Quote Originally Posted by henk4 View Post
CAFE test....
Do you actually mean that you are not allowed to shift gears in a manual in a different way than the CAFE test prescribes? Is it not enough if the manaul gearbox achieves the same test results as the autobox under laboratory conditions?
No, I can shift how ever I wish hence the reason why I typically beat the EPA numbers with my car. I short shift more that the test allows. But the car companies have to test the car to the test. While an automatic can shift at just the right time (say into second at 8mph) for best mileage a manual with the same ratio is required to shift into second no sooner than 15mph*. So let's take the new Porsche with the DSG and manual boxes. Let's assume that both boxes have identical ratios (no idea if that's true). The computer controlled box almost certainly will deliver better mileage because it will shift at the perfect time, not at 15, 25, and 40 mph.

The car companies are required to sell cars that meet fleet fuel economy averages that are based on those tests. So even if I can get say an average 32mpg out of a car, if the test says the car gets 28 then the manufacture records 28. With the new (and stupid) CAFE increases the car makers will have to look for every little advantage. The easy stuff (improvements in aero, fuel injection etc) have already been had. Now we have to go after the fruit at the top of the tree. Unlike the late 70s, these days a good auto will return better numbers than a manual which is an incentive for car makers to dump the manual. Not only do they have to pay to develop a second transmission, they also have to take a hit in their CAFE numbers.

*In searching for some up to date info I found an interesting post which I can't find again (will have to post it later). Anyway, it talked about the shift lights that make it into some cars. Those lights are a partial way around the standardized shift points. With the lights a manufacture can run the test with the speeds indicated by the lights instead of the ones I listed above. However, the manufacture, according to the post I read, must then study real life drivers and show what percent follow the lights (say 30%). Now they get to do a weighted average of the original test number and the presumably better mileage achieved by following the shift points. So you now have a CAFE mileage number that is .30*the shift light average + .7* the standard shift point average. Confusing huh.