There must be a better way to get an XDF?

[Mykk]

Hey guys, I've been working off and on making the XDF for my project going on months now. But, there has got to be another more efficient way to do this? The car is a '95 BMW 540i/6 running a non-security version of the computer known as the 404 DME. I'm using an Ostrich 2.0 as the chip and I'm flashing via TunerProRT. The available XDF online for the application is incomplete. For my vehicle several of the control tables are labeled incorrectly and don't effect what they say they do. I decided to make my own XDF, using XDFcreator. But, the creator doesn't have the 404 DME values preloaded and instead used another DME's values that yielded results. Okay, so now I've got 63 unlabeled and unknown 2d and 3d maps in raw data. A few I was able to recognize. Others I'm literally just driving the vehicle and seeing under what driving conditions are they used by cell hit tracing in hopes that gives me a clearer picture of their function. I can identify between ignition timing and fueling tables by looking at them. So far I've identified: Mass Air Flow voltage vs airflow table, Injector offset, A 4 cell RPM dependent table that seems to globally shift ignition timing, four ignition timing maps one identified as part throttle another that cell tracing uses during idle but doesn't actually seem to adjust timing, Ignition Dwell times, Global fueling raw data, a 4x3 fuel table I haven't found in cell hit tracing yet, TPS Accelerator Shot fuel add, WOT fuel, a coolant temp fuel or timing add, three more tables with coolant temp x-axis that haven't been hit on cell tracing yet..... From what I can tell, out of the 63 tables I'm still missing a few basic controls. I used WinOLS to see if I couldn't find a few more maps, it found 3 that were not in the creator. But those three don't seem critical. There has got to be an easier way. How are you guys making your XDF's from scratch?

[steveo]

they're made either by data sheets from programmers (which are rare and usually have to be acquired by back door means), or by painstaking disassembly of the ecm program. seeking out tables in raw form is a good starting point, but doesn't get you far.

there is no 'easier' way, you are doing the easy way (which also yields the least amount of useful tables)

[Mykk]

Thanks for the response, I realize this will give me just the basic most surface level maps to work with. I just don't know of any other way.