I was assembling some datalog shots from various model year and trim evos today to show mph at 7,000 and 7,500 rpm differences between them.

I quickly found out that the sample rate for MPH in evoscan has far too slow sample rate and that the data couldn't be trusted as I saw different mph at the same rpm from 4 different cars with the same gearbox, tire and model year. It seems the stair stepping of the MPH has huge jumps.

The problem, as you can see in the logs below is that the MPH stair steps and is not a smooth input like RPM is. I have to admit, I have never really looked at MPH closely in my datalogs before so this is the first time I saw the problem.

Is there a way to get a higher MPH/speed sample rate?

*LOGS UPDATED TO SMOOTH AVERAGED DATA FROM DLL: (thanks NJ!)

 

You can create a power graph and then copy the power graph into the main screen using the Tools>click to sample graph on the power graph menu bar. That will display the speed as it is calculated by DLL.

 

^why dont you open the log files in excel, and just find the averge for the steps? cant excel do an inferrence? and make a smooth line?

 

That worked perfect, thanks NJ!

 

Ok logs updated in first post. It still feels like it's too low resolution for comparison purposes since DLL is just averaging coarse data.