New Toyota Recall
- Thread starter Boomer
- Start date
- Status
Originally Posted By: ls1mike
Quote:
ShiningArcanine
Honestly, the sudden acceleration issues are most likely the result of the drive by wire transition Toyota is making at the moment. Having drive by wire meant that physical constraints that a mechanically linked system placed on the design of things were lifted, so engineers were free to do things that they could not do previously. As a consequence, they learned how not to do things through the experience of being the first ones there (as far as mass production cars are concerned).
The fact that they were first mattered here...
but here...
Quote:
ShiningArcanine: I was under the impression that they adopted first, but who did it first is irrelevant. From a statistical standpoint, they are not far from where they were before
But since they are not first it does not matter?
Which is it?
I don't think GM shared their DBW knowledge with Toyota, so sort of reverse engineering for all intents and purposes, they were stepping into unknown territory. Does that clear up the apparent contradiction satisfactorily for you?
Quote:
ShiningArcanine
Honestly, the sudden acceleration issues are most likely the result of the drive by wire transition Toyota is making at the moment. Having drive by wire meant that physical constraints that a mechanically linked system placed on the design of things were lifted, so engineers were free to do things that they could not do previously. As a consequence, they learned how not to do things through the experience of being the first ones there (as far as mass production cars are concerned).
The fact that they were first mattered here...
but here...
Quote:
ShiningArcanine: I was under the impression that they adopted first, but who did it first is irrelevant. From a statistical standpoint, they are not far from where they were before
But since they are not first it does not matter?
Which is it?
I don't think GM shared their DBW knowledge with Toyota, so sort of reverse engineering for all intents and purposes, they were stepping into unknown territory. Does that clear up the apparent contradiction satisfactorily for you?
Originally Posted By: Norse
Originally Posted By: ShiningArcanine
Originally Posted By: ls1mike
Originally Posted By: ShiningArcanine
Honestly, the sudden acceleration issues are most likely the result of the drive by wire transition Toyota is making at the moment. Having drive by wire meant that physical constraints that a mechanically linked system placed on the design of things were lifted, so engineers were free to do things that they could not do previously. As a consequence, they learned how not to do things through the experience of being the first ones there (as far as mass production cars are concerned).
What?
Just about everything GM makes now and since about 02 has been been Drive by wire. My 04 Silverado was, GTO's are, newer W-bodies, My Equinox is, My 03 CTS was. The cars involved in the recall were not the first mass produced ones there. The cars involved are all newer then when GM stared using it. It was first used in 1997 by the C-5 Corvette.
Man you really love Toyota, because the current system they have now, was not the first one there. Dosen't matter if it is there DBW or not, they made a mistake plain and simple.
I was under the impression that they adopted first, but who did it first is irrelevant. From a statistical standpoint, they are not far from where they were before.
Toy boys make me laugh! If it's so irrelevant who was first to DBW, why did you try to make it appear that Toy was so cutting edge by stating they were?
I am sure that it was cutting edge as far as the engineers working on it were concerned.
Speaking from actual experience, if you wanted to do something like molecular dynamics simulations on computational clusters, the fact that it has been done before does not automatically make it any easier for you, because the people who have done it are not there holding your hand showing you where the various pitfalls are. I was doing undergraduate research in molecular dynamics simulations in a lab that gave no support to undergraduate students and things did not go well. I am sure that it is the same situation for all engineers working on all problems that are not textbook cases.
I doubt GM was willing to have their engineers that implemented drive by wire show Toyota's engineers the ropes when they did it.
Does that satisfy your nit-picking?
Originally Posted By: ShiningArcanine
Originally Posted By: ls1mike
Originally Posted By: ShiningArcanine
Honestly, the sudden acceleration issues are most likely the result of the drive by wire transition Toyota is making at the moment. Having drive by wire meant that physical constraints that a mechanically linked system placed on the design of things were lifted, so engineers were free to do things that they could not do previously. As a consequence, they learned how not to do things through the experience of being the first ones there (as far as mass production cars are concerned).
What?
