Monday, September 21, 2015

German Software Engineering

Do you trust your car's computer Eli asks

The announcement that VW had incorporated software in it's diesels, that turned the emissions system on when it was being tested and off at all other times has, well, think electric fan and moving manure. 

It has been well known for a long time that the manufacturers' test results for mileage and emissions were a lot better than independent labs or on the street tests, and that the various responsible agencies in the US and Europe turned the other way rather than looking at this issue in detail.

Within a few days it will be clear or at least clearer how far the scandal extends, to other manufacturers, and to other types of engines.  Will BMW and Mercedes try and toss VW under the bus?  Will this reach US, Japanese and Korean manufacturers.

A related question is that after this scandal will the manufacturers be able to continue protecting their computer code under the Digital Millenium Copyright Act.  The bad faith that VW showed its customers and the regulators is going to make that very hard to defend.


Anonymous said...

As someone who has owned VW vehicles in the past and until recently kept them on my evaluation list for my next car, I hope that VW has the book thrown at it, that it is a very heavy, very expensive book and that criminal charges for the individuals responsible are not ruled out.

And the same goes for any other manufacturer that engaged in such deceptive practices.

Everett F Sargent said...

"It has been well known for a long time that the manufacturers' test results for mileage and emissions were a lot better than independent labs or on the street tests, and that the various responsible agencies in the US and Europe turned the other way rather than looking at this issue in detail."


I don't doubt what you say, but the days of the op-ed with NO links ... I guess are still with us?

Op-ed's, for me in general, died like over three decades ago, before there was a public internet even. Meaning that I don't read them because they are almost always bias-with-intent-hit pieces.

So say that VW is guilty as charged, they deserve the maximum fines allowable under our current laws or whatever judicial decisions are handed down.

However, if this was an SOP industry wide practice (e. g. they all did it) and say the EPA knew this or purposely looked the other way, what happens then? Do 'we the people' sue Uncle Sam the auto industry or both?

As to the DMCA, are you suggesting that those computer embedded microprocessor codes be placed in the public domain and accessible/modifiable by the end user?

If given that chance, I'd modify the firmware for maximum MPG efficiency for a maximum highway speed of 55 MPH (I sort of did that (yeah, I limited my maximum highway driving speed to 55 MPH), drove 5,500 miles and averaged 41.5 mpg (tires were somewhat overinflated to reduce rolling resistance) with a 1996 Mazda 626 I4 (standard tranny SI) which the EPA said should only get like 31 mpg highway).
(I don't think it matters if this estimate was with the old standard or the new standard as the old standard would have been higher than the new standard)

I can also see a lot of 'rednecks' and 'hobbiest' doing the exact opposite.

But I'm an old fart, time isn't an issue, I don't need to see 666 national parks in like one day even.

Also, since CI (diesel) gets better MPG versus SI, but the cost of diesel fuel in MS was over $0.50/Gal (when gas was like $3.50/Gal) but now is like ($1.79/Gal (regular) versus $2.01 (diesel)), I think CI wins in the MPG battle.

Time to get an old (circa 2005) small standard tranny diesel VW and get like 70 MPG and limiting my speed to say 45-50 MPH (whatever the optimum MPG turns out to be).


Brian said...

Eli's first link has an update, saying sales have now been halted in Canada too. Where it ends, nobody knows.

Anonymous said...

Come on, Eli. This is nothing compared to Deflagate. Please leave auditors alone.

Everett F Sargent said...

Well that was rather simple ...



40 CFR Part 86, Appendix V to Part 86 - The Standard Road Cycle (SRC)

Testing procedures (like ASTM or ISO or in this case EPA) are all rather well defined or codified.

VW 'hacked' known test sequences, that embedded microprocessor code is bound to come out in discovery.

Sort of like digital DNA. VW is DAA (Dead After Arrival).

BBD said...


There have been rumblings about this for a couple of years in the UK.

My wife just replaced a diesel VW Golf with a *petrol* engine car because we knew from the media that the diesel test results were misleading. This includes CO2, particulates, NOx. We've know it for some time. This prompted a planned change from diesel to petrol.

We are not unusually well informed on this topic. We do not work for the automotive industry. I repeat, this has been in the papers for the last couple of years, on and off.

What shocked me was that this has taken so shamefully long to break and when it did, it came from the US (what's going on with OUR regulatory bodies and OUR testing labs FFS? That's what I want to know.)

BBD said...

I should also add that there is apparently clear evidence that the practice of cheating emissions tests is industry-wide.

BBC News 14 March 2013 admittedly does not mention software tweaks but points to systematic deception.

Anonymous said...

Here you go, BBD, read it and weep, innumerate, illucid crank.

BBD said...

[number string]

You seem to forget that I was correct all along, which highlights how unwise it is to use Greenpeace as a source of information on energy.

Now, I don't wish to be unnecessarily rude (I leave that to you), so I will simply suggest you let your grotesque embarrassment on the other thread fade into the past rather than reminding us all what a colossal prat you made out of yourself.

Anonymous said...

You're right, how could I have missed that. DLR is an unsavory research organization, and some shill for the nuclear industry already has it all figured out. I get that. Shoot the messenger, Greenpeace is an easy target. The nuclear industry, not so much.

Blogger profile said...

