Volkswagen cheats EPA Tests: Millions of Vehicles affected

Seen something interesting in the news or on the intertubes? Discuss it here.

Moderators: Zamfir, Hawknc, Moderators General, Prelates

KnightExemplar
Posts: 5494
Joined: Sun Dec 26, 2010 1:58 pm UTC

Volkswagen cheats EPA Tests: Millions of Vehicles affected

Postby KnightExemplar » Tue Sep 22, 2015 8:10 pm UTC

This is all over the news right now. I'll edit some news article into this post eventually.

In any case... thoughts? Volkswagen's stock is down like 30%, and rumors are that the company is reserving $7 Billion, maybe more, to deal with this issue. The hammer hasn't come down on Volkswagen yet, but this looks extremely serious.

http://qz.com/507767/volkswagen-admits- ... worldwide/
Last week, the US Environmental Protection Agency said that nearly 500,000 Volkswagen cars sold in the country between 2009 and 2015 had “defeat device” software installed, which is designed to detect when a car is being tested for emissions and lower them accordingly. Meanwhile, out on the road, these cars released up to 40 times more pollution than allowed by the rules.
First Strike +1/+1 and Indestructible.

User avatar
Dauric
Posts: 3995
Joined: Wed Aug 05, 2009 6:58 pm UTC
Location: In midair, traversing laterally over a container of sharks. No water, just sharks, with lasers.

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Dauric » Tue Sep 22, 2015 8:23 pm UTC

Linkage from NPR

VW had software installed on their 4-cylinder diesel engines that detected when it was connected to an emissions test machine and altered the performance of the engine to meet emissions standards, but when disconnected the engine operated at 'normal' performance levels that emitted 40 times more pollutants than are allowed by EPA standards.

VW is also being investigated by South Korea over this issue.
We're in the traffic-chopper over the XKCD boards where there's been a thread-derailment. A Liquified Godwin spill has evacuated threads in a fourty-post radius of the accident, Lolcats and TVTropes have broken free of their containers. It is believed that the Point has perished.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Tyndmyr » Tue Sep 22, 2015 9:09 pm UTC

Woohoo! Time to buy a VW.

User avatar
SDK
Posts: 703
Joined: Thu May 22, 2014 7:40 pm UTC
Location: Canada

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby SDK » Tue Sep 22, 2015 9:19 pm UTC

My 2012 VW Golf makes me sad today...

At least it still gets good mileage. :roll:
The biggest number (63 quintillion googols in debt)

KnightExemplar
Posts: 5494
Joined: Sun Dec 26, 2010 1:58 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby KnightExemplar » Tue Sep 22, 2015 9:22 pm UTC

SDK wrote:My 2012 VW Golf makes me sad today...

At least it still gets good mileage. :roll:


Well, this only affects the "Clean Diesel" models. Ironic naming scheme, isn't it?

IIRC, some Diesel models use urea, which may (or may not) be affected by this particular scandal. Details are still coming in of course...

EDIT: Here's a quote:

http://www.businesswire.com/news/home/2 ... gHIIpcjx_A
According to the EPA, Volkswagen installed its deceptive software, called a “defeat device,” in at least the following diesel models of its vehicles: Jetta (model years 2009 – 2015), Beetle (model years 2009 – 2015), Audi A3 (model years 2009 – 2015), Golf (model years 2009 – 2015) and Passat (model years 2014 – 2015).


Uh oh... well... you might want to get some of that class-action lawsuit money coming in.

Tyndmyr wrote:Woohoo! Time to buy a VW.


All affected cars are off the market as of now. You might manage to buy it up from SDK over here though :-p
First Strike +1/+1 and Indestructible.

User avatar
SDK
Posts: 703
Joined: Thu May 22, 2014 7:40 pm UTC
Location: Canada

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby SDK » Tue Sep 22, 2015 9:35 pm UTC

"Previously enjoyed 2012 VW Golf Wagon TDI. AS SEEN ON TV!"

It definitely does affect my model - 4 cylinder TDI engine. I plan to call the dealership later this week, see if they want to give me some free stuff. 8-)
The biggest number (63 quintillion googols in debt)

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Tyndmyr » Tue Sep 22, 2015 9:37 pm UTC

KnightExemplar wrote:
Tyndmyr wrote:Woohoo! Time to buy a VW.