Man you really love Toyota, because the current system they have now, was not the first one there. Dosen't matter if it is there DBW or not, they made a mistake plain and simple.
I was under the impression that they adopted first, but who did it first is irrelevant. From a statistical standpoint, they are not far from where they were before.
Toy boys make me laugh! If it's so irrelevant who was first to DBW, why did you try to make it appear that Toy was so cutting edge by stating they were?
I am sure that it was cutting edge as far as the engineers working on it were concerned.
Speaking from actual experience, if you wanted to do something like molecular dynamics simulations on computational clusters, the fact that it has been done before does not automatically make it any easier for you, because the people who have done it are not there holding your hand showing you where the various pitfalls are. I was doing undergraduate research in molecular dynamics simulations in a lab that gave no support to undergraduate students and things did not go well. I am sure that it is the same situation for all engineers working on all problems that are not textbook cases.
I doubt GM was willing to have their engineers that implemented drive by wire show Toyota's engineers the ropes when they did it.
Does that satisfy your nit-picking?
OVERKILL
$100 Site Donor 2021
Originally Posted By: ShiningArcanine
I will definitely agree that it is easy for little bugs to exist in software that cause major problems, but I am not sure if having the brake override is a good idea.
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
Quote:
The operation of the brakes and engine have been separate for decades. In driver's ed, they tell you that when you are stopped on a hill, you should hold down the brake with your left foot and then press the accelerator with your right so you do not roll backwards when you release the brake.
Not in the drivers ed course I took.... Try that with a standard and see how the clutch pedal fits into the equation.
Quote:
Simply programming the car to ignore the accelerator when the brake is engaged would prevent that from working.
That's not how it works. If it detects significant throttle position and the brakes happening at the same time, it cuts power. So if your car decides to stick at WOT, and of course you lack any mechanical connection to the pedal, so you can't simply pry it up with your foot to bring it OUT of WOT, the ECM cuts power; like a rev limiter. It doesn't shut the car OFF.
Quote:
There is room to have a mph threshold in the code so it works properly, but that is not in the specification which people are saying should be given to Toyota's programmers and I do not have faith that a typical programmer would realize the issue before implementing it. If the change is made, I would not be surprised to hear about an increase in the incident rate of Toyota vehicles colliding with cars because they rolled backwards on hills.
People are saying Toyota's implementation should be the same as those used by other manufacturers, which aren't having cars roll down hills by the way.....
Quote:
I think adding that as a feature could mask the extent of the underlying problem.
Which is that Toyota has some screwed-up programming bug that they can't find, so they have been waging floor mat and now gas pedal campaigns to redirect the attention while they search for the real problem.
Quote:
in summary, I do not think modifying the firmware for this is the best idea in terms of design. As far as damage control, now that cars with such an issue are on the market, it is probably the best thing Toyota could do, but I do not think that vehicles should come that way from the factory. I find it difficult to put into words, but my impression of such a feature (cutting off the engine when the brake is applied) is that it will negatively affect a vehicles' predictability, particularly because hiccups where acceleration does not occur in a predictable fashion will go unreported, leading to decreased predictability as time goes on.
Well, you see, the companies that ACTUALLY pioneered this technology (read: NOT Toyota, BMW has been using it since 1988) have managed to NOT kill people with their implementation and still implement reasonable fail-safes to deal with things like stuck throttle blades and the like.
I know it is hard to fathom the Japanese; especially THE Japanese company: Toyota, learning something from its competitors, but in this case, since everybody BUT Toyota has been doing it right, and Toyota is about a decade late on being the "Pioneer" of the technology (which throws that argument far, FAR out the window), maybe opening their eyes and getting their noses out of the air might actually result in them learning something about a system they apparently don't know how to implement correctly.
I will definitely agree that it is easy for little bugs to exist in software that cause major problems, but I am not sure if having the brake override is a good idea.
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
Quote:
The operation of the brakes and engine have been separate for decades. In driver's ed, they tell you that when you are stopped on a hill, you should hold down the brake with your left foot and then press the accelerator with your right so you do not roll backwards when you release the brake.
Not in the drivers ed course I took.... Try that with a standard and see how the clutch pedal fits into the equation.
