But that's actually irrelevant for the cited report. What the referenced graph says is that from 2008 to 2014, auto manufacturers declared a certain increase in emissions/economy, real world people reported a different amount, and those amounted to, unique to GM in this report, a negative percentage in how well they achieved their claims. That is, on average, GM car fuel economy got worse over this period, or people's driving habits changed for GM cars. (I can't discount the possibility of something else is going on, such as an influx of a different type of car for GM over that period, I haven't read yet how they normalized the study and accounted for things like that).
Manufacturers report fuel consumption during a controlled environment, with specific driving habits, etc.
Real world people drive insane (at least in the US), and most wouldn't even re-pass their drivers-license test, so most assuredly do not have the best driving habits.
So drawing the conclusion that manufacturers are "lying" from the discrepancies between those two data sets is ridiculous.
It's irrelevant because unless both a) people changed their driving habits on average over that time frame to cause more emissions and b) that change was primarily limited to drivers of GM vehicles, then driver habits do nothing to explain the data.
> So drawing the conclusion that manufacturers are "lying" from the discrepancies between those two data sets is ridiculous.
Perhaps you should pay attention to actually what I said, not what you think I said. I didn't say GM was lying, I implied their numbers deserve extra scrutiny.
The whole controversy is over manufacturers reporting false numbers, i.e. lying.
Saying a manufacturer deserves extra scrutiny can absolutely be inferred to be an accusation of them reporting false numbers, i.e. lying.
My reading comprehension isn't at fault here.
Secondly, it's still irrelevant.
You would have to prove that an improvement at throttle position 0.10 over 5 years equates directly to an improvement at throttle position 0.40.
How an engine performs in condition A has no inherent relation to its performance in condition B.
So yes, I stand by what I said. Maybe before accusing people of "not paying attention", you should look at your own language and see if it's ambiguous?
> Saying a manufacturer deserves extra scrutiny can absolutely be inferred to be an accusation of them reporting false numbers, i.e. lying.
> My reading comprehension isn't at fault here
Except that I very clearly said "I can't discount the possibility of something else is going on, such as an influx of a different type of car for GM over that period, I haven't read yet how they normalized the study and accounted for things like that."
> Maybe before accusing people of "not paying attention", you should look at your own language and see if it's ambiguous?
You ignored clear text from me saying that it didn't necessarily indicate GM was at fault. What am I supposed to infer from that?
As to the matter at hand, you still seem to be under the impression that driver habits can account for a change in the numbers we are talking about, when that would mean that GM drivers specifically drive very differently than all other car drivers in the report. We are talking about number across the industry, categorized by manufacturer. Everything listed on the graph in question shows automakers overreporting gains, which could be driver habits. GM specifically shows vastly different numbers from event every other manufacturer, not just the reported numbers. I fail to see how driver habit accounts for a specific manufacturer being so out of bounds with it's own reported numbers (keep in mind engines are often shared across multiple manufacturers).
> You ignored clear text from me saying that it didn't necessarily indicate GM was at fault. What am I supposed to infer from that?
I didn't ignore anything.
I said the source of data is not irrelevant. That's my entire argument.
And I was specifically refuting your thesis that the source of data is irrelevant.
and, "as to the matter at hand", yes, driver habits can account entirely for the discrepancy in numbers. Flawed data can account for it. Poor sampling can as well.
The report seems to have come from a self reporting site in the EU. Are you asserting that a US car manufacturer has sold a uniform distribution of make/model/engine cars across the globe?
Are you also asserting that this site is getting a proper randomized sample of the car-buying population, such that they don't all have a similar complaint? In my experience, the people who report such things are one of two cases 1) Extremely Satisfied, or the much more common 2) Extremely Dissatisfied.
Are you then also asserting that if GM optimizes for test conditions, which they believe to be similar to normal driving habits, that their engines would be optimized to perform well under conditions that people actually drive in? That's just sloppy. That's like comparing data from your Dev environment to that from Production, I believe there have been a few academic articles to refute that as well.
But, ultimately, are you asserting that we should just accept the source of data as a source of truth, and never question whether or not that it's valid? Should I just go buy a bunch of tabloids and take that as my "new news", since.....the source is irrelevant, right?
> And I was specifically refuting your thesis that the source of data is irrelevant.
Please stop trying to tell me what my thesis is, as you've misunderstood my statements. The portion of the comment I was responding to is "In my anecdotal experience, drivers tend to get much lower than the manufacturer mileage because they optimize for things other than fuel economy (e.g., speed)." That, by itself, is irrelevant to the question as to why one manufacturer has vastly different numbers than another. As to testing methodologies, yes, this could entirely be testing error. I've never stated my opposition to that possibility. I have repeatedly stated my only intention was to say these numbers need more investigation.
I'm done trying to discuss this with someone who ignores what I actually state in favor of what they decide I must be implying.
The whole controversy is over manufacturers reporting false numbers, i.e. lying.
