The Rivet: First Look At This Potential Zwift Competitor

To be fair, using the term ‘Zwift Competitor’ might be a bit…early. Like looking at a pile of ingredients on a table and calling it a cake. But damn, it looks pretty. And everyone likes eating cake batter anyway. Also, it was responsible for a fair bit of distraction from my planned morning. Sorry MARQ in-depth review.

Anyways, back to that pile of ingredients. Lama shot me a text earlier this morning asking if I’d seen the Rivet cycling dude or not. After an initial Googling fail landed me upon a bad-boy heartthrob cycling t-shirt store, I found the actual site, which in turn got me distracted watching his Twitch feed. The one-man-band of a developer was doing a 24hr ride to kick-off the launch of his platform, named The Rivet. All while his social media manager (aka…friends), supplied him with a Greek salad and noodles on the bike to keep him fueled.

At first I was impressed by the game’s graphics, based on the Unreal engine. But to be honest, I think I’m actually more impressed listening to him answer the sprinkle of occasional questions into the chat window from the assortment of 1,000 people that have stopped by to watch. Almost every question prompted a flood of ideas he has to implement, or things he doesn’t quite think are perfect in the game yet. All of them well thought through.

At present, the biggest ‘selling’ point, if you will, is steering and braking, plus of course the graphics you’ve already seen above. The biggest downsides is essentially missing everything else. But so was Zwift when it first started, at least compared to today. So after watching and listening in the background for an hour or so, I decided to jump in and give things a whirl.

And if text and photos are more your thing, then onwards into the rest of the post!

A Quick Ride:

As of current, the application is split into two pieces: The main app that runs on a desktop computer (Windows or Mac), and then the companion app that runs on your smartphone (iOS or Android). You’ll need both pieces in order to make it work. In my case I used the Mac version on my computer, and then installed the iOS app on my iPhone.

Once that’s done you’ll have to do a bit of geekery to display the IP address of your computer and then insert that into the phone app, so it can find its desktop friend. At which point you can connect the two. Obviously, down the road that’ll happen automatically. And ideally, any absolute dependency would go away. Though the current phone requirement is for steering. But more on that in a second.

As far as creating a username goes, that’s not there yet either. Basically this is kinda like going into an arcade and just playing on the machine as-is. The platform supports pairing to smart trainers and power meters, with plans for other sensors shortly. They’ve also already lit up and tested smart trainer control Tacx trainers.

So I dragged out a Tacx NEO that was sitting nearby and got it paired up to the companion app. You’ll scan for nearby BLE power meters.


Then I needed to get my phone on my handlebars. That’s because the app uses the phone’s accelerometer for steering and control of your line. In my case I just used my Quadlock mount/case on my bike and I was good to go.

With that, I started pedaling. There’s honestly not a ton of other configuration options at this point, aside from a slew of graphics related ones.

There’s a singular course right now, which is a hilly and forested mountain route, which is what I’d be riding. And at present the only camera view is as seen here, due to Patrick’s (that’s the founder/one-man-band) concerns around steering and having a camera in the wrong place could cause confusion.

There’s really only two core functions you’ll need to master, the first being steering. It’s pretty easy, you just turn your handlebars. Of course, if you’ve got a front wheel block, that might be slightly more difficult, but that’s a problem for a different day.

I didn’t have too many issues with steering, but I also didn’t do any descents, which might be more sensitive to it given the faster speed. Of course, the concept of steering isn’t new to indoor training. Tacx had it years ago in their Tacx Desktop suite and their steering utility. But beyond their apps there wasn’t any mainstream adoption. Still, I think it’s inevitable that steering will make it into mainstream platforms at some point.

The second is braking, which you can do by simply tapping the gigantic red ‘BRAKE’ button on the smartphone app:

In the event you fail at the only two things you’ve got to do (brake and steer), then you’ll eventually fly off the road and subsequently crash. When you crash you simply reset back to the middle of the road perhaps 50-100m prior to the point you crashed.


Otherwise, you simply ride. The graphics look great, even when not put at their highest levels. You can see them below, as well as in the video up above.

Of course, the downside is there is no other gamification or additional features at this point. The focus has primarily been on the physics of the game and the underlying engine. Plus of course the scenery of the 11.8KM route.

It’ll be interesting to see where the platform goes. For example, during his 24hr livestream, the game was actually running remotely in a datacenter on a remote computer there (‘in the cloud’). You can see it looks fantastic in the Twitch stream, of course, at present that’s not required. But the idea of it is something that is quickly gaining momentum in the gaming industry, with projects like xCloud and Google Stadia. He noted he’s excited about this trend and “totally convinced that’s the way everything is going”, so here’s to hoping that the company invests earlier in that space (they’ve already applied to be part of Google’s Stadia program) as a way to potentially have a leg up on their competitors which haven’t really dug into it yet. But at the same time, he pointed out that the business/monetization side of some of these platforms isn’t super clear yet either.

And he also accurately notes that there are elements of ‘friction’ that make these setups a bit more complex, such as indoor trainers that need to connect via Bluetooth or ANT+ from your living room to a cloud platform. Obviously, he’s making it work today for the demo/livestream. But whether or not that scales at the speed required in full-on high stakes racing remains to be seen.

