[DChaffins]

Surely now something will be done.


http://blogs.insideline.com/roadtest...y.html#poll-62

[BMWrules7]

The delay you refer to is a safety delay. Pretend for a moment that you are a computer programmer.

Your program monitors events. For example, you monitor the accelerator. Now, hypothetically the accelerator returns a number between 1 and 1024 depending on how far you depress the gas pedal.


Ok, now remember your ass is on the line. You have to distinguish between a legitimate depress of the accelerator and an accidental one, or even a malfunctioning gas pedal.

Remember, everything is a number.

It would appear that BMW's drive by wire system wants double assurance that you mean to "go". Hence, the programer is reading the numbers generated by the gas pedal and the programmer is not willing to unload the actual throttle until he is sure that his reading of the gas pedal is consistent with someone that wants to speed off instead of someone that meant to hit the brakes but hit the gas pedal instead, or a broken gas pedal.

Again, the vehicle is all about safety first. So this mentality is the reason for your throttle delay.

If BMW won't even turn on the rain sensing wipers automatically without having you first press a button, do you think these guys are going blindly turn your throttle input into an instant revving of the engine? The answer, of course is "no!"

Now, you have to appreciate the European state of mind here. Just think of Airbus vs. Boeing jets.

Airbus will not let you fly outside the flight envelope. Airbus uses joysticks and a long history of building excellent fly by wire aircraft. But, the computer won't let the pilot stall the plane.

Boeing, except the really new stuff that is fly by wire, will let you stall the plane without any intervention. A "stall" for the non airplane enthusiast is when the plane stops flying and turns into a rock and starts to fall out of the sky. A stall does not refer to an engine failure.

So, a BMW behaves a bit more like an Airbus. It double checks your inputs to make sure you mean what you say. The penalty here is latency. Once the BMW is sure you mean to floor the gas then all heck breaks loose so that it can meet its performance specs. It makes up a but for the latency by being a little choppy. But it's the only way to keep those 0 to 60 numbers in spec.

You may ask, "why the latency? I am a good driver." Just like the designer of airbus believes---the car is delivered to the entire world and not everyone is a good driver. So, they design the car to be operated by the lowest common denominator and both BMW and Airbus feels this concept saves lives.

Now, back to our computer program. Are you really going to accelerate the vehicle without being sure the input pattern is consistent with a true depress of the accelerator? Don't you want to wait just a split second to make sure it's not a faulty input from a potentially defective accelerator?

I hope this explains the delay.

Originally Posted by DChaffins

Surely now something will be done.


http://blogs.insideline.com/roadtest...y.html#poll-62

[grimlock]

I think its a 'cultural' issue wrt. the drivetrain code..
I side with the Germans.. I like the way it drives now. Perhaps they should make an NA version to deal with the complaints..

[The Mewster]

I don't have a lot of experience with autos, other than with the VAG DSG/S-tronic double clutch box (which is effectively a manual from a power transfer perspective) but from what I understand about fluid-clutch based automatic gearboxes, you're never going to get immediate direct power transfer due to the inherent lag involved in the fluid-clutch (hence them being the only option on really high torque engines).

I'd have thought if you're into your driving enough to want immediate throttle response (which I have it say I'm not!) then you're far better off with a manual.

But, happy to be proved wrong by someone with more experience of auto boxes than me!

[ibimmer954]

This is definitely a complex behavior that results from the implementation of both the accelerator and transmission. In the U.S., the only tranny option available for the F10 528i is the 8-speed automatic which is a traditional auto tranny. No other option is available for the 528i.

But even if the 6MT were available with a real clutch, the implementation of the accelerator software comes into play. Yes, you have "instant" tranny response because the clutch is a "hard" link. But if the engine computer doesn't crank up the throttle for an extra 300 milliseconds, you are still going to feel a 'delay'.

Dual-clutch autos are not available for the F10 yet. Will probably have to wait for the M5 for that option.

You think it's bad now? Wait until they go to complete steer-by-wire. It's coming eventually. The current complaints about steering "feel" will be nothing compared to the inevitable steer-by-wire system.

