Av MPG Readout vs Calculated

[Longshot270]

If recording more increases your % error then you shouldn't even be watching it to begin with. Your chance of error goes down when the number of samples goes up.

Here is a quick example of why short term averages are bad. For easy math (and since I'm not positive on the official rates other than the data stream rate) we'll say the programmer takes 5 readings per second.

Lets say you have two different calculations going. One set of calculations has a small number of readings at 1000 readings and the other has a large number of readings at 10000 readings. If the current average is 16 mpg for both then it is easy to see how a single stretch of road could impact the mathematical formula.
You climb up a hill with an IECON reading of 5mpg for the 10 seconds it takes to climb up that hill. Then you slowly roll down the other side of the hill with zero throttle input (maxing out the IECON at 40mpg) dragging on the brakes (to not speed and ram the little old lady ahead of you) for 15 seconds. Your average in this section of road is 26 mpg according to the programmer.
Factor that into the "small" calculation and once you are back on the road your average is 17.1. The large equation comes out to 16.1. If you then compare to the "full" amount of 100,000 readings, you end up with 16.0, or no noticeable change. Keep in mind, this is the impact of only 25 seconds of drive time, now consider what happens when you are cruising through traffic or hilly areas where you are switching between the gas and brake pedals as well as changing speed often. If you were able to freely roll down the hill to shorten the recording time, the average would not be thrown off as much. Unfortunately the computer cannot account for elevation changes, lack of throttle input, resistance from wind and tires, etc. Until then the average will not be completely accurate except under ideal conditions.

[ticopowell]

Originally Posted by Longshot270

If recording more increases your % error then you shouldn't even be watching it to begin with. Your chance of error goes down when the number of samples goes up.

Here is a quick example of why short term averages are bad. For easy math (and since I'm not positive on the official rates other than the data stream rate) we'll say the programmer takes 5 readings per second.

Lets say you have two different calculations going. One set of calculations has a small number of readings at 1000 readings and the other has a large number of readings at 10000 readings. If the current average is 16 mpg for both then it is easy to see how a single stretch of road could impact the mathematical formula.
You climb up a hill with an IECON reading of 5mpg for the 10 seconds it takes to climb up that hill. Then you slowly roll down the other side of the hill with zero throttle input (maxing out the IECON at 40mpg) dragging on the brakes (to not speed and ram the little old lady ahead of you) for 15 seconds. Your average in this section of road is 26 mpg according to the programmer.
Factor that into the "small" calculation and once you are back on the road your average is 17.1. The large equation comes out to 16.1. If you then compare to the "full" amount of 100,000 readings, you end up with 16.0, or no noticeable change. Keep in mind, this is the impact of only 25 seconds of drive time, now consider what happens when you are cruising through traffic or hilly areas where you are switching between the gas and brake pedals as well as changing speed often. If you were able to freely roll down the hill to shorten the recording time, the average would not be thrown off as much. Unfortunately the computer cannot account for elevation changes, lack of throttle input, resistance from wind and tires, etc. Until then the average will not be completely accurate except under ideal conditions.

Your math makes perfect sense to me, and unless it was recording wrong on something, the longer it's in the more accurate it is, but I still dont know why my average mpg's change so much. I am driving back tomorrow and Ill report on how they do, but I expect them to drop to about 12 or less going up parley's canyon (pretty big canyon here), then raise to about 15-17 as I hit the park city exit, then lower to 14 by the time I am to Evanston... and this is without ever resetting the Gryphon. the only time I have ever seen it reset is when I unplug it when going to the dealer to get free oil changes. If anyone's Average mpg's are more steady than mine then please let me know, and even though I got an oil change a little more than a week ago (aka not much time calculating) I have seen this same behavior every time I drive my truck ever since I got the Gryphon, and it's been both trucks too... maybe I have a bad hardware something that doesn't record as many points of data... Oh well though, I have had mine for almost a year in this truck, and it was almost a year in my other truck too, and I am not planning on changing because of a miscalculation in my percieved mileage... I know I suck at getting good mileage, but I didnt get a truck to get mpg's, I got it because it's useful, and safe, and also fun ... oh and BTW it hasn't helped get any girls either... maybe that will change in texas...

[Longshot270]

Originally Posted by ticopowell

Your math makes perfect sense to me, and unless it was recording wrong on something, the longer it's in the more accurate it is, but I still dont know why my average mpg's change so much. I am driving back tomorrow and Ill report on how they do, but I expect them to drop to about 12 or less going up parley's canyon (pretty big canyon here), then raise to about 15-17 as I hit the park city exit, then lower to 14 by the time I am to Evanston... and this is without ever resetting the Gryphon. the only time I have ever seen it reset is when I unplug it when going to the dealer to get free oil changes. If anyone's Average mpg's are more steady than mine then please let me know, and even though I got an oil change a little more than a week ago (aka not much time calculating) I have seen this same behavior every time I drive my truck ever since I got the Gryphon, and it's been both trucks too... maybe I have a bad hardware something that doesn't record as many points of data... Oh well though, I have had mine for almost a year in this truck, and it was almost a year in my other truck too, and I am not planning on changing because of a miscalculation in my percieved mileage... I know I suck at getting good mileage, but I didnt get a truck to get mpg's, I got it because it's useful, and safe, and also fun ... oh and BTW it hasn't helped get any girls either... maybe that will change in texas...