All affected cars are off the market as of now. You might manage to buy it up from SDK over here though :-p


Tempting. Only if it's actually defective, mind you.

User avatar
Paul in Saudi
Posts: 262
Joined: Fri Sep 07, 2012 12:52 pm UTC
Location: Dammam, Saudi Arabia

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Paul in Saudi » Tue Sep 22, 2015 11:21 pm UTC

It will be very, very hard to put a good spin on this. This was not a mistake, it was criminal misconduct. Further, even if they cop a deal with the US, the EU and other markets may very be able to hammer VW.

commodorejohn
Posts: 1197
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby commodorejohn » Tue Sep 22, 2015 11:35 pm UTC

Meanwhile, my 1969 Beetle continues to be exempt from California emissions standards...
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

User avatar
sardia
Posts: 6813
Joined: Sat Apr 03, 2010 3:39 am UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby sardia » Wed Sep 23, 2015 12:55 am UTC

SDK wrote:"Previously enjoyed 2012 VW Golf Wagon TDI. AS SEEN ON TV!"

It definitely does affect my model - 4 cylinder TDI engine. I plan to call the dealership later this week, see if they want to give me some free stuff. 8-)

No you don't. What's gonna happen is they'll fix the "defect" and then you'll have a crappy normally performing diesel engine. The whole point of the emissions cheating software isn't just to spew pollution. There's performance and mileage gains to be had.

Derek
Posts: 2181
Joined: Wed Aug 18, 2010 4:15 am UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Derek » Wed Sep 23, 2015 2:41 am UTC

Paul in Saudi wrote:It will be very, very hard to put a good spin on this. This was not a mistake, it was criminal misconduct. Further, even if they cop a deal with the US, the EU and other markets may very be able to hammer VW.

Good spin? Well, the engines had better mileage in dirty mode, that's the best you can do. It's hard to put good spin on cheating.

commodorejohn
Posts: 1197
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby commodorejohn » Wed Sep 23, 2015 3:06 am UTC

I dunno, a lot of people care way more about what kind of mileage their car gets than what it may or may not be emitting.

That said, the people actually in charge of enforcing emissions controls are probably not among them.

In any case, I foresee a spike in the resale value of affected vehicles in the near future.
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

BattleMoose
Posts: 1993
Joined: Tue Nov 13, 2007 8:42 am UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby BattleMoose » Wed Sep 23, 2015 5:24 am UTC

What is shocking is how brazen this misconduct is. I always knew these large corporations couldn't be trusted but to this degree of abuse? The bigger issue for me here is not so much the extra nitrous oxides but the complete failure in trust but such corporations and the public. And I wouldn't assume this is particular to Volkswagen, perhaps they are the only ones to get caught so far.

Clean coal? Can we trust it? Or will it just be more abuse. Even if it is a legitimate technology, there is just no way, not after this. (And I haven't forgotten deepwater horizon either)

User avatar
Thesh
Made to Fuck Dinosaurs
Posts: 6598
Joined: Tue Jan 12, 2010 1:55 am UTC
Location: Colorado

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Thesh » Wed Sep 23, 2015 5:25 am UTC

commodorejohn wrote:In any case, I foresee a spike in the resale value of affected vehicles in the near future.


How so? You'll probably fail emissions tests if you don't get the update.
Summum ius, summa iniuria.

commodorejohn
Posts: 1197
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby commodorejohn » Wed Sep 23, 2015 5:39 am UTC

Only if you live in a place which enforces them and requires regular testing. Which, in the US, is pretty much just California, unless I'm mistaken. Residents of other states should theoretically be able to enjoy improved performance and mileage at no penalty.
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

User avatar
Thesh
Made to Fuck Dinosaurs
Posts: 6598
Joined: Tue Jan 12, 2010 1:55 am UTC
Location: Colorado

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Thesh » Wed Sep 23, 2015 5:46 am UTC

I assumed it was everywhere, but it's not just California:

https://en.wikipedia.org/wiki/Vehicle_i ... ted_States
Summum ius, summa iniuria.

Derek
Posts: 2181
Joined: Wed Aug 18, 2010 4:15 am UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Derek » Wed Sep 23, 2015 5:47 am UTC

commodorejohn wrote:Only if you live in a place which enforces them and requires regular testing. Which, in the US, is pretty much just California, unless I'm mistaken. Residents of other states should theoretically be able to enjoy improved performance and mileage at no penalty.