[DuneMan]

Originally Posted by BMWrules7

Don't you want to wait just a split second to make sure it's not a faulty input from a potentially defective accelerator?

Of course not!
That split second could be the end of me, if I'm at an intersection or yielding for traffic. The car should behave exactly as I expect it.

If BMW sees a high risk in their throttle (going off on it's own), then stop making cars. But the solution should not be to have another brain inside the car making it's own decision regardless of the driver's commands.


There's a saying: two captains will sink the ship.
Well i say: two brains will wreck the car.

Sorry, but my brain should be the only one making the decision while I'm driving my car.

[raleedy]

That's the zaniest explanation I've heard for a defect from anyone other than Microsoft. Why don't they just pop up a little "Are you sure?" window on the iDrive when you press "Go"? I don't believe it for a moment.

[grimlock]

Yes it is asking too much to have a car behave the way you want it. If you were the only person on the road, maybe, but there are other considerations, such as the safety of others, and even yourself. That's why it is accepted for(+by) society to inter those individuals who may be detrimental to the rest. Because a car is a 2 tonne piece of metal that can move at 300kpm, you don't think there should be any limits imposed on individual liberty in its use? Common!

[Vinfast]

I have instant throttle response on my 550i M-sport. I love it ---- it's not like my 08 550i M-Sport which did have a split second delay or hesitation. I'm glad they fixed it on the 550i.

[BMWrules7]

You make a great point! Plus, the problem gets worse when you start to consider steer by wire. Any latency here could be a problem too!

Again, this is why there is room for tuners to install a chip or other software. Dinan can take more risks in this area and they can engineer their systems to respond to accelerator inputs faster than BMW. It all comes down to liability and who has the deeper pockets!

Just think for a moment... If you had responsibility for designing a fly by wire accelerator pedal, how would you do it?

Would you scan the pedal a few times with an algorithm that ensures the driver means to accelerate or would you just increase the throttle to full speed with no safety checks?

I will say this, however, that BMW does have some secret settings to reduce latency iff: you complain about it.

Does it not make sense to send your most safest algorithms out to the masses but still allow an improved but riskier setting for the one percent of the people that complain?

I would think so.

But, you can be sure that a software update will reset the setting. Also, a prudent engineer might make such a riskier setting last only a month before it resets thinking you won't even notice.

Who knows, but software makes such tactics possible, right?

Just be glad bill gates did not design the algorithms for the accelerator pedal! Otherwise, when you floor the accelerator a message would appear on the idrive before the car would move and say "are you sure?"

But, in all seriousness the "are you sure" message is actually being asked every time you press the accelerator pedal (deep inside the software of course) and this is the reason for that damming latency!

Originally Posted by DuneMan

Of course not!
That split second could be the end of me, if I'm at an intersection or yielding for traffic. The car should behave exactly as I expect it.

If BMW sees a high risk in their throttle (going off on it's own), then stop making cars. But the solution should not be to have another brain inside the car making it's own decision regardless of the driver's commands.


There's a saying: two captains will sink the ship.
Well i say: two brains will wreck the car.

Sorry, but my brain should be the only one making the decision while I'm driving my car.

[BobRBob]

Originally Posted by BMWrules7

The delay you refer to is a safety delay. Pretend for a moment that you are a computer programmer.

Your program monitors events. For example, you monitor the accelerator. Now, hypothetically the accelerator returns a number between 1 and 1024 depending on how far you depress the gas pedal.


Ok, now remember your ass is on the line. You have to distinguish between a legitimate depress of the accelerator and an accidental one, or even a malfunctioning gas pedal.

Remember, everything is a number.

It would appear that BMW's drive by wire system wants double assurance that you mean to "go". Hence, the programer is reading the numbers generated by the gas pedal and the programmer is not willing to unload the actual throttle until he is sure that his reading of the gas pedal is consistent with someone that wants to speed off instead of someone that meant to hit the brakes but hit the gas pedal instead, or a broken gas pedal.

Again, the vehicle is all about safety first. So this mentality is the reason for your throttle delay.

