Is it the case that we understand exactly how the stock ecu identifies what is and is not knock? Or have we not dissambled this logic?

 

thats the million dollar question man, like I posted before, can you really believe any datalogger since its not a full memory dump, but instead just a trasfere of data in real time. I just began to fiddle with matlab to look at signals in real time in the frequency domain to get a real picture of all that is going on so I can ultimately understand there algorithms. Make filters ect.. and see what happens.
I dont trust 100% any data logger out there, but they do a terrific job and is very usefull nevertheless.

 

Yes, the routines have been disassembled and are very complex. You can read about it over at aktivematrix. I forgot what thread it is in, but Bez talked about the disassebmly somewhere. That's actually where all of the newly loggable knock varibles that you see in the newer EvoScan came from.


Eric