Pennsylvania has mandatory yearly emissions tests.

commodorejohn
Posts: 1197
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby commodorejohn » Wed Sep 23, 2015 6:26 am UTC

Hmm, it's more common than I thought, I guess. Still, a fairly healthy chunk of the country has no emissions testing whatsoever.
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

User avatar
Diadem
Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Diadem » Wed Sep 23, 2015 8:21 am UTC

All car manufacturers cheat with emission and mileage tests. This is well known. Cars often get only like half of the stated mileage. Of course most of those cheats are clever legal loopholes, and thus not actually illegal. But it wouldn't surprise me at all if more manufacturers have done illegal stuff too.

It's obviously criminal, and they deserve a huge fine for the environmental damage this has caused. That being said, I don't quite see how this can be interpreted as consumer fraud. Selling someone a car that performs better than promised is not fraud.

I drive a 2014 Volkswagen Golf diesel, and it's an awesome car. Before buying this one, I've tested most cars in its price range*. The Golf has the best performance, and by far the best handling. It also scores very high on luxury and spaciousness. All around an awesome car. The car is very well designed, with an amazing eye for detail. But if they cheated over here in Europe I well, then I guess good design wasn't the only reason for its great performance.

* Prices for lease cars are weird in The Netherlands. Because the employer pays for the operational cost, most of the cost left for the employee is taxes, and these taxes are mostly based on carbon emissions. My company mandated that I'd get a diesel, and I wanted one that fell in the lowest tax bracket. Ignoring cars too small to fit me comfortable, I was left with a dozen or so choices.
Last edited by Diadem on Wed Sep 23, 2015 8:53 am UTC, edited 1 time in total.
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister

quantropy
Posts: 195
Joined: Tue Apr 01, 2008 6:55 pm UTC
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby quantropy » Wed Sep 23, 2015 8:28 am UTC

I cna't help wondering how many people knew about this. Upper management are now ordering investigations into the affair, implying they weren't totally in the loop beforehand. Maybe a few techies implemented the idea, and everyone else believed the wonderful figures they obtained. Do VW have independent audits of their software?

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7602
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Zamfir » Wed Sep 23, 2015 10:52 am UTC

There's two different questions there: how many poeple knew that his particular trick was used, and how many people knew that the emissions in real life were far higher than in the formal test?

The answer to the first could be relatively few people, in theory. Though hardly 'a few techies', for a motor management system used on a variety of different models.

But not to the second. Small diesel engines are a core technology for the company, and emissions regulations are perhaps the single largest hurdle for them. The NOx-performance tradeoff is basic knowledge. VW might well have spent billions over the years just pushing that envelope, and you can bet that they know exactly what they can and cannot achieve.

Lots of people must have known that this engine had a particularly large difference between the formal tests and real life results. Quite possibly up the very highest levels - VW built a big US factory anticipating US expansion, and a part of that strategy was to meet strict US regulations at a lower price point than the competition who needed urea systems.

So, perhaps the highest bosses didn't literally order this twerk. But they must have ordered an engine that would meet emissions test without regard for real life emissions, and pointedly decided not to ask the lower bosses how they got there.

To make matters worse, Wolfsburg lost 5 to 1 to Bayern.

elasto
Posts: 3778
Joined: Mon May 10, 2010 1:53 am UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby elasto » Wed Sep 23, 2015 12:26 pm UTC

Diadem wrote:It's obviously criminal, and they deserve a huge fine for the environmental damage this has caused. That being said, I don't quite see how this can be interpreted as consumer fraud. Selling someone a car that performs better than promised is not fraud.

I drive a 2014 Volkswagen Golf diesel, and it's an awesome car. Before buying this one, I've tested most cars in its price range*. The Golf has the best performance, and by far the best handling.

How does it perform better than promised? Most likely the mileage was based on when it was running dirty, while emissions were based on when it was running clean. So at best it would perform exactly as promised mileage-wise (but worse than promised emissions-wise).

Now, in order to meet emissions standards, it will have to run clean all the time, meaning its mileage during daily use will drop drastically. (Unless your country doesn't do emissions tests of course, in which case you can probably skip getting the software patch)

That's fraud. You paid for your car and now its performance will be nerfed. You will deserve to receive compensation.