If BMW won't even turn on the rain sensing wipers automatically without having you first press a button, do you think these guys are going blindly turn your throttle input into an instant revving of the engine? The answer, of course is "no!"

Now, you have to appreciate the European state of mind here. Just think of Airbus vs. Boeing jets.

Airbus will not let you fly outside the flight envelope. Airbus uses joysticks and a long history of building excellent fly by wire aircraft. But, the computer won't let the pilot stall the plane.

Boeing, except the really new stuff that is fly by wire, will let you stall the plane without any intervention. A "stall" for the non airplane enthusiast is when the plane stops flying and turns into a rock and starts to fall out of the sky. A stall does not refer to an engine failure.

So, a BMW behaves a bit more like an Airbus. It double checks your inputs to make sure you mean what you say. The penalty here is latency. Once the BMW is sure you mean to floor the gas then all heck breaks loose so that it can meet its performance specs. It makes up a but for the latency by being a little choppy. But it's the only way to keep those 0 to 60 numbers in spec.

You may ask, "why the latency? I am a good driver." Just like the designer of airbus believes---the car is delivered to the entire world and not everyone is a good driver. So, they design the car to be operated by the lowest common denominator and both BMW and Airbus feels this concept saves lives.

Now, back to our computer program. Are you really going to accelerate the vehicle without being sure the input pattern is consistent with a true depress of the accelerator? Don't you want to wait just a split second to make sure it's not a faulty input from a potentially defective accelerator?

I hope this explains the delay.

Are you actually sure that this is how it works or are you just speculating?

[Needsdecaf]

Originally Posted by BMWrules7

The delay you refer to is a safety delay. Pretend for a moment that you are a computer programmer.

Your program monitors events. For example, you monitor the accelerator. Now, hypothetically the accelerator returns a number between 1 and 1024 depending on how far you depress the gas pedal.


Ok, now remember your ass is on the line. You have to distinguish between a legitimate depress of the accelerator and an accidental one, or even a malfunctioning gas pedal.

Remember, everything is a number.

It would appear that BMW's drive by wire system wants double assurance that you mean to "go". Hence, the programer is reading the numbers generated by the gas pedal and the programmer is not willing to unload the actual throttle until he is sure that his reading of the gas pedal is consistent with someone that wants to speed off instead of someone that meant to hit the brakes but hit the gas pedal instead, or a broken gas pedal.

Again, the vehicle is all about safety first. So this mentality is the reason for your throttle delay.

If BMW won't even turn on the rain sensing wipers automatically without having you first press a button, do you think these guys are going blindly turn your throttle input into an instant revving of the engine? The answer, of course is "no!"

Now, you have to appreciate the European state of mind here. Just think of Airbus vs. Boeing jets.

Airbus will not let you fly outside the flight envelope. Airbus uses joysticks and a long history of building excellent fly by wire aircraft. But, the computer won't let the pilot stall the plane.

Boeing, except the really new stuff that is fly by wire, will let you stall the plane without any intervention. A "stall" for the non airplane enthusiast is when the plane stops flying and turns into a rock and starts to fall out of the sky. A stall does not refer to an engine failure.

So, a BMW behaves a bit more like an Airbus. It double checks your inputs to make sure you mean what you say. The penalty here is latency. Once the BMW is sure you mean to floor the gas then all heck breaks loose so that it can meet its performance specs. It makes up a but for the latency by being a little choppy. But it's the only way to keep those 0 to 60 numbers in spec.

You may ask, "why the latency? I am a good driver." Just like the designer of airbus believes---the car is delivered to the entire world and not everyone is a good driver. So, they design the car to be operated by the lowest common denominator and both BMW and Airbus feels this concept saves lives.

Now, back to our computer program. Are you really going to accelerate the vehicle without being sure the input pattern is consistent with a true depress of the accelerator? Don't you want to wait just a split second to make sure it's not a faulty input from a potentially defective accelerator?

I hope this explains the delay.

I am sorry, but this is a total bunch of hooey.

Put the transmission in sport mode. The delay / hesitation is gone COMPLETELY.

