You're probably paying attention to the CNW "junk science".

To quote from http://www.pacinst.org/topics/integrity_of_science/case_studies/hummer_vs_prius.pdf:

Another example of an unusual assumption and choice of data is the reported distribution of energy across the different phases of vehicle life. The CNW results suggest that the majority of energy is consumed during the production of the vehicle. These results are at odds with every other study weâ(TM)ve seen on the energy life-cycle costs of automobiles. Other studies independently conclude that the vast majority of energy is consumed during âoevehicle operations,â with lesser quantities used during materials acquisition, fabrication, and vehicle disposal. For example:

â A report produced by a British research firm concluded that more than 90% of all energy used in the motor industry went to vehicle operation; less than 10% went to manufacturing and production.

â The British auto industry trade group estimated in their 2006 sustainability report that life cycle CO2 emissions â" a strong proxy for energy â" are allocated 10% to manufacturing; 85% to use; and 5% to disposal.

â The Center for Sustainable Systems of the University of Michigan, which pioneered and refined the tool of life-cycle assessment, conducted a joint project with Chrysler, Ford,General Motors, the Aluminum Association, the American Iron and Steel Institute, and the American Plastics Council. They analyzed the life-cycle energy costs of the 6 systems, subsystems, and 644 discrete parts and components composed of 73 different materials comprising a typical North American mid-sized car and concluded that more than 85% of all energy is the result of using the car, not making, assembling, repairing, or disposing of it.

â A comprehensive energy life-cycle analysis of a Volkswagen Golf Mark 3 concluded that 73% of total energy is consumed during the use and disposal phases, 11% in materials production, 8% in vehicle manufacturing, and 8% in fuels manufacturing.

â The MIT study, âoeOn the Road in 2020,â reported on a comprehensive energy life-cycle analysis and found that 80% to 90% of all energy was used in the operation stage; 7% to 12% in the materials production stage, and the remainder in vehicle assembly, distribution, and disposal.

â A 2006 study from Argonne National Laboratory concluded that around 75% of all hybrid and internal combustion vehicle energy use comes from the operation of the vehicle. The rest comes mostly from producing the fuels and the manufacture and disposal of the vehicle and its materials.

How long do you have to run the new car before the amount of fuel you've saved is more than the amount used to build the new car?

Never. It takes vastly more energy to produce a new car than the car will ever consume. Even thirsty old Volvo 240s will be thundering away well past the end of their 22-year design lifespan, still consuming 24mpg and *still* nowhere near the amount of energy it took to make them.

Even the more far-out wacky environmental groups are agreed on this - it makes no economic or ecological sense to keep churning out new cars that are only a tiny bit cleaner than the old cars they replace, taking ten times as much energy to produce.

http://www.atarimuseum.com/videogames/consoles/7800/games/ says:
"Note: If you are going to Mirror these sources or place them onto your own site, please have the respect and courtesy to include with them - Source: www.atarimuseum.com as these wouldn't exist if I hadn't of climbed into a filthy dumpster at 3am in the morning behind the old Atari building in Sunnyvale and salvaged them and restored them from their diskettes."

http://www.cleanmpg.com/forums/forumdisplay.php?f=23 has some hints. They also have specific articles for some cars like the Prius. See http://www.cleanmpg.com/forums/showthread.php?t=1224.

http://www.motherjones.com/news/feature/2007/01/king_of_the_hypermilers.html is an article about a hypermiler (Wayne Gerdes) who achieves 59 mpg in his non-hybrid 05 Honda Accord.

Gerdes was part of a team of who set a record (which has been since beaten) of ~110 mpg in a Prius over 1397 miles on a single tank.

There are some other tips at http://www.hypermiling.com/.