User avatar
Quercus
Posts: 1810
Joined: Thu Sep 19, 2013 12:22 pm UTC
Location: London, UK
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Quercus » Wed Sep 23, 2015 12:40 pm UTC

elasto wrote:
Diadem wrote:It's obviously criminal, and they deserve a huge fine for the environmental damage this has caused. That being said, I don't quite see how this can be interpreted as consumer fraud. Selling someone a car that performs better than promised is not fraud.

I drive a 2014 Volkswagen Golf diesel, and it's an awesome car. Before buying this one, I've tested most cars in its price range*. The Golf has the best performance, and by far the best handling.

How does it perform better than promised? Most likely the mileage was based on when it was running dirty, while emissions were based on when it was running clean. So at best it would perform exactly as promised mileage-wise (but worse than promised emissions-wise).

Now, in order to meet emissions standards, it will have to run clean all the time, meaning its mileage during daily use will drop drastically. (Unless your country doesn't do emissions tests of course, in which case you can probably skip getting the software patch)

That's fraud. You paid for your car and now its performance will be nerfed. You will deserve to receive compensation.

Also, many people factor how environmentally friendly a car is into their purchasing decisions, therefore deceiving them about emissions is quite definitely fraud, regardless of the effect on performance.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Tyndmyr » Wed Sep 23, 2015 2:16 pm UTC

sardia wrote:
SDK wrote:"Previously enjoyed 2012 VW Golf Wagon TDI. AS SEEN ON TV!"

It definitely does affect my model - 4 cylinder TDI engine. I plan to call the dealership later this week, see if they want to give me some free stuff. 8-)

No you don't. What's gonna happen is they'll fix the "defect" and then you'll have a crappy normally performing diesel engine. The whole point of the emissions cheating software isn't just to spew pollution. There's performance and mileage gains to be had.


Precisely.

Enjoy the crap out of your "defective" model, knowing the company will take the blame.

commodorejohn wrote:Only if you live in a place which enforces them and requires regular testing. Which, in the US, is pretty much just California, unless I'm mistaken. Residents of other states should theoretically be able to enjoy improved performance and mileage at no penalty.


MD requires testing frequently. With hefty fines if you do not pass on time. If, say, you moved, and the notification got sent to your old address, so you don't realize until next year, enjoy hundreds of dollars in fines. Even if your vehicle is still fairly new, and still passes.

Enjoy living in a place that isn't a liberal dystopia.


As for how many people know... Not sure, exactly. Modern cars have traction control and stuff, so special code, etc normally exists so they don't freak out when only a single axle is moving for testing. Logically, they must have just added some extra code to that particular branch. That isn't necessarily something that would be widespread knowledge, but it's almost certainly intentional.

I'm not sure if EPA uses the same testing hardware for both mileage and emissions. A difference may exist that allows optimization for both separately, but I'm not certain this is the case.

User avatar
Diadem
Posts: 5654
Joined: Wed Jun 11, 2008 11:03 am UTC
Location: The Netherlands

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Diadem » Wed Sep 23, 2015 2:24 pm UTC

elasto wrote:
Diadem wrote:It's obviously criminal, and they deserve a huge fine for the environmental damage this has caused. That being said, I don't quite see how this can be interpreted as consumer fraud. Selling someone a car that performs better than promised is not fraud.

How does it perform better than promised? Most likely the mileage was based on when it was running dirty, while emissions were based on when it was running clean. So at best it would perform exactly as promised mileage-wise (but worse than promised emissions-wise).

If it's running clean during tests, and dirty in normal use, then compared to the test I get a car with worse emissions, but better mileage and performance.

