[Snipesy]

Hi guys. I made the "BiScan app" which many of you use now. A program used to control your car among other things using Torque as a sort of middle man.

Allow me to introduce you to the next generation of android obd2 scan tools. The Terminal (look I spent enough time coming up with BiScan and GMX gimme a break here).

What makes The Terminal so great?

• Built around Android O. Android O brings some changes to how services are done, and it is crucial that The Terminal follows these changes from the ground up. (Well honestly everything would be fine, but pays to prepare ahead of time).
• It's designed to be free! A major issue that the BiScan app is is no one is really sure if it works until they fork over the money. With the app being free you can see if the features work before you purchase. Even trial them.
• Designed to work OBDLink MXs and LXs, some things wont work with LXs. Traditional ELM devices are being cut for various reasons. Others might be added down the road.
• The app is based around the idea of buying "Feature Packs". Ex. You can buy all the "GM Diesel PIDs" and all the "GM Gas Stuff" separately. Some features will collide a bit, and were still working on some kinks. If you have BiScan you will get GM Diesel PIDs and the assorted control packs for free.
• The ability to program and read from ECUs. This is a very iffy feature. At this time I don't know how far I am going to take it, but it's there. It works. Yes you can flash your ECM from your phone. There are alot of safety systems that go into ensuring reliability, but there is nothing I can do if you walk away from your car, phone in pocket.
• Really fast PID scan rate. How fast? Pretty fast.

It's gonna be great. I dont normally try to sell things until I can get the app on people's phones, but I'm pretty excited how things are currently moving. The goal is to release alongside Android O but that's not necessarily a requirement. (No you dont need Android O to run the app).

Some other notes

• No Android Auto. Sorry. Blame google. You can't really do much with Android Auto other than notifications really.

 

[Cruz15]

Its Great your developing a new app.
Ok, What wont work with the LXs? I have an LX.

 

[Snipesy]

Its Great your developing a new app.
Ok, What wont work with the LXs? I have an LX.

The LX doesn't communicate with every module. Which for the most part isn't really a problem.

 

[Cruz15]

Is it worth buying the MX? If I have an LX?

 

[EricSmit]

This is neat!

 

[Snipesy]

Is it worth buying the MX? If I have an LX?

Its unlikely you'd ever need it.

 

[DieselGuy]

I can be a app tester

 

[oldestof11]

Is this available yet?

 

[Snipesy]

Is this available yet?

Soon™.

 

[sparkola]

Oh yeah

 

[sparkola]

Will this make it so I won't have to go in to have a new body side monitor module "programmed " at a the dealership?

 

[Snipesy]

Will this make it so I won't have to go in to have a new body side monitor module "programmed " at a the dealership?

I doubt it. The problem is I can't really give you those programs since theyre techniclly GM'S IP.

 

[sparkola]

But doesn't the purchase of the vehicle grant me the right to use their IP? They licensed the use when they sold me the product and in Illinois here, possession is 9/10ths the law ?

 

[Snipesy]

But doesn't the purchase of the vehicle grant me the right to use their IP? They licensed the use when they sold me the product and in Illinois here, possession is 9/10ths the law ?

Regardless no one is going to go to jail over sharing the software. GM, Bosch, Delphi, whoever.... They never defend their stuff. The best they can do is send strongly worded letters, for a while. Anything else would make a judge laugh.

 

[starspangled6.0]

Sounds amazing. Let me know if you need an app tester.

 

[Snipesy]

Will do.

I've took a small break for vacation but I'll be back at it soon.

If anyone is interested, the app runs on an HTTP based API which will be publicly available. Meaning you can remotely talk to the app from anywhere in the world if you set it up correctly (it has strong authentication so don't be worried about Russians getting in). It will work out of box on your local network, but if you want to talk to your car outside your local network you will need a VPN. I may in the future have a server which will do NAT punching for you so you can talk directly (or maybe indirectly).

I also recently got what I am dubbing 'Platform PIDs' working. Platform PIDs work a bit different from normal PIDs... instead of requesting some PIDs from the ECU it just taps into the signal your Instrument Cluster and others read. These will have a refresh rate of anywhere from 70 to 20 hertz. This is literally faster than your phone screen probably refreshes, so I doubt you will notice any lag. When you glance at your gauge the value or needle you see is what that condition is at that exact moment, not half a second ago.

 

[Snipesy]

This style of gauge I am particularly proud of. (lag is from gif)

 

[sparkola]

Any updates?

 

[Snipesy]

Any updates?

Lots of backend work... It's a massive project. Unfortunately nothing really interesting to really to show unless you like code. I'm hoping it will be easier to generate an entire map of the network, which you can use to directly control each module or what have you in a graphical interface. 80% of the work is probably going to be here, which is kind of unfortunate when it comes to showing off. Good news is once it makes everything else possible.

HTTP api got slightly replaced in favor of web sockets. Eaiser on data, more efficient. The HTTP API is still there but I'm not really sure how much of it is going to be used.

Network discovery hasn't been put in, so although the entire thing can work remotely it's actaully impossible to do that atm (I just haven't had to push to do it yet). If you've seen how spotify does it that is kind of what I am going for. Eventually I'd like a server to do NAT punching so you can do it outside your LAN.

Annnd I still need to implement OAuth so you know, Russians don't hack your car. That is a priority before I even think about distributing it to anyone.

Annd then I need to refine the UI. Finish the rudimentary guage view, make the table view at least... Add diagnostic management (cels, common stuff like that). From there Ill probably sell it in a sort of early access and start the BiScan to new app migration.

After that I will look at other things... Graph views is definitely up there, but at 40 frames per second this isn't as simple as it sounds so I will probably outsource this to people who know what they are doing.

 

[CraigoM]

How is the project going? Interested I trying it out