Quote:
Simply programming the car to ignore the accelerator when the brake is engaged would prevent that from working.
That's not how it works. If it detects significant throttle position and the brakes happening at the same time, it cuts power. So if your car decides to stick at WOT, and of course you lack any mechanical connection to the pedal, so you can't simply pry it up with your foot to bring it OUT of WOT, the ECM cuts power; like a rev limiter. It doesn't shut the car OFF.
Quote:
There is room to have a mph threshold in the code so it works properly, but that is not in the specification which people are saying should be given to Toyota's programmers and I do not have faith that a typical programmer would realize the issue before implementing it. If the change is made, I would not be surprised to hear about an increase in the incident rate of Toyota vehicles colliding with cars because they rolled backwards on hills.
People are saying Toyota's implementation should be the same as those used by other manufacturers, which aren't having cars roll down hills by the way.....
Quote:
I think adding that as a feature could mask the extent of the underlying problem.
Which is that Toyota has some screwed-up programming bug that they can't find, so they have been waging floor mat and now gas pedal campaigns to redirect the attention while they search for the real problem.
Quote:
in summary, I do not think modifying the firmware for this is the best idea in terms of design. As far as damage control, now that cars with such an issue are on the market, it is probably the best thing Toyota could do, but I do not think that vehicles should come that way from the factory. I find it difficult to put into words, but my impression of such a feature (cutting off the engine when the brake is applied) is that it will negatively affect a vehicles' predictability, particularly because hiccups where acceleration does not occur in a predictable fashion will go unreported, leading to decreased predictability as time goes on.
Well, you see, the companies that ACTUALLY pioneered this technology (read: NOT Toyota, BMW has been using it since 1988) have managed to NOT kill people with their implementation and still implement reasonable fail-safes to deal with things like stuck throttle blades and the like.
I know it is hard to fathom the Japanese; especially THE Japanese company: Toyota, learning something from its competitors, but in this case, since everybody BUT Toyota has been doing it right, and Toyota is about a decade late on being the "Pioneer" of the technology (which throws that argument far, FAR out the window), maybe opening their eyes and getting their noses out of the air might actually result in them learning something about a system they apparently don't know how to implement correctly.
Originally Posted By: OVERK1LL
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
OVERKILL
$100 Site Donor 2021
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
So YOUR opinion is that it is OK to have Kamikaze cars so as long as Toyota isn't borrowing engineering experience from its competitors that aren't killing their clientèle?
Originally Posted By: OVERK1LL
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
So YOUR opinion is that it is OK to have Kamikaze cars so as long as Toyota isn't borrowing engineering experience from its competitors that aren't killing their clientèle?
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
So YOUR opinion is that it is OK to have Kamikaze cars so as long as Toyota isn't borrowing engineering experience from its competitors that aren't killing their clientèle?
It is my opinion that reliance on such a mechanism to avoid dealing with the actual issue is poor engineering. It is also my opinion that such a thing would make the shareholders happy, as it would allow the company to avoid dealing with the actual issue.
I do not care whether or not Toyota does this, but I think it would be a poor design decision. Being a computer science student, my classes are divided among two schools of thought, science and engineering. The science classes deal with mathematics while the engineering classes deal with actually making stuff like software. As far as what I have learned from the engineering side of computer science goes, putting a patch over an issue without actually solving it is a poor method of doing things in that it causes problems down the road and I see this as being an analogous situation.
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
So YOUR opinion is that it is OK to have Kamikaze cars so as long as Toyota isn't borrowing engineering experience from its competitors that aren't killing their clientèle?
It is my opinion that reliance on such a mechanism to avoid dealing with the actual issue is poor engineering. It is also my opinion that such a thing would make the shareholders happy, as it would allow the company to avoid dealing with the actual issue.
I do not care whether or not Toyota does this, but I think it would be a poor design decision. Being a computer science student, my classes are divided among two schools of thought, science and engineering. The science classes deal with mathematics while the engineering classes deal with actually making stuff like software. As far as what I have learned from the engineering side of computer science goes, putting a patch over an issue without actually solving it is a poor method of doing things in that it causes problems down the road and I see this as being an analogous situation.