If tests for emissions, mileage and performance are three separate tests done at three separate moments, then what you say could be possible. But how does the car detect what type of test it's being subjected to? And why would anyone in their right mind set up a test protocol like that. I admit that the people making the regulations being out of their mind isn't such a far-fetched possibility, regulations for cars (at least in The Netherlands) are staggeringly stupid. But I'd still like some citation on this claim.

elasto wrote:Now, in order to meet emissions standards, it will have to run clean all the time, meaning its mileage during daily use will drop drastically. (Unless your country doesn't do emissions tests of course, in which case you can probably skip getting the software patch)

Will there be a mandatory software patch? That seems unlikely. I don't think many politicians are going to stick their neck out over punishing consumers for a multinational's fraud.
It's one of those irregular verbs, isn't it? I have an independent mind, you are an eccentric, he is round the twist
- Bernard Woolley in Yes, Prime Minister

Chen
Posts: 5579
Joined: Fri Jul 25, 2008 6:53 pm UTC
Location: Montreal

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Chen » Wed Sep 23, 2015 2:25 pm UTC

Tyndmyr wrote:As for how many people know... Not sure, exactly. Modern cars have traction control and stuff, so special code, etc normally exists so they don't freak out when only a single axle is moving for testing. Logically, they must have just added some extra code to that particular branch. That isn't necessarily something that would be widespread knowledge, but it's almost certainly intentional.


You'd think there'd at least be coders, software team leads, software managers and all who would know about this type of code that was put in there. That said I'm used to working with aircraft engine software which has extremely strict guidelines for development. The automotive industry might be less robust in their softwar development process, since that robustness is costly. Still any complex piece of code is going to be documented SOMEWHERE. Clearly they took time to develop this code, so they wouldn't risk losing it completely because they hadn't written down it's requirements somewhere. Not only that, there was clearly someone who wrote and implemented the code. You'd think that person would question what was being put into the software, and that their manager would hear about it, etc etc.

User avatar
Quercus
Posts: 1810
Joined: Thu Sep 19, 2013 12:22 pm UTC
Location: London, UK
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Quercus » Wed Sep 23, 2015 2:28 pm UTC

What I'd be interested in knowing is if implementing this deception was an edict from high up, or rather a programming team being pressured to increase performance without failing emissions tests, and choosing to take that requirement rather too literally. I doubt we'll ever find out though, because if it was the former and done "properly" there will be enough deniability to make it look like the latter.

User avatar
bigglesworth
I feel like Biggles should have a title
Posts: 7461
Joined: Sat Apr 07, 2007 9:29 pm UTC
Location: Airstrip One

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby bigglesworth » Wed Sep 23, 2015 2:37 pm UTC

What I'm waiting for is the reveal that all the other major manufacturers have done something like this. I don't believe VW would do it alone.
Generation Y. I don't remember the First Gulf War, but do remember floppy disks.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Tyndmyr » Wed Sep 23, 2015 2:46 pm UTC

Chen wrote:
Tyndmyr wrote:As for how many people know... Not sure, exactly. Modern cars have traction control and stuff, so special code, etc normally exists so they don't freak out when only a single axle is moving for testing. Logically, they must have just added some extra code to that particular branch. That isn't necessarily something that would be widespread knowledge, but it's almost certainly intentional.


You'd think there'd at least be coders, software team leads, software managers and all who would know about this type of code that was put in there. That said I'm used to working with aircraft engine software which has extremely strict guidelines for development. The automotive industry might be less robust in their softwar development process, since that robustness is costly. Still any complex piece of code is going to be documented SOMEWHERE. Clearly they took time to develop this code, so they wouldn't risk losing it completely because they hadn't written down it's requirements somewhere. Not only that, there was clearly someone who wrote and implemented the code. You'd think that person would question what was being put into the software, and that their manager would hear about it, etc etc.


*shrug* Not my area, but software development styles vary greatly. Some people are very much about reviewing every little thing, some people test code in production.

It seems reasonable that at least a few people are in on it, but who ordered it will likely be a point of some contention.

elasto
Posts: 3778
Joined: Mon May 10, 2010 1:53 am UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby elasto » Wed Sep 23, 2015 2:51 pm UTC

Diadem wrote:If tests for emissions, mileage and performance are three separate tests done at three separate moments, then what you say could be possible. But how does the car detect what type of test it's being subjected to?


It was quite sophisticated:

BBC wrote:Full details of how it worked are sketchy, although the EPA has said that the engines had computer software that could sense test scenarios by monitoring speed, engine operation, air pressure and even the position of the steering wheel.

When the cars were operating under controlled laboratory conditions - which typically involved putting them on a stationary test rig - the device appears to have put the vehicle into a sort of safety mode in which the engine ran below normal power and performance. Once on the road, the engines switched from this test mode.

The result? The engines emitted nitrogen oxide pollutants up to 40 times above what is allowed in the US.