Back to the example of when Zwift creator Jon Mayfield first posted Zwift screenshots to the Slowtwitch forums, versus what Patrick posted today, I’d say Zwift was a bit more refined at that point (including things like summary screens, computer generated racers, and more). On the flipside, Rivet has the advantage of a far more cohesive trainer industry compared to 6-7 years ago, and a far better educated consumer base to work from compared to then. Not to mention easier access now to investment funding that ultimately drives employing developers.

Going Forward:

As the title of this section implies, it’s all about what’s ahead of the company. As it stands today, no, they don’t really have a completed product. But they do have an idea. And frankly even more importantly – they have a dude that’s fanatical about the development of the platform. Seriously, the guy’s mid-way through a 24hr ride on it as part of his one-man-band promotion of it. Yes, you can watch it live here for another few hours.

Realistically what The Rivet needs is investment funding and a gaggle of developers. Once they have that, they can go forth on adding all the features they need to be a legit Zwift competitor. Stuff like adding in multiplayer support (planned by winter), as well as creating the gamification draw that Zwift has nailed. Certainly, one doesn’t have to gamify everything, but if the company is looking for investor money, then realistically that’s where investors want to ensure the platform has something that’s going to keep significant quantities of people addicted to it (and paying for it).

Like I said in the video, nobody is going to go and cancel their Zwift membership today to jump to Rivet. Though, I think for geek-inclined folks it might be fun to take the course for a lap. And for investors to judge whether there’s enough of a business case to help fund it. I know from talking to numerous people in the industry, there’s huge appetite to help fund Zwift competitors – especially ones that are positioned like this without significant overhead.

With that, thanks for reading!

DC Rainmaker:

View Comments (49)

  • I find it funny that we have steering, in the real world when riding above a few kph you actually counter steer to go around corners (steer right to turn left), this is so natural most folk are not even aware of this, bet they don’t implement that as part of the steering, my turbo bike still sits on the original Tacx steering rig (Tacx iMagic) as I’ve no room to get a front wheel on my bike, used it once in Tacx, switches steering off and never used it again #WasteOfTime

    • Hey Mike!

      It's Patrick here. Thanks for your comment :).

      If you read above I made some comments about steering in The Rivet that might interest you. You points are well taken though - steering is not for everyone (which I have definitely learned in all the feedback I've received in the launch). At risk of promising everything to everyone I think in time we can have our cake and eat it - an option to steer and an option to not. We're not building a rocket to the moon here it's just a video game!

      I also agree with you that if we are to make steering a 'thing' it has to be executed absolutely perfectly. What we have now is just a v1. If we win you round I'll know we'll have nailed it!

    • I agree with you. Steering will be a tough one to simulate. You countersteer to initiate a lean, then slightly steer back into the corner as you lean. As you say, the countersteer is practically imperceptible when you do it, but fundamental to negotiate turns. In my mind, the leaning and lateral acceleration are the sensations that you'd want to replicate from the real world, but I don't see how you'd accomplish that any current stationery trainer. The idea of steering your bars in the direction of your turn to track a roadway seems a bit gimmicky if your goal is hyperrealism. I'd be fine with it as a tool to choose your path at a junction, but I don't see it playing out to negotiate curves in a roadway.

    • I actually bought the animations from someone but somehow didn't notice the hip rocking at the time ( think I must've only looked from the font). I then tried to fix it before realising the documentation for 3dsmax (3d modelling + animation software) is thousands of pages long. So it is what it is for the moment I'm afraid!

  • Will be interesting to see if they get to the point of a Garmin Connect IQ app in place of the phone.

  • I don’t know that I see that steering feature to be a plus. On a real bike, you wouldn’t turn the handlebars like that, you’ll use body English and lean into the turn. With a winter of Zwift, I already find some bad habits working their way into my riding (not looking far enough ahead, not paying attention to traffic, etc). Can’t wait to launch myself over the bars trying to execute a gentle turn at speed.

    Can you turn the steering off?

    • Hey Matthew!

      Patrick from The Rivet here :).

      That's interesting about developing bad habits. I spent 24 hours on the virtual bike and successfully rode to work the next day though, so it luckily didn't effect me too much!

      But to address your point - right now you can't turn the steering off. In time that will definitely be added though as many have asked for it! I can see a scenario in which some 'game modes' require this ^^ 'real' steering where you have total freedom over the road, some use 'lanes' you can move between by steering, and some keep you fixed on a pre-defined path.

      I am actually going to be experimenting with the lane steering idea pretty soon (once some of the bugs and UX issues are fixed). I'm going to livestream myself testing it, get feedback and then release it for the community to try ASAP. My approach to this is going to be super open and community driven. I've set up a subreddit for people to keep up to date if you ever hang out over there : .


    • Right now you can't, though the founder talked a bit during the livestream on everything being a bit of a 'let's see how it goes' sort of thing. For example, sensitivity being one of those things.

1 2 3

By continuing to browse the site you are agreeing to the use of cookies