And if it was a "BMW" concept, why wouldn't other BMW's have the same thing? The 328 loaner I drove had no such issue.

[BMWrules7]

I wish it were true sir. But, facts are facts.

Go to any legacy linkage based car and rev up the engine. You will feel the car shake the moment you depress the throttle.

Now, go to your fly by wire bimmer. You can rev the engine and have the accelerator pop back to default before the engine responds.

Anyone familiar with a 70s muscle car will feel the difference.

Go to an old suburu. It is not possible to rev the engine and take your foot off the pedal before the engine responds. Regrettably, the same is not true for BMW.

Before you start telling the world my post is a bunch of "hooey," I suggest you do some research or perform some testing.

This is not a hard test gentlemen.

Because we are dealing with computers, this test must be done with the tranny in drive. Nevertheless, I found the safety delay exists in neutral as well.

[Needsdecaf]

Originally Posted by BMWrules7

I wish it were true sir. But, facts are facts.

Go to any legacy linkage based car and rev up the engine. You will feel the car shake the moment you depress the throttle.

Now, go to your fly by wire bimmer. You can rev the engine and have the accelerator pop back to default before the engine responds.

Anyone familiar with a 70s muscle car will feel the difference.

Go to an old suburu. It is not possible to rev the engine and take your foot off the pedal before the engine responds. Regrettably, the same is not true for BMW.

Before you start telling the world my post is a bunch of "hooey," I suggest you do some research or perform some testing.

This is not a hard test gentlemen.

Because we are dealing with computers, this test must be done with the tranny in drive. Nevertheless, I found the safety delay exists in neutral as well.

Seriously, just stop. Or please produce evidence that this parameter is designed into the software. Your "facts" as you say. I say you have no "facts" just your theory. What "research" have you done besides sitting in the car pressing the accelerator?

My 2004 Volvo had a turbo and an automatic transmission, yet does not exhibit any of this behavior. But wait, what's this? It had a drive by wire "throttle". So tell me, why didn't it exhibit this behavior in the least? I just got rid of the car with 105k miles on it, most by me, the rest by my wife, with NONE of this behavior.

And don't tell me to do research or testing. I actually know quite a bit about the system in that car. Did you know that it actually isn't a drive by wire "throttle" in the least? It's actually a "torque demand" switch. Basically, the computer monitors your pedal input and determines, based on the pedal position and the rapidity of which it was pressed, how much torque, as a % of maximum, you are looking for. It then adjusted the throttle, the timing, the fuel, the boost (via the boost control valve) to compensate. I've reviewed constant lines of logged data on that system and understood all the parameters of pedal position, boost, timing, intake air temp, etc. None of this type of behavior existed...and this from the sensible, safety-first company called Volvo.


Finally, as I said again, why does the tip in delay vanish completely on the 535i when the transmission is put in sport mode?

Sorry, but your theory just doesn't hold any water.

[EvolutionTheory]

I'm sorry but until you find a source from BMW describing this as the cause, your entire post should be considered false. I have heard many different reasons for the cause in delay from "starting in 2nd gear" to "Software needs to adjust to you" to "Software needs a patch", but I've never heard or read this explanation. Not only does your explanation sound illogical and un-safe (Are you really sure you need to accelerate??) it's just a plain bad idea to create that option and not inform your dealers or your customers.

Originally Posted by BMWrules7

The delay you refer to is a safety delay. Pretend for a moment that you are a computer programmer.

Your program monitors events. For example, you monitor the accelerator. Now, hypothetically the accelerator returns a number between 1 and 1024 depending on how far you depress the gas pedal.


Ok, now remember your ass is on the line. You have to distinguish between a legitimate depress of the accelerator and an accidental one, or even a malfunctioning gas pedal.

Remember, everything is a number.

It would appear that BMW's drive by wire system wants double assurance that you mean to "go". Hence, the programer is reading the numbers generated by the gas pedal and the programmer is not willing to unload the actual throttle until he is sure that his reading of the gas pedal is consistent with someone that wants to speed off instead of someone that meant to hit the brakes but hit the gas pedal instead, or a broken gas pedal.