I dunno about you, but that certainly sounds sophisticated enough to be able to tell the difference - especially given that the mileage test is done at a time and place of the manufacturer's choosing, without any third-party oversight:

Which wrote:Don’t be fooled by the name of the official driving cycle currently used to calculate mpg figures – the New European Driving Cycle (NEDC) hasn’t been new since it was last updated in 1997, and the basis of the test was first introduced in the 1970s.

The test’s lack of real-world driving scenarios and numerous loopholes make the figures it generates unachievable when you get behind the wheel of a car. Here are some of the most notable:

- All manufacturers follow the same test procedure, but can select any accredited lab to use for the test. It’s very hard to get truly repeatable and comparable results when using multiple labs.

- The test cycle only includes urban (in town) and extra urban (out of town) driving, but while it reaches a top speed of 75mph for 10 seconds, it doesn’t include any sustained motorway driving. This is the type of driving where many cars consume the most amount of fuel in our tests.

- Many modern cars have adaptable driving modes to make them more economical or sportier. Our testing finds that ‘eco’ modes can make cars feel unresponsive. Manufacturers can elect to carry out the official test cycle using these, but we think most drivers won’t opt to use them.

- The test is conducted with all ancillary loads turned off, including air conditioning, lights and heated windows, thereby making the car more efficient.

- There is a tolerance for the testing to be carried out at 1.2mph below the required speed, meaning less fuel is used, although the speeds used are already quite pedestrian.

- Roof rails, extra lights and even the door mirror on the passenger side can be removed – this makes the car lighter and therefore more fuel efficient.

- There is no restriction for the air pressure in the tyres, meaning manufacturers can use higher-than-recommended pressures to reduce rolling resistance, taking load off the engine and reducing fuel use.

- There is no official body in place to police the testing procedure and monitor results from lab to lab.

- If all that wasn’t enough, the rules allow carmakers to arbitrarily knock 4% off the results at the end of the cycle.


Diadem wrote:Will there be a mandatory software patch? That seems unlikely. I don't think many politicians are going to stick their neck out over punishing consumers for a multinational's fraud.

I think it's extremely likely given how damaging the emissions are to the environment; Pollutants 40 times higher than the legal maximum..?

http://www.bbc.co.uk/news/business-34324772
http://www.which.co.uk/cars/choosing-a- ... -test-mpg/
Last edited by elasto on Wed Sep 23, 2015 3:05 pm UTC, edited 1 time in total.

commodorejohn
Posts: 1197
Joined: Thu Dec 10, 2009 6:21 pm UTC
Location: Placerville, CA
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby commodorejohn » Wed Sep 23, 2015 3:00 pm UTC

Tyndmyr wrote:Enjoy living in a place that isn't a liberal dystopia.

Too late, I'm in California these days. I just get around it by driving a car manufactured before 1975.
"'Legacy code' often differs from its suggested alternative by actually working and scaling."
- Bjarne Stroustrup
www.commodorejohn.com - in case you were wondering, which you probably weren't.

User avatar
Echo244
Posts: 511
Joined: Wed May 20, 2015 9:49 am UTC
Location: Ping! Ping! Ping! Ping!

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Echo244 » Wed Sep 23, 2015 3:12 pm UTC

It really sounds like a programmer somewhere came up with what they thought was a really cool idea, and nobody in management either knew enough of the technical details, or asked enough, to know that it had Bad Idea written all over it. Isn't there an xkcd somewhere close to this?

Image

Replace "Fire insurance policy" with "vehicle emissions test parameters", remove the second "insurance" and it's all good to go... apart from the part where we only just found out about this...
Unstoppable force of nature. That means she/her/hers.
Has committed an act of treason.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Tyndmyr » Wed Sep 23, 2015 3:14 pm UTC

Yeah. Though it's an outside chance, it could have been a "clever" fix by one dude.

Stuff like "detecting it's in a test" is already standard. That isn't fishy at all. The fishy part is changing performance to circumvent the test.

And...that kind of happens in certain instances. Performance IS optimized for mileage tests to a degree, that may not reflect real world performance. It is possible that someone didn't realize the implications of what they were doing.

Chen
Posts: 5579
Joined: Fri Jul 25, 2008 6:53 pm UTC
Location: Montreal

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Chen » Wed Sep 23, 2015 3:28 pm UTC

For that kind of software change though, I'd have to imagine there's at least a few other checks and signatures needed before it actually gets implemented into code that's going out into the field. I mean hell just the software testing should have showed something.