No, actually it isn't. The controversy is over manufacturers tuning their vehicles' engine behavior to give economy for the tests and for the tests only. By your criteria, that's not a problem but VW has already been bitten really hard by the revelation that they did this.
...You would have to prove that an improvement at throttle position 0.10 over 5 years equates directly to an improvement at throttle position 0.40....
The EPA has essentially said that if this isn't a least somewhat true, the manufacturers are breaking the law.
I'm not sure what your definition of alter is, but for me to alter means to lie.
They weren't simply optimizing for test conditions, they were outright lying during test conditions.
that's what the EPA is saying is against the law.
Here's the thing, and this is my main point:
If the EPAs test conditions reflected normal driving habits, there would be zero difference between Manufacturer reported numbers and real world reported numbers.
If the EPAs test is completely different from normal driving, then it stands to reason that there is no inherent correlation between manufacturer numbers and reported numbers.
IDK if you've ever been through an emissions test (like the ones GM actually goes through), but they're a complete joke as to simulating real world behavior.
So I threw together a histogram [1] of car CO2 emissions as reported by manufacturers for 1200+ model/engine combos. It's not exactly uniformly distributed, to say the least, there is a strong peak at 99 g/km etc. Now that's maybe not conclusive evidence that they are lying, but it's certainly suspicious.
I'm not sure what that proves, at least relative to what I'm saying?
IFF the EPA's regulatory test procedure accurately described/expressed normal driving habits, then the only way that numbers would differ between the real world and the reported numbers would be fraud/lying/alteration, like what Volkswagen Group engaged in (let there be no mistake, it WAS fraud/lying, not a case of poor optimization).
BUT the EPA's regulatory test DOES NOT accurately depict normal driving conditions.
In the current case, a manufacturer has to:
- Realize and Formally accept that the test procedure is inaccurate
- Optimize for:
- Simulated driving conditions, aka the Test Procedure
- Real life driving
To me, it looks like GM just didn't do any of that, and only optimized for the simulated conditions and not real world road conditions.
Depending how far out of wack the test procedure really is could explain the entire data-set.
So the point is that these official numbers should be taken with a generous pinch of salt.
Now, when you talk about GM, remember this is data from Germany, so GM essentially means Opel. Now I believe that before the quite successfull Insignia which launched in 2008, Opel were mainly known for cheap and efficient and small cars like the Corsa and Astra, while now they have a more premium lineup, and "Turbo" versions of several models.
Yes exactly, with a pinch of salt. Exactly what I was trying to say.
And yeah, I've driven a Chevy Spark, and that thing was so slow I could get out and walk faster. I could totally see people driving with their foot to the floor in it.
IDK if you've ever been through an emissions test (like the ones GM actually goes through), but they're a complete joke as to simulating real world behavior.
Does not matter. Your argument is something like, that by dumb luck auto manufacturers produced good emissions results and terrible results in the field.
As understand it, the EPA is saying that if a manufacture is engaged in your two stage approach, "Optimize for:- Simulated driving conditions, aka the Test Procedure [and achieve high fuel economy] - Real life driving [and achieve high power, low fuel economy]" then they are engaging in systematic fraud and are going to get hit big time.
You seem to think it's just a matter of technicalities in the test condition. I'd like references for that.
No. That's not my argument. It has absolutely nothing to do with dumb luck.
The EPA has a set of standards, and manufacturers optimize for those. That's it.
Cheating != optimization. Cheating is like a kid copying someone else's answers. Optimization is studying before the test. Completely different things. And optimization isn't illegal.
If you still don't get it, I don't know what to say, I can't explain it any further.
OK, I get that you saying "Cheating != optimization". That may be your opinion. It may be true in a moral sense.
However, I have good evident that VW's optimization was considered criminal in this instance. And Indeed, since many manufacturers do this, things are pretty dicey.
"“Developing an engine software to optimize certain aspects of an operation cycle that you know the parameters of is a challenge, but it is very possible,” says Thiruvengadam. “Knowing when to switch to the EPA-favorable cycle is the trick; it could be set up to detect the absence of steering-wheel movement, or, and this is known, we often turn off the traction control for testing purposes.” Either way, the result is the same: it turns the emissions controls on for EPA testing and off for real-world driving. Somewhat ironically, the presumed benefits of turning off the controls for normal driving include improved fuel economy and engine power." [1]
It's the old saw about the drunk looking under the light post for his keys, when he probably lost them in the alley; the light's better over here.
If you make laws that can't be verified then you can't enforce them. So you test for things you think you can actually test and hope everything works out.
Wiring up an exhaust sensor to everybody's car every time they need to renew their tabs and having them drive around the city is so infeasible as to be laughable. So they put your car on a dyno or even have you put it in neutral with the parking brake on and have you rev the engine a little bit and see if the tailpipe is a mess.
Nevermind that a lot of cars belch smoke only under load.
