kristopher_d

Okay, I'm tired of waiting. I think it's time we get together and offer a bounty to the open source community for cracking the challenge response algorithm for entering program mode in the ECU.

nycgps

iTrader: (1)

no point.

it requires a lot of stuff.

First, someone gotta have the car,

Second, that person have to *give up* his ECU (which means the car)

3rd , someone in the community have to know how auto computer works, I think this is a 32-bit ECU. which is quite Advance for cars.

4th, that person has to know what Rotary engine works 100%

5th, you cannot just reverse engineering it. because you're going to **** alot of stuff up in the process.

6th, they dont program it in C/Java/C#, Im not 100% sure but its either Assemable or Machine code. not that many people know them.

and so and so.

Cracking ECU code is not as easy as cracking your Windows EXE files. Most EXE I can patch them maybe a few Bytes and I can make it work (say, License code), but reverse engineering takes alot of trial and error tries.

and remember, if something in the process ***** up. the ECU (and the car) is a bye bye.

kristopher_d

Sorry, I obviously wasn't clear. We don't need a whole tuning package. There a couple of firms that can do that already. What we need is the ability to answer the ecu's challenge.

Basically, according to my sources, when you try to put the ECU in program mode, it sends a number as a challenge. If you don't respond with the right number as a response, it wont go into program mode. If you give the right response, it will go into program mode. Then, the exiting methods of loading maps will do for now. But the key is unlocking that challenge response algorithm. Then we can load maps and start writing new features like launch control and anti-lag (for turbos), etc.

There may or may not be a need to sacrifice an ECU, and the programming language used to generate the proper response doesn't matter.

MazdaManiac

iTrader: (3)

Just buy an old WDS and leave it at that. It will answer the passcode on its own.

kristopher_d

Right. And there are enough of those lying around for everyone who wants to use the factory PCM instead of piggy backing. I know you and others have had great results with piggy-backs. At the same time, I can't help but think there's more potential in the car if the stock PCM is actually in control and doing what we want. Plus, it's just a more elegant solution.

Hell, I'll put my money where my mouth is, and offer the first $100 of a bounty for BSD licensed software that can reliably put the RX-8 PCM into program mode using the diagnostic port, a generic bi-directional OBD connection, and a standard PC.

nycgps

iTrader: (1)

it still requires reserves engineering of the whole.

in order to get the challenge code, first you have to *dump* the whole rom into your computer. and the challenge code is probably some one way hash or something base on what your VIN number or whatever (just a guess, never work with the PCM before)

if its that easy, the market should be flooded with portable flasher already.

kristopher_d

I didn't say it was easy. But the open source community is motivated as much, or more, by pride as any thing else. I'm just suggesting we offer a little honey and see what happens. If the pot's big enough, some anonymous hacker with inside knowledge may release something.

Phil's 8

iTrader: (3)

Wishful thinking, you could not collect enough money to make it worth while. The RX8 community is way to small to be even noticed by the people that are capable. You'll just have to wait for one of the companies that have put us on the back burner.

rotorocks

Won't be happening for another few years, or until you can pick up a used 8 for a few hundred bucks. then all the teens will jump on the wagon, and start spending mom and dad's money on modding, and finally we'll be noticed.

kristopher_d

You overestimate the monetary motivation of the open source community. The money is just something to draw attention to the problem. Most open source work is done for free/free.

http://www.enginuity.org