"You seem to forget that I was correct all along, which highlights how unwise it is to use Greenpeace as a source of information on energy. "

It's very unwise to use thinktanks because they're paid to have an opinion. It's also unwise to use the generating industry because they're making money off the big systems, whereas the renewable systems are making a hash of their systems' profitability.

Given your failures in the past to get past your ideological blinkers (mostly because you don't KNOW that they are there, you "think" that you're "fair and balanced"), you really should know better than to make such asinine statements.

Blogger profile said...

" Here you go, BBD, read it and weep, innumerate, illucid crank."

Well, your post was entirely innumarate and illucid crank.

Care to give any evidence *at all* for your claim about the pdf, or shall we chalk it down to partisan hate?

BBD said...


Well, your post was entirely innumarate and illucid crank.

It's always projection...

Care to give any evidence *at all* for your claim about the pdf, or shall we chalk it down to partisan hate?

Sure. The pdf is the usual bollocks. See p. 6:

The cost of transition.

The introduction of renewable technologies slightly increases the cost of electricity generation, compared to the IEA scenario, though the difference is marginal - only some 0.2 to 2 US cents/kWh (excluding integration costs for storage or other load-balancing measures.).

Storage and grid enhancements, including interconnectors are not costed. That's how clowns like Teske get their numbers to 'work'.

This stuff is arschpapier. Nobody with a clue takes it seriously. QED.

Anonymous said...

Oh boo hoo, the very first NECESSARY step to even begin to address the problems that fossil fuels YOU have created for yourself with fossil fuels is going to cost you. Hint - it's going to hurt too.

Innoculations generally do. But you somehow think you are immune.

Blogger profile said...

""Well, your post was entirely innumarate and illucid crank."

It's always projection..."

cf the post I was replying to when it claimed:

"OpenID 8c7793aa-15b2-11e5-898a-67ca934bd1df said...

Here you go, BBD, read it and weep, innumerate, illucid crank."

Yup, always projection. Unlike OID here, mine actually had some evidence to support it.

"Storage and grid enhancements, including interconnectors are not costed."

They're not costed with ANY of the power systems. Not in the past and not in their future.

So it isn't a new cost for renewables AT ALL.

Again, you're retarded crusade for Nuclear Uber Alles blinds you with hate and anger.

Blogger profile said...

"Nobody with a clue takes it seriously. QED."

IOW !"I'm right because I don't take it seriously, you're wrong if you do because only people without a clue take it seriously".

This is a completely asinine and clueless "argument by assertion".

However, you haven't got anything else, you don't have the mental capacity to do any better and you don't WANT to try.

BBD said...

Oh for pity's sake, sod off.

Anonymous said...

Not a chance. On the contrary, the imminent collapse of modern civilization demands even more urgent and pointed commentary and messaging. If anything, I will start naming names, and crimes.

I don't care where you think you are with the problem, I want solutions.

You apparently have nothing but feel you need to muddle the discussion.

Blogger profile said...

" Oh for pity's sake, sod off. "

Really? That's supposed to "work" as a rebuttal???

It IS entirely innumarate and illucid crank.

Marlowe Johnson said...


It's well known that laboratory test cycles overstate the fuel economy benefits of vehicles compared to real world driving, and in the past this was mainly due to conservative assumptions about driver behaviour (e.g. air conditioning use, acceleration, etc). To make matters even more interesting, different jurisdictions use different test methodologies. What the ICCT report shows (incidentally I've worked with them a number of times over the years and they're top notch) is that manufacturers have been actively pushing the envelope to increase their ratings. Given the increasing stringency of fuel economy rules this is hardly surprising.

However, these sorts 'tricks' don't apply to criteria air pollutants which are treated with end-of-pipe technologies. Now you could argue that willingness to game the system in one area is suggestive of gaming in another, but I'd hesitate to draw too many conclusions about the behaviour of the industry as a whole in this area.

Jan Galkowski said...

This is clearly cheating and it suggests that the engineers involved are not following the ethics codes and guidelines of many engineering societies, including those of, for instance, the IEEE. Its first principle says:

"to accept responsibility in making decisions consistent with the safety, health, and welfare of the public, and to disclose promptly factors that might endanger the public or the environment"

and its third says:

"to be honest and realistic in stating claims or estimates based on available data"

and its last says:

"to assist colleagues and co-workers in their professional development and to support them in following this code of ethics."

I think the conflict of interest which already exists because an employer pays you always ought to be tempered by heightened sensitivity to the possibility that the employer will use that asymmetry to do something contrary to The Good, especially for engineers and scientists.

In some cases, the asymmetry is particularly bad, such as when test engineers report, ultimately, to the head of the organizational unit which has everything to gain by pushing a product out the door, working or not.

Jan Galkowski said...

... And on another matter, closer to the busines we're most concerned about, with all these "commitments" and INDCs being made by countries, even granted that they are not overwhelmingly in line with what's needed, and the announced choice of "cap 'n' trade" as a policy of choice by others, I wonder if (a) there will be an independent scientific check on the progress the globe is making at reducing its GHG emissions, (b) if governments will believe the result when it (inevitably) is found to fall short of promises, (c) if governments will act on the result when it is found, and (d) how good attribution will get, by country.

On the latter points, it sounds to me like a good bit of engineering, research, mathematics, and economics to be done to develop an impartial means of monitoring emissions and finding ways of attributing these to specific countries if not cities and locales.