Software and electrical engineers look here

[jonbonazza]

Being that a good 90%of the twins community is some sort of engineer, i think we have the perfect setup to follow through with an idea that i have had for a while now.

I think it would be very interesting if there existed a completely free and open source ecu flashing solution that didnt just target one platform, but instead was modular to a degree that anyone with the know how could develop a plugin for the software for their particular car. Basically, it would peovide the foundation that communities could build upon.

What do you all think? Is there enough interest in something like this? If so, anyone want to join in on this endeavor and help make it a reality?

Sent from my Nexus 4 using Tapatalk 4

[userjack6880]

It'd be a neat thing to do. I don't have a whole lot of time myself to dedicate to the project, but I would love to do so.

We just need to be careful and make sure we don't use anything proprietary. Reverse engineering would be fine. Public specs as well. Just nothing that requires licensing.

[jamesm]

Big problem is that this requires very specific (and thus rare) knowledge of protocols and things like that. It's also labor intensive. Most solutions end up becoming proprietary because of the amount of work involved. we're not talking about a graphing library or web server here... this is pretty serious low-level stuff. basically what you're describing is a library that would abstract away the gritty details of j2534 or whatever protocol you need to use to do this sort of thing, and present some generic interface we could use in a higher-level language to then make cool things. the problem is that there are lots of different ways these things are done, and different manufacturers extend the standards with their own additional crap. it's a murky mess of confusion, at least to me lol. it'd be very difficult to make something generic and abstract enough that it could work on any ecu and mere mortals could implement anything useful with it.

That said if there was a serious effort i'd love to be a part of it, but I have none of said specific protocol knowledge.

[robot]

Check out the romraider forums. They are sort of trying to do this. There is a whole open source tuning of ecus community out there,

[jamesm]

Originally Posted by robot

Check out the romraider forums. They are sort of trying to do this. There is a whole open source tuning of ecus community out there,

ecuflash is not open source.

[jonbonazza]

I understand it would be no simple feat to do, but i am also very certain that it is possible. I dont expect to make something that any average joe could develope a plugin for since as you said, there are a lot of intricacies with the various protocols out there, but someone with the know how and technical skills should still be able to develop a plugin given a common interface.

I think ill start brainstorming designs and see what i can come up with. I work with low level protocols and serialization on a dailey basis with my day job, so perhaps that experience will provide some hints.

Is there enough interest to get a google group going for discussion and brainstorming? I dont expect anyone to devote a shit load of tome to this, but would be nice to get a knowledgable community around it and if we can com up with a solid design, maybe even kickstart it so we can devote a little more time with compensation.

Sent from my Nexus 4 using Tapatalk 4

[Raiin]

I'd be interested in the project as well. I oversee a considerably large sized software development team who is always interested in 'third-time' projects. Especially open-source or community focused projects. Looking forward to seeing what comes of this.