Blog post author here - Didn't expect a hack I did 5 years ago to show up here!

This project was a weekend project that I took on. I used to professionally reverse engineer cars for building self-driving research vehicles for companies.

Here's a demo of it using a VR headset that I made shortly after the blog post: https://youtu.be/jfWcgWdSK28

To answer the other questions, I made it work directly with PCs as a USB accessory. It doesn't require the Xbox gamepad in that variant.

I never bothered building out the FFB for the steering wheel - but it would be a nice addition. That said, slightly jacking the nose of the car up gave enough resistance on the wheels but it was still easy enough to turn.

It's still been a great tool for me to get seat time. Recently, I just taught at Spa-Francorchamps in Belgium and the Nürburgring in Germany. I used this sim rig setup to get properly acquainted with the tracks before teaching in person.

Last, along with a production company in NY, we used the same technology in a big rig truck to make it into a super realistic truck simulator for Pilot Logistics. I think there's media on it floating around the internet somewhere.

Cheers and happy to answer any questions!

This looks like a fun project, but that's about where I'd draw the line. A few issues right off the bat:

- In all such comparisons, people conveniently leave out build and research time/costs. I suspect that at least a few hundred hours went into this project, which would significantly increase the cost, if we account for hourly rates. Whenever I've run the numbers on similar custom hardware projects, I've come to the conclusion that if I'd simply worked the hours as overtime, I would be getting a much better product even if I bought high-end equipment, and quite a bit cheaper. So, these are strictly for people who enjoy the journey, and as such I find such comparisons insanely misleading.

- What exactly are the wheels doing, when you're turning the steering wheel? Is the car stationary in the garage, or is it lifted? I probably wouldn't want to spend dozens of hours turning my car's wheels in a stationary position, it puts strain on things.

- While on face value it seems like this should be about as real as you can get with sim equipment, I'd argue that it's actually fairly low- to mid-tier. Neither the steering nor the brake pedal will have any adaptive feedback, which is fairly easily obtainable with direct drive wheels and in some high-end pedals.

So all in all, quite a cool project, but not particularly useful or reasonable, IMHO.

I don’t like that the Xbox adaptive controller is necessary for this. Microsoft gets praise for how it enables alternate controllers - and it’s great for DIY - but that is only necessary because they use a cryptographic protocol to disallow allow unlicensed controllers in the first place. There’s no reason home brew or third party USB game pads should not work on the Xbox in the first place. Sony is the same. Nintendo at least, while they don’t ‘support’ third party pads, don’t use cryptography to disallow them.

There's been a recent article [1] about a very similar students' project at Darmstadt University. Interesting to see that the post referenced here precedes this by about five years.

[1] https://www.heise.de/en/news/Car-hacking-real-cars-turned-in...

I’ve been told that a local wealthy fellow turned his preferred racing car into a sim rig when he got too old/rich to risk racing it for real. It’s in his garage and is up off the ground. The car has been completely redone to make it a realistic sim machine—screens in the windshields, vibration motors, pedal sensors, the works. The person that told me that didn’t know how much they’d spent but it was in the six figures.

Reminded me of the Ridge Racer Full Scale recovery/restoration story: https://arcadeblogger.com/2022/11/20/the-last-ridge-racer/

In which a very limited edition version of Ridge Racer in arcades used a full-size Mazda MX-5 (technically the Eunos Roadster) shell and panoramic projection.

That's one way to get a workout. Without the engine, there's no power steering, right?

What about the simulated acceleration on the body by tipping the sim rig? That's going to be a tall ask in a real car sitting still.

interesting project indeed, but lack of steering feedback might already be a deal breaker for most. in terms of emulating the car you'd practice virtually for real-world driving, there is probably a nice market for easy-to-convert platform for getting input from cars.