"Uh why is it when we turn this constant on, all the emissions control for the car suddenly comes on. Actually wait, why are all the emissions controls OFF to begin with?"

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Tyndmyr » Wed Sep 23, 2015 3:31 pm UTC

Chen wrote:For that kind of software change though, I'd have to imagine there's at least a few other checks and signatures needed before it actually gets implemented into code that's going out into the field. I mean hell just the software testing should have showed something.

"Uh why is it when we turn this constant on, all the emissions control for the car suddenly comes on. Actually wait, why are all the emissions controls OFF to begin with?"


Yeah, I've worked in plenty of environments were "well, if it builds, and passes minimal functional testing, it's good" is normal. The idea that every variable gets tested seems...unlikely outside of extremely rigid environments. Signatures? Meh. You *might* have peer review.

quantropy
Posts: 195
Joined: Tue Apr 01, 2008 6:55 pm UTC
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby quantropy » Wed Sep 23, 2015 3:39 pm UTC

I can believe that management accepted the 'too good to be true' results without too many questions. But what about the competition? I would expect them to get hold of a vehicle and do a lot of testing to see how the results were achieved, and so realise what was going on. In which case they'd want to blow the whistle - unless they were doing the same thing.

speising
Posts: 2363
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby speising » Wed Sep 23, 2015 3:41 pm UTC

All of this isn't actually new. There were cars that changed their profile when the hood was open, which indicated a test run.

User avatar
Quercus
Posts: 1810
Joined: Thu Sep 19, 2013 12:22 pm UTC
Location: London, UK
Contact:

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Quercus » Wed Sep 23, 2015 3:42 pm UTC

Tyndmyr wrote:Yeah, I've worked in plenty of environments were "well, if it builds, and passes minimal functional testing, it's good" is normal. The idea that every variable gets tested seems...unlikely outside of extremely rigid environments. Signatures? Meh. You *might* have peer review.

And given how well vehicle cybersecurity has been going lately I think it's safe to say that car manufacturers are not paragons of best development practices.

User avatar
Zamfir
I built a novelty castle, the irony was lost on some.
Posts: 7602
Joined: Wed Aug 27, 2008 2:43 pm UTC
Location: Nederland

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Zamfir » Wed Sep 23, 2015 4:02 pm UTC

https://www.tno.nl/media/1969/investigations_emission_factors_euro_6_ld_vehicles_tno_2013.pdf (English summary starts at page 6)
Interesting report from 2013, comparing test NOx emissions with real-world test. They conclude that actual emissions have not improved in a decad or more, even though the official tested limit has been reduced.

Image

The report already mentions that some NOx reduction technologies are turned on during the cycle test, but not in real life. There is a subtle difference, I suppose, between software that reacts to conditions that appear in the official test, and software that positively looks for evidence that it is being tested in a laboratory. VW crossed the line into that second area, but other manufacturers might have walked very close up to that line without fully crossing it.

Tyndmyr
Posts: 11443
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: Volkswagen cheats EPA Tests: Millions of Vehicles affect

Postby Tyndmyr » Wed Sep 23, 2015 4:10 pm UTC

Quercus wrote:
Tyndmyr wrote:Yeah, I've worked in plenty of environments were "well, if it builds, and passes minimal functional testing, it's good" is normal. The idea that every variable gets tested seems...unlikely outside of extremely rigid environments. Signatures? Meh. You *might* have peer review.

And given how well vehicle cybersecurity has been going lately I think it's safe to say that car manufacturers are not paragons of best development practices.


Yeah, my expectations are lower, I think. It's "welp, at least this screw up isn't directly killing people". Because that *totally* happens too.

The idea that we've already hit an ideal tradeoff between emissions and fuel efficiency is a bit unhappy, but possible, I suppose. I can definitely believe that manufacturers have been trying to see how close they can come on fudging the numbers without actually being illegal. That game really does inevitably lead to this point, because SOMEONE will eventually go to far. So, what's the next step? Dial back emissions regulations? Or are we just gonna accept a level of number fudging in order to have "progress"?


Return to “News & Articles”

Who is online

Users browsing this forum: Leovan, speising and 17 guests