Again, the vehicle is all about safety first. So this mentality is the reason for your throttle delay.

If BMW won't even turn on the rain sensing wipers automatically without having you first press a button, do you think these guys are going blindly turn your throttle input into an instant revving of the engine? The answer, of course is "no!"

Now, you have to appreciate the European state of mind here. Just think of Airbus vs. Boeing jets.

Airbus will not let you fly outside the flight envelope. Airbus uses joysticks and a long history of building excellent fly by wire aircraft. But, the computer won't let the pilot stall the plane.

Boeing, except the really new stuff that is fly by wire, will let you stall the plane without any intervention. A "stall" for the non airplane enthusiast is when the plane stops flying and turns into a rock and starts to fall out of the sky. A stall does not refer to an engine failure.

So, a BMW behaves a bit more like an Airbus. It double checks your inputs to make sure you mean what you say. The penalty here is latency. Once the BMW is sure you mean to floor the gas then all heck breaks loose so that it can meet its performance specs. It makes up a but for the latency by being a little choppy. But it's the only way to keep those 0 to 60 numbers in spec.

You may ask, "why the latency? I am a good driver." Just like the designer of airbus believes---the car is delivered to the entire world and not everyone is a good driver. So, they design the car to be operated by the lowest common denominator and both BMW and Airbus feels this concept saves lives.

Now, back to our computer program. Are you really going to accelerate the vehicle without being sure the input pattern is consistent with a true depress of the accelerator? Don't you want to wait just a split second to make sure it's not a faulty input from a potentially defective accelerator?

I hope this explains the delay.

[BMWrules7]

For those of us with quick acting neurons we recognize a delay. For those with less responsive neurons I suggest you use a clock.

The BMW throttle has a considerable delay when compared to legacy linkage vehicles. This delay is related to safety mechanisms in the fly by wire accelerator.

I think your description of the accelerator completely qualifies as a fly by wire system. X increases Y; or Y increases X. What's the difference if it's done by the computer?

Originally Posted by Needsdecaf

Seriously, just stop. Or please produce evidence that this parameter is designed into the software. Your "facts" as you say. I say you have no "facts" just your theory. What "research" have you done besides sitting in the car pressing the accelerator?

My 2004 Volvo had a turbo and an automatic transmission, yet does not exhibit any of this behavior. But wait, what's this? It had a drive by wire "throttle". So tell me, why didn't it exhibit this behavior in the least? I just got rid of the car with 105k miles on it, most by me, the rest by my wife, with NONE of this behavior.

And don't tell me to do research or testing. I actually know quite a bit about the system in that car. Did you know that it actually isn't a drive by wire "throttle" in the least? It's actually a "torque demand" switch. Basically, the computer monitors your pedal input and determines, based on the pedal position and the rapidity of which it was pressed, how much torque, as a % of maximum, you are looking for. It then adjusted the throttle, the timing, the fuel, the boost (via the boost control valve) to compensate. I've reviewed constant lines of logged data on that system and understood all the parameters of pedal position, boost, timing, intake air temp, etc. None of this type of behavior existed...and this from the sensible, safety-first company called Volvo.


Finally, as I said again, why does the tip in delay vanish completely on the 535i when the transmission is put in sport mode?

Sorry, but your theory just doesn't hold any water.

[EvolutionTheory]

I don't think there was a question over whether or not a delay exists.. everyone agrees there's an unexpected delay. The issue is that your explanation does not make practical sense on an automobile, especially when the company its self hasn't explained to the dealers, its service department, or its customers directly the cause as you've written it.

I'd love to know that your explanation is the true cause and BMW made a huge mistake here it could correct in the future, but there's nothing to suggest your explanation is the actual correct explanation beyond something that just seems to make sense to you.

Originally Posted by BMWrules7

For those of us with quick acting neurons we recognize a delay. For those with less responsive neurons I suggest you use a clock.

The BMW throttle has a considerable delay when compared to legacy linkage vehicles. This delay is related to safety mechanisms in the fly by wire accelerator.