To make it steady out you just have to leave it a long time. The influence earlier was from JUST 25 seconds, if you are cruising along at 60, that is not even half a mile...now consider you can get around 400 miles per tank, and the average doesn't really steady out until after you've driven through the second tank. Because of the swings from regular to 40 mpg

I haven't tested this in a while but I bet if you used the cruise control all the time the average may be more accurate. Part of the inaccuracy is from letting off the pedal while rolling. All the time you spend braking is counted as 40 mpg per reading. I think back and the computer rarely uses 0 throttle so you wont have nearly as much 40 mpg getting thrown into the mix. Only problem is you can't use cruise control in varying traffic.

The only reason I look for ways to improve mileage is because I'm cheap and need to stretch the tanks.

You might be able to get some girls out in Del Rio, last time I was passing through I wanted to take my time.

[88Racing]

Ok Jim,
I'll agree on the overwritting data but the most accurate data comes from 4 tank fulls of gas ...that's the optimized amount of samples according to Bill with out reseting each time....

[ticopowell]

So I just got back from utah, and I have had at least 5 tanks of gas since the last time the gryphon was unplugged, and it still varied from 14.x to 17.x on the average mpg's I had it in cruise control most of the trip except for the stops and when I made sure to not hit the people going 10 under the speed limit , I have a little fun program on my phone, and on the way back from Utah I gained 18,431 feet and lost 16,069 feet, with a tailwind I got about 15 mpg's (done by hand) and my gryphon shows about 16.5ish... so the trend continues haha .
My worry still is that mine is fluctuating too much, aka more than it should and more than everyone else's is... but I am happy about the 15 mpg's, I got only 13.7ish on the way to utah... :/.

The trick in Del Rio is to figure out how to outshine all the other brand new 2nd Lt's that are girl hunting! haha

[Jim Allen]

I recall Bill telling me to reset every tank but what you say makes sense, Longshot. I'll leave it alone for a while and see how it goes.

A related phenomenon I recently noted is that my mpg readouts were significantly reduced for a while after I disconnected the battery to erase the adaptive memory. I reverted from an AEM CAI back to stock, so I reset my Gryphon to L1 (because Bill had put the special CAI tune in L2) and disconnected the battery to start at square one again. For about 35 miles or so, driven over a road I know well the reading I should be getting, the (instant and average) mpg were down 25 percent or so. At least it started that way. It gradually worked back up to normal levels as I drove. Bear in mind I use that stretch of road as a "track" for testing mpg when I get new test product to play with, so I am very familiar with what I should be getting on the instant reading, even by the sections of road. By the time I was 2/3s of the way, it was more or less back to normal. Engine readapting?

[Longshot270]

Originally Posted by Jim Allen

I recall Bill telling me to reset every tank but what you say makes sense, Longshot. I'll leave it alone for a while and see how it goes.

A related phenomenon I recently noted is that my mpg readouts were significantly reduced for a while after I disconnected the battery to erase the adaptive memory. I reverted from an AEM CAI back to stock, so I reset my Gryphon to L1 (because Bill had put the special CAI tune in L2) and disconnected the battery to start at square one again. For about 35 miles or so, driven over a road I know well the reading I should be getting, the (instant and average) mpg were down 25 percent or so. At least it started that way. It gradually worked back up to normal levels as I drove. Bear in mind I use that stretch of road as a "track" for testing mpg when I get new test product to play with, so I am very familiar with what I should be getting on the instant reading, even by the sections of road. By the time I was 2/3s of the way, it was more or less back to normal. Engine readapting?

The reason it starts out low then builds back up to the normal range is from accelerating then cruising. When you accelerate you are throwing low numbers into your average. As you cruise you are adding a longer list of normal values. As your normal value count increases the % error from the abnormal values (acceleration values) will diminish. Eventually you'll have enough "good" values that the "bad" values will not be significant. It would only take 21 good readings to make one bad reading disappear from the average. That is why you can make that average shoot up by flooring it for a second or two then letting off and rolling. It only takes a few seconds to go from 30 to 70 but the truck is heavy enough to roll for a while. If you roll long enough, the low readings get displaced by the high ones.

I'll get a datalog later on today to demonstrate this. It'll give me another reason to drive to Lowes for some stuff.