pcmhacking.net

Another update. I've been working on our file format and we have designed it in a way that will allow unlicensed users to edit other users files.

Eg you do a read of a car, license the strategy to your own personal account (so you can edit/read/write as many times as you want) then send the file to anyone.

If the person you send it to has no licenses (eg they haven't spent a cent) they can edit your file and save it and send it back to you. When you re-open the file it will check if you have a licensed template and merge the changes back into the binary (this all happens in the background seamlessly) and you can write it to your car.

We feel this will be a big step up over other tuning companies that do not let you edit files you haven't licensed. I believe the reason they do it is due to their file format this would be easily cracked.

The way we have designed the file format this is basically uncrackable as the addressing information is not stored in the file or the application. Hence if you haven't got a licensed template there is literally nothing to crack.

So this is a win win for everyone, hopefully will mean tuners (and DIY guys) can remotely "tune" peoples cars much easier. I know I've wanted to edit other peoples files to help out before but it is close to impossible to do.

Curious on peoples thoughts.

That sounds like a better idea than the current practice. The person writing the file needs to have a licence for that ecu to be able to reflash the ecu. It would be nice if the licence can be transferred. If the person pays shop a and wants to fix minor problems they do not have to buy a licence as well. Shop a can transfer the licence after finishing the tune. Usually if some needs this they are not interested in talking to the original shop.

Transferring licenses will be way too hard to implement or police, no one else offers this for similar reasons. Would open it up a huge number of exploits.

SCT etc get around this by making each user buy a $600 hand controller. So if you wanted to flash it after paying a workshop you'd have to buy your own license which would still be way cheaper, also you can still edit your tune unlike SCT.

HPT make everyone buy a cable ($1k) and a license ($240), SCT make everyone buy the software ($4k) and a hand controller ($600).

Buying two of our licenses and an open port (~$200) would be nothing in comparison.

Some more teaser screenshots. Almost implemented flashing into the editor. Teaser shots of the new exhaust cam stuff we are testing out.

To do ghost cam mod(that all the kiddies love) wouldn't you want to retard the low load & low rpm points only?

VX L67 Getrag wrote:To do ghost cam mod(that all the kiddies love) wouldn't you want to retard the low load & low rpm points only?

That's what I imagine yes, I haven't had time to actually play with it yet. from the factory they seem to run ~20 degrees once load is above 1.0. Though it seems inconsitent, eg my HACCK tune was 0 everywhere, I just put some random numbers in to test the map saved properly, those numbers are not representative of anything.

This is awesome! please look at the TDV6 diesel ECM next!

ejukated wrote:This is awesome! please look at the TDV6 diesel ECM next!

What vehicle are they?

Ford Territory its the 2.7 v6 diesel. It uses a Siemens SID204 ECU, same combo is also used in Jaguar and Land Rover. I think the seed/key is different on the AU Fords as the usual euro can connect but they key is invalid. I have a bench ECU if you want me to test anything.

ejukated wrote:Ford Territory its the 2.7 v6 diesel. It uses a Siemens SID204 ECU, same combo is also used in Jaguar and Land Rover. I think the seed/key is different on the AU Fords as the usual euro can connect but they key is invalid. I have a bench ECU if you want me to test anything.

Would be multiple months work to get started on that. I have no tools that can access it currently nor a list of PIDs etc.

You need to map a base calibration to then create address templates for the hundreds of firmware revisions. If you have literally nothing as a base it takes a long time to get started as you have to map PIDs to memory addresses, then decipher the map struct format and find common interpolation routines etc then from there you can find all the maps. Once you find all the maps you need to figure out what exactly they are mapping and they actually do. If you can find an A2L file for a single calibration this saves you a few months of base work. These are usually leaked or bought under the table from OEM employees.

Once we get the BA/BF/FG nailed we will look at other ECUs, first we need to get this finished and polished though.

You also have to weigh up demand vs other ECUs, if there's only a handful of people interested it would make a poor time investment. I think we will set up a public voting system for what ECUs have the most demand, also for new features etc at some point, this will be a good way to gauge what people are interested in.