I think your description of the accelerator completely qualifies as a fly by wire system. X increases Y; or Y increases X. What's the difference if it's done by the computer?

[grimlock]

I'm going to do a little experimenting to see if I can trigger any 'fat foot' safeguard.. only 310NM torque here so its relatively safe

[mspeasl]

Originally Posted by EvolutionTheory

I don't think there was a question over whether or not a delay exists.. everyone agrees there's an unexpected delay. The issue is that your explanation does not make practical sense on an automobile, especially when the company its self hasn't explained to the dealers, its service department, or its customers directly the cause as you've written it.

You're correct about the dealers not knowing and lack of support from BMW. To them they must fall back on the old cliché, "It's a normal condition"

BS.......!!

[BMWrules7]

Sure. You only believe things that BMW says.

Why would BMW disclose to anyone that their cars have a built in latency on the accelerator as a means to improve safety?

BMW will never admit it. Unless, of course they get sued for a runaway gas pedal accident. If they ever had such a lawsuit and they did not settle then you might find some juicy testimony in the record.

But, you want us all to believe that there is zero latency attributable to safety. BMW would be insane not to have specific algorithms to ensure the driver's inputs on the gas pedal are consistent with the driver's intent.

In this beautiful and most litigious USA, I can say with absolute certainty that any fly by wire throttle system will have logic to prevent accidents from malfunctions and mistakes.

You think BMW is going to avoid putting such safety logic in their cars because they want to please only you?

A faster processor and a better gas pedal ain't going reduce the latency. The latency is based upon human factors and this will always require a 100 ( or whatever) or more millisecond delay. It gives you the chance to change your mind. And, as such it is safer in most cases. Not all cases, but see my post about airbus above for a better understanding.

Just like the brake force system designed to apply full brake pressure when you only tap the brakes at the right time is based on human factors too. For some reason people don't slam the brakes when they should. And this system compensates beautifully doesn't it?

And so does the delay in the throttle. If BMW took away the delay you would have all kinds of people blaming the fly by wire throttle as "too sensitive" and "dangerous."

Most people don't complain about the delay. But BMW is not going to tell anyone about this feature unless they get sued from a related accident.

Finally, just because BMW won't say what I am saying does not make my point false. You don't have to believe me. Nevertheless, lack of BMW's blessing does not somehow invalidate my points that the latency is an intentional safety feature of the throttle system.

Originally Posted by EvolutionTheory

I'm sorry but until you find a source from BMW describing this as the cause, your entire post should be considered false. I have heard many different reasons for the cause in delay from "starting in 2nd gear" to "Software needs to adjust to you" to "Software needs a patch", but I've never heard or read this explanation. Not only does your explanation sound illogical and un-safe (Are you really sure you need to accelerate??) it's just a plain bad idea to create that option and not inform your dealers or your customers.

[DuneMan]

Originally Posted by BMWrules7

Again, this is why there is room for tuners to install a chip or other software. Dinan can take more risks in this area and they can engineer their systems to respond to accelerator inputs faster than BMW. It all comes down to liability and who has the deeper pockets!

But that's my point... the risk is in the status quo. There is much much greater daily risk that can come from throttle delay/hesitation, then from a runaway throttle.


Auto makers thrive on consumer appeal not on lack of law suites.
If what you say is true, then BMWs are being designed by a bunch of lawyers. Not Engineers!

in short, I don't buy your argument at all... sorry.

[delvek]

Originally Posted by EvolutionTheory

I don't think there was a question over whether or not a delay exists.. everyone agrees there's an unexpected delay. The issue is that your explanation does not make practical sense on an automobile, especially when the company its self hasn't explained to the dealers, its service department, or its customers directly the cause as you've written it.

I'd love to know that your explanation is the true cause and BMW made a huge mistake here it could correct in the future, but there's nothing to suggest your explanation is the actual correct explanation beyond something that just seems to make sense to you.

I have owned 2 F10's neither display(ed) the situation you discuss.

My car tiptoes through parking lots and hauls ass down autobahn's the momment I push on the accelerator with no indecision, it only asks my foot to instruct.

May I suggest ankle/calf workout program?