When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
to date, 997.2 MPG (7 months ownership) is 22.9 mpg, However, the computer will show a higher value. I've plotted the actual (y-values) vs. the indicated (computer) values (20 tanks of gas so far) with the following results:
As you can see, the computer gives a reasonable linear correlation between indicated and actual miles/gallon. However, it (at least on my 997.2) reads a bit over 2 mpg high (y = 1.0107x - 2.0672; R² = 0.92231 where y is the actual mpg, x is the computer indicated MPG and R² = is the Correlation Coefficient which is a goodness of fit parameter). [the upper grey line is y = 1x, with zero offset] Actual mpg at a fixed 65 mph (SD to LA freeway driving, no traffic) is a bit of 29 mpg.
On my 996.2 (2 years of ownership followed by IMS failure) ave mileage was 24.2 mpg (hi-29.1 on a road trip, low was 19.5 LA freeway traffic)
• correlation on ODO computer was y = 0.9936x - 0.4759; R² = 0.89209
What this tells me is that the computer does a good job (±1%) of computing the actual mpg, subject to a fixed offset. I don't know why both my 911's had an offset. If it was wrong tire size, the slope would be a number not equal to 1. I wonder if this offset could be corrected by reprograming the electronics that do the calculations.
to date, 997.2 MPG (7 months ownership) is 22.9 mpg, However, the computer will show a higher value. I've plotted the actual (y-values) vs. the indicated (computer) values (20 tanks of gas so far) with the following results:
As you can see, the computer gives a reasonable linear correlation between indicated and actual miles/gallon. However, it (at least on my 997.2) reads a bit over 2 mpg high (y = 1.0107x - 2.0672; R² = 0.92231 where y is the actual mpg, x is the computer indicated MPG and R² = is the Correlation Coefficient which is a goodness of fit parameter). [the upper grey line is y = 1x, with zero offset] Actual mpg at a fixed 65 mph (SD to LA freeway driving, no traffic) is a bit of 29 mpg.
On my 996.2 (2 years of ownership followed by IMS failure) ave mileage was 24.2 mpg (hi-29.1 on a road trip, low was 19.5 LA freeway traffic)
• correlation on ODO computer was y = 0.9936x - 0.4759; R² = 0.89209
What this tells me is that the computer does a good job (±1%) of computing the actual mpg, subject to a fixed offset. I don't know why both my 911's had an offset. If it was wrong tire size, the slope would be a number not equal to 1. I wonder if this offset could be corrected by reprograming the electronics that do the calculations.
Really good data! Regarding the offset with slope of 1: Could this be Porsche's approach to cheating? Seriously, the temptation to to tweak a consumer experience is mighty tempting.
to date, 997.2 MPG (7 months ownership) is 22.9 mpg, However, the computer will show a higher value. I've plotted the actual (y-values) vs. the indicated (computer) values (20 tanks of gas so far) with the following results:
As you can see, the computer gives a reasonable linear correlation between indicated and actual miles/gallon. However, it (at least on my 997.2) reads a bit over 2 mpg high (y = 1.0107x - 2.0672; R² = 0.92231 where y is the actual mpg, x is the computer indicated MPG and R² = is the Correlation Coefficient which is a goodness of fit parameter). [the upper grey line is y = 1x, with zero offset] Actual mpg at a fixed 65 mph (SD to LA freeway driving, no traffic) is a bit of 29 mpg.
On my 996.2 (2 years of ownership followed by IMS failure) ave mileage was 24.2 mpg (hi-29.1 on a road trip, low was 19.5 LA freeway traffic)
• correlation on ODO computer was y = 0.9936x - 0.4759; R² = 0.89209
What this tells me is that the computer does a good job (±1%) of computing the actual mpg, subject to a fixed offset. I don't know why both my 911's had an offset. If it was wrong tire size, the slope would be a number not equal to 1. I wonder if this offset could be corrected by reprograming the electronics that do the calculations.
As a scientist, I heartily endorse calibration graphs! Of course, I prefer curves to lines!
T
06-05-2016, 10:12 AM
