern cars have an "eco display" or similar, which gamifies the experience. My previous car had some progress bars for different aspects, and I spent several boring commutes trying to get to a perfect 100%. (I did it once, I was so happy! Usually only got it to 98% before I hit the city and got stuck in traffic again)
Making automatics more efficient is a technological optimization problem. Making manuals more efficient is a human skill problem. I know which side I'd bet on in the long run. (Hint: Not the human)
That has the potential to be really dangerous though, at least where I drive(a third world country). If a car doesn't do what you expect it to when you provide a certain input, that's a recipe for disaster.
Large six cylinder revving to mid-range (say, 7500) with ample torque at all range can drive the car in overdrive even at city speeds, but an automatic will treat most change in throttle as a downshift. Because they have no ability to understand your intent.
I guess self driving car with an automatic can really fix that, especially because they can get geared optimally to the driving inputs the software tells them and they have 'inside brain' information.
I'm not sure I buy this. I noticed when I switched to automatic from manual I would drive in ways I would never have done with the manual because, well, it was easier than having to care about RPM, target speed, and current gear state: I'm not worried about automatic drivers being bad drivers, I'm worried about them not caring.
Here's the site: http://www.autoweek.nl/verbruiksmonitor
The vast majority, certainly >90%, of people who claim to be driving more efficiently manually are simply mistaken.
It's mostly observation bias. They recognize those two situations a day where they really have the upper hand. For a second or two. They don't realize that in most other driving situations they don't even come close.
Day to day driving is wildly different from that, there are only very few situations where people come anywhere close to the maximum acceleration their car would offer. With manual transmission, you can attain submaximal acceleration by flooring the throttle in a low gear. This is much more fuel efficient than getting the same amount of acceleration with a throttled engine at higher rpm. Automatic transmissions make that more difficult or even impossible (depending on implementation).
Worst case: people are unhappy with the acceleration they can get at low rev, because the transmission only allows low rev at throttled state and they don't consider it socially acceptable to rev up. The engine would be capable of much more torque at the desired rpm, if only the throttle could be opened without the transmission shifting for higher rpm. Next time they will get a bigger engine that will be throttled even more. ICE inefficiency rises with high rpm and closed throttle (opening the throttle raises power output more than consumption). Ideal ICE efficiency is when the gear is selected so that torque at open throttle matches the power demand. Manual can come much closer to that than automatic. (Unfortunately, many people don't know that and drive manuals like automatic, e.g. never at really low rpm, never at really open throttle, because they think that opening the throttle would be wasteful)
I'm not going to say "you're wrong" but I think there are plenty of situations where you know or can anticipate what gear you need to tackle a certain scenario, where an auto simply can't know.
For example, a set of 90 degree bends with (very) short straights between them (this is a real scenario where I live and why I've thought about this at length already). In a manual, you might ride the entire set of bends out in a low gear so you get both the most acceleration between bends and not have any gear changes mid-bend so you're ready to accelerate out quickly. An automatic, on the other hand, would be inclined to change up gear between bends and then have to change back down again once you want to accelerate out as it wouldn't really know what was going on.
My car's transmission has a wide variety of options, including a "race" mode, and I usually drive manual instead, because the semi-automatic options seem to be terrible at anticipating anything and coaxing them through throttle alone is an art beyond me.
I know what you mean, the similarity between an automatic and a race driver was only meant to be about the specific context of fuel efficiency, not about shaving of a few more second from your lap times. But in the end, both of those wildly different optimization goals can benefit from a driver's superior ability to anticipate future power demand. And both goals create comparable challenges in "coaxing them through throttle alone", which describes my frustration with automatics pretty well. In day to day driving, people seem to want a certain amount of acceleration at non-aggressive rpm, and automatics that make it hard or even impossible to tap the full low-rev torque of a given engine make people buy bigger engines than they would need if the transmission would better support their use case.
Give me an automatic that allows me to apply unrestricted (by backshifting) torque at any rpm and I would look at it much more favorable.
But note that the transmission in F1 cars is semi-automatic [1]
[1] https://www.formula1.com/en/championship/inside-f1/understan...
Grin. But seriously, at some point, it simply becomes a question of low level control vs improved high level functioning. I love manuals, and drive one, but the writing is clearly on the wall for good reasons. Fortunately, buying a hybrid forces the issue anyway, as others have noted.
I used to code intros in assembler on the Amiga, and did some assembler on UltraSPARC, so the joke is completely lost on me. I love assembler and low level control.
You get mad because your assumptions about your own superiority get placed under doubt by reality and physics? Must be hell of a way to live.
Sounds like a dunning-kruger effect and some anger management issues rolled into one.
Now you're incorrectly assuming that the engineers' and programmers' assumptions about how I drive are correct, and they know better than me. Who are they and what do they know? Who am I and what do I know? What if I'm a professional driver or an engineer just like them, or both? You don't know either of the two parties, so how could you possibly judge this correctly? The answer is - you cannot. You are berating me for challenging what you perceive to be an authority, as if I hadn't studied physics and mechanical engineering and knew nothing about those subjects, and even go so far as to imply I'm incompetent and suffer from an illusory superiority complex based on Dunning-Kruger research.
Is this what we're coming down to nowadays, if someone says "I know what I'm doing, and I don't want someone else to tell me how to do it", they'll be sticker-slapped with a superiority complex? That's the society you want to live in? And that is okay?