OVERKILL
$100 Site Donor 2021
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
So YOUR opinion is that it is OK to have Kamikaze cars so as long as Toyota isn't borrowing engineering experience from its competitors that aren't killing their clientèle?
It is my opinion that it is reliance on such a mechanism is poor engineering.
It is not a reliance, you seem to be missing that part. It is a SAFEGUARD in case of a failure. This could be a mechanical failure like a stuck throttle blade, or a software failure, such as the ECM randomly commanding WOT (which is speculated to be Toyota's issue).
In either instance, the event would be picked up by the ECM and throw a code, so there would be evidence of the occurrence, without that nasty death thing associated with not having it there.
It is not being proposed as a SOLUTION to the problem. It is being proposed as a SAFEGUARD to prevent the problem from KILLING PEOPLE.
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Well, it was apparently a good idea for all the other manufacturers who use it and don't have vehicles randomly killing people or going out of control....
That is a matter of opinion. There is no right or wrong opinion when it comes to that modification.
So YOUR opinion is that it is OK to have Kamikaze cars so as long as Toyota isn't borrowing engineering experience from its competitors that aren't killing their clientèle?
It is my opinion that it is reliance on such a mechanism is poor engineering.
It is not a reliance, you seem to be missing that part. It is a SAFEGUARD in case of a failure. This could be a mechanical failure like a stuck throttle blade, or a software failure, such as the ECM randomly commanding WOT (which is speculated to be Toyota's issue).
In either instance, the event would be picked up by the ECM and throw a code, so there would be evidence of the occurrence, without that nasty death thing associated with not having it there.
It is not being proposed as a SOLUTION to the problem. It is being proposed as a SAFEGUARD to prevent the problem from KILLING PEOPLE.
If the ECM malfunctions, this feature would likely not function either. You need to have a completely redundant system implement it for it to actually be worth doing, otherwise, it would not even qualify as a good workaround.
OVERKILL
$100 Site Donor 2021
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
And it is not like it is unreasonable to expect the ECM to run multiple systems. One that commands control, while another that can interrupt that control if an issue is detected.
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
And it is not like it is unreasonable to expect the ECM to run multiple systems. One that commands control, while another that can interrupt that control if an issue is detected.
Last edited by a moderator:
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
Wow, that was a fast reply. I thought of mentioning that right after my post and quickly edited it to add it.
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
Wow, that was a fast reply. I thought of mentioning that right after my post and quickly edited it to add it.
OVERKILL
$100 Site Donor 2021
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
Wow, that was a fast reply. I thought of mentioning that right after my post and quickly edited to add it.
I see that. LOL.
We use redundant systems all the time in the IT industry, as I would assume, given your position, you would be familiar with.
When it comes to safety; and most of the time it is in regards to the safety of data, there is no such thing as overkill.
Redundant routers, redundant hard drives, servers, all with fall overs and fail safes. Think of something as simple as flood protection on a LAN, or the ability for a router to poison routes when another router goes out of control/into a loop.
We deal with redundancy and "what if scenarios" all the time. I would expect that every possible scenario due to component failure would be explored by the engineers working on this system. And if this is an OVERSIGHT by those engineers, then simply shame on them for not looking to see if there was something they missed that their competitors didn't.
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
Wow, that was a fast reply. I thought of mentioning that right after my post and quickly edited to add it.
I see that. LOL.
We use redundant systems all the time in the IT industry, as I would assume, given your position, you would be familiar with.
When it comes to safety; and most of the time it is in regards to the safety of data, there is no such thing as overkill.
Redundant routers, redundant hard drives, servers, all with fall overs and fail safes. Think of something as simple as flood protection on a LAN, or the ability for a router to poison routes when another router goes out of control/into a loop.
We deal with redundancy and "what if scenarios" all the time. I would expect that every possible scenario due to component failure would be explored by the engineers working on this system. And if this is an OVERSIGHT by those engineers, then simply shame on them for not looking to see if there was something they missed that their competitors didn't.
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
And it is not like it is unreasonable to expect the ECM to run multiple systems. One that commands control, while another that can interrupt that control if an issue is detected.
If the ECM's OS kernel were to crash, all systems it controls would be affected, not just one. Addressing this issue with changes in software as a preventative measure is not enough. You need redundant systems. I believe that the F-22 has triple redundant systems where each system has completely different software running for this exact reason.
There are ways of improving a computer system to make it more reliable so it can handle mission critical tasks, but the change that has been suggested is more of a hack than anything else.
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
And it is not like it is unreasonable to expect the ECM to run multiple systems. One that commands control, while another that can interrupt that control if an issue is detected.
If the ECM's OS kernel were to crash, all systems it controls would be affected, not just one. Addressing this issue with changes in software as a preventative measure is not enough. You need redundant systems. I believe that the F-22 has triple redundant systems where each system has completely different software running for this exact reason.
There are ways of improving a computer system to make it more reliable so it can handle mission critical tasks, but the change that has been suggested is more of a hack than anything else.
OVERKILL
$100 Site Donor 2021
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
And it is not like it is unreasonable to expect the ECM to run multiple systems. One that commands control, while another that can interrupt that control if an issue is detected.
If the ECM's OS kernel were to crash, all systems it controls would be affected, not just one. Addressing this issue with changes in software as a preventative measure is not enough. You need redundant systems. I believe that the F-22 has triple redundant systems where each system has completely different software running for this exact reason.
There are ways of improving a computer system to make it more reliable so it can handle mission critical tasks, but the change that has been suggested is more of a hack than anything else.
But if the hack can save lives while the root of the problem is being found, I think it is perfectly reasonable. As I said, other manufacturers have a system (as I described in my other post) that safeguards in the event of a discrepancy between commanded throttle and actual throttle, as well as the reduction in power output if significant acceleration and braking are being commanded at the same time.
Obviously a PROPER implementation of those safeguards will require significant time. But a hack that is at least functional and can save lives is better than nothing at this point.
If the OS kernel on the ECM crashes, I think we have bigger issues
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
And it is not like it is unreasonable to expect the ECM to run multiple systems. One that commands control, while another that can interrupt that control if an issue is detected.
If the ECM's OS kernel were to crash, all systems it controls would be affected, not just one. Addressing this issue with changes in software as a preventative measure is not enough. You need redundant systems. I believe that the F-22 has triple redundant systems where each system has completely different software running for this exact reason.
There are ways of improving a computer system to make it more reliable so it can handle mission critical tasks, but the change that has been suggested is more of a hack than anything else.
But if the hack can save lives while the root of the problem is being found, I think it is perfectly reasonable. As I said, other manufacturers have a system (as I described in my other post) that safeguards in the event of a discrepancy between commanded throttle and actual throttle, as well as the reduction in power output if significant acceleration and braking are being commanded at the same time.
Obviously a PROPER implementation of those safeguards will require significant time. But a hack that is at least functional and can save lives is better than nothing at this point.
If the OS kernel on the ECM crashes, I think we have bigger issues
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
Wow, that was a fast reply. I thought of mentioning that right after my post and quickly edited to add it.
I see that. LOL.
We use redundant systems all the time in the IT industry, as I would assume, given your position, you would be familiar with.
When it comes to safety; and most of the time it is in regards to the safety of data, there is no such thing as overkill.
Redundant routers, redundant hard drives, servers, all with fall overs and fail safes. Think of something as simple as flood protection on a LAN, or the ability for a router to poison routes when another router goes out of control/into a loop.
We deal with redundancy and "what if scenarios" all the time. I would expect that every possible scenario due to component failure would be explored by the engineers working on this system. And if this is an OVERSIGHT by those engineers, then simply shame on them for not looking to see if there was something they missed that their competitors didn't.
I think given our backgrounds, we see eye to eye on these sorts of design considerations quite a bit.