Bottom line is: a car with an automatic transmission does not do what I want, how I want, when I want it, and does not drive the way I want and expect it to drive. It fails my requirements, whereas a manual transmission and a dry clutch don't try to meet my requirements or expectations - what I want, how I want it, and when I want it is solely up to me. If I mess up, that's on me as well. It's my problem, my responsibility, and that's great.
As far as I'm concerned, since driving is a very important part of my life and identity, dictating, through the use of transmission or shift points, for example, is attempting to dictate how I live. I, an individual, equate that with someone telling me what I must believe, and what I must accept, "because it's better for me", and this goes against what I believe is my basic right to self-determination. That is one of the reasons I reject the automatic. If I have no direct say in it, then at least I can indirectly vote with my wallet.
Must be hell of a way to live.
I deal with it.
Now DCTs in road-going cars: undeniably automatics.
Probably not for old "3 gear with torque converter" or similiar.
However, it's more convoluted to use. You have to engage the clutch, put the shifter in neutral, and release the clutch again. Compared to automatic where you just have to hold your foot on the brake pedal.
Stop/start reduces fuel consumption on test cycles like NEDC for both automatics and manuals, but I'll bet in the real world the benefit is mostly seen with automatics as many manual drivers will rarely bother to use it.
This is how you should drive if you know you're going to be stopped for more than a few seconds. 'Convoluted' doesn't come into it - it's the default.
Now, when driving stop/start equipped vehicles, I always try to use it. But in a manual, it does takes more thinking and effort and I wonder how many casual drivers actually bother.
The thing is, most of the time you don't really know how long you're going to be stopped for. Will that light take 3 seconds or 30? A manual makes you decide whether to stop the engine, but an automatic just does it every time without having to think about it or perform extra steps. (although, you can override it if you know it's going to be a short stop by just pressing the brake lightly.)
I'm glad start stop is supported on manuals this way. It makes sense. I thought the stop start systems didn't really save much fuel, but I've read some real world tests that clock in at 8~10% savings. All that idle time does really add up.
Now I just don't care about that, since it was too annoying at stop lights, which has led to incidents of auto-stop and immediate start and even a half auto-stop. Didn't notice and bad effects on the car.
If you take a driving test in the UK and do this, that would be a minor fault. You can only have a few minor faults before you fail the test. The rationale is that you need to be in full control of the car at all times, and when you set off you mustn't drift backwards at all. The easiest way to do this without needing three legs is to use the handbrake. Unless you have a diesel engine that has enough torque without any pressure on the accelerator pedal to pull away.
For example if you are turning right and waiting for oncoming traffic to clear, then someone pauses and flashes their lights to let you turn, if you then turn that would be a minor fault in Driving Test conditions - where you are supposed to ignore the driver inviting you to turn and sit like a diddy.
Also you are allowed 15 minor faults before failure, which is quite a lot really.
And now days, newwer cars have the "hill hold" feature which automatically prevents you rolling backwards anyway.
I usually keep my foot to the side, as I was explicitly taught in driving school by a professional instructor to do. And if the master cylinder ever starts to malfunction, since the plate spacing is really tight, keeping the clutch pedal depressed while stopped could prematurely burn up the clutch and the torque plate.
So, it may require more thinking, but I can rest my feet while the light is red and it actually feels good.
Few years ago, I had colleagues who came from a certain foreign country and were not used to manual transmission. They got a company car with manual transmission. The clutch was burned in one month.
A clutch is destroyed by friction - slippage between the two plates that comprise it. This is a useful characteristic when pausing on a hill for a few seconds but I suppose there must be some people that keep the clutch partially engaged for a very long time.
But if you are sitting at a junction with the clutch entirely depressed for a long time the plates are disengaged and no damage is done. That's a bad habit for other reasons, such as if you get rear ended your car is still in gear and could then cause a further accident.
I generally don't. I too have driven several manual pickups, and only once had a squeaky throwout bearing. But I bought that one used so who knows what it had gone through before me.
Some engines are more susceptible to this than others.
Putting the car in neutral is basically one meaningful action - changing gear, yes you have to depress the clutch move the gearstick to neutral and then let the clutch go - but to describe that as "cycle the clutch twice" is madness. Am I missing something?
No madness involved.
(It happens to involve the clutch, but that is "automatic", as in "muscle memory", even though I drive less than once a month)
I would like to know:
1. How long you need to idle to recharge the electricity to battery you use to restart the engine?
2. Does NEDC and others require that the battery has same state of charge at the end of the cycle as when beginning?
Stop/start will automatically disable if the battery has a low state of charge.
It's a pain...and depending on the break, it really may become pain.
Usually if I drive my manual private car, I disengage the gear already on the way to the red light. It just rolls. I break, motor turns off. When I want to start, I push the clutch, motor turns on, off you go.
The good thing is, If I want the motor to launch, I can by pushing the clutch slightly. I can also release it and put my legs wherever I want. I can relax.
Holding one leg constantly on the break is just annoying. For you, your leg and the guy behind you who has to look into that bright red light pointlessly shining in his eyes.
It's a pain...and depending on the break, it really may become pain.
Oh, with an automatic you have the option of putting it in Park also. Certainly, it's a good idea to do that if you're stopped for more than a few seconds. You don't have to keep your foot on the brake, but doing so is enough to activate stop/start.
That's both a long-winded and incorrect way of putting it. If you are stopped, then you're either already in neutral with the clutch released or the clutch is already engaged. There is no other way.
Really, the rule is just "you have to have the clutch released". For experienced drivers of manual transmissions, this is one smooth action.
https://www.fueleconomy.gov/feg/PowerSearch.do?action=noform...