Anyway, I read an article a while back about how Toyota was trying to cut costs by integrating all of the on-board computer systems in their cars into a single system. I wonder if these issues could be the result of that. Perhaps the engineers did not use a realtime operating system kernel needed in such applications. Maybe they have some sort of race condition in their code that occurred as a programmer's mistake when the systems were being integrated. Come to think of it, the issue could be as simple as not having used ECC memory. The transmission issues from 3 years ago satisfactorily demonstrated to me that Toyota's computer expertise is a weakness of theirs, so it would not surprise me if it got out that they did not use ECC memory in their cars.
Originally Posted By: ShiningArcanine
Originally Posted By: OVERK1LL
Originally Posted By: ShiningArcanine
If the ECM malfunctions, this feature would not function either.
Depends on the nature of the malfunction I suppose. Ideally, you'd want it as a separate system.
Wow, that was a fast reply. I thought of mentioning that right after my post and quickly edited to add it.
I see that. LOL.
We use redundant systems all the time in the IT industry, as I would assume, given your position, you would be familiar with.
When it comes to safety; and most of the time it is in regards to the safety of data, there is no such thing as overkill.
Redundant routers, redundant hard drives, servers, all with fall overs and fail safes. Think of something as simple as flood protection on a LAN, or the ability for a router to poison routes when another router goes out of control/into a loop.
We deal with redundancy and "what if scenarios" all the time. I would expect that every possible scenario due to component failure would be explored by the engineers working on this system. And if this is an OVERSIGHT by those engineers, then simply shame on them for not looking to see if there was something they missed that their competitors didn't.
I think given our backgrounds, we see eye to eye on these sorts of design considerations quite a bit.
Anyway, I read an article a while back about how Toyota was trying to cut costs by integrating all of the on-board computer systems in their cars into a single system. I wonder if these issues could be the result of that. Perhaps the engineers did not use a realtime operating system kernel needed in such applications. Maybe they have some sort of race condition in their code that occurred as a programmer's mistake when the systems were being integrated. Come to think of it, the issue could be as simple as not having used ECC memory. The transmission issues from 3 years ago satisfactorily demonstrated to me that Toyota's computer expertise is a weakness of theirs, so it would not surprise me if it got out that they did not use ECC memory in their cars.
OVERKILL
$100 Site Donor 2021
Originally Posted By: ShiningArcanine
I think given our backgrounds, we see eye to eye on these sorts of design considerations quite a bit.
Anyway, I read an article a while back about how Toyota was trying to cut costs by integrating all of the on-board computer systems in their cars into a single system. I wonder if these issues could be the result of that. Perhaps the engineers did not use a realtime operating system kernel needed in such applications. Maybe they have some sort of race condition in their code that occurred as a programmer's mistake when the systems were being integrated. Come to think of it, the issue could be as simple as not having used ECC memory. The transmission issues from 3 years ago satisfactorily demonstrated to me that Toyota's computer expertise is a weakness of theirs, so it would not surprise me if it got out that they did not use ECC memory in their cars.
And it very well could be something that simple!! Something so basic, yet with such dire consequences.
I think given our backgrounds, we see eye to eye on these sorts of design considerations quite a bit.
Anyway, I read an article a while back about how Toyota was trying to cut costs by integrating all of the on-board computer systems in their cars into a single system. I wonder if these issues could be the result of that. Perhaps the engineers did not use a realtime operating system kernel needed in such applications. Maybe they have some sort of race condition in their code that occurred as a programmer's mistake when the systems were being integrated. Come to think of it, the issue could be as simple as not having used ECC memory. The transmission issues from 3 years ago satisfactorily demonstrated to me that Toyota's computer expertise is a weakness of theirs, so it would not surprise me if it got out that they did not use ECC memory in their cars.
And it very well could be something that simple!! Something so basic, yet with such dire consequences.
Hmmm, you mean it wasn't sum dum 'merican worker somewhere? Who'da thunk it?
OVERKILL
$100 Site Donor 2021
Originally Posted By: Norse
Originally Posted By: OVERK1LL
Kamikaze cars
I like that! Do you mind if I borrow that?
Go right ahead
Originally Posted By: OVERK1LL
Kamikaze cars
I like that! Do you mind if I borrow that?
Go right ahead
- Status
Similar threads
