No Linux support yet?

(Andreas Wolff) #1

Just came back after a one year break and still find myself forced to an OS I do not use actively.
I would be really great if you at ZWIFT could/would support Linux as OS!

Luckily I still have an “oldish” Windows machine so I can use if for now …

Hoping for many supporters on this topic!

Enjoy riding!


(Paul Allen) #2

Last I heard Zwift has no plans for Linux support (search the old forums for the threads on it).

(David Gibbs (Red Rider)) #3

It boils down to their user base.

Every platform they support has R&D costs and a complete duplication of QA effort … so to support Linux they would have a very large expense. Considering the percentage of their user base that uses Linux (I’m guessing less than 1%), it just isn’t cost effective.

Have you tried running Zwift in WINE?

(Daren Chandisingh [Vision]) #4

But their user base boils down to what they support. Take Android, for example. You could say their Android user base is close to zero, but that’s because they are only just in beta for support.

(David Gibbs (Red Rider)) #5

What I meant is what portion of their user base uses Linux AT ALL … LOTS of their customers use Android, but a small fraction of their user base uses Linux.

It makes business sense to invest in an Android client, as lots of their customers will be able to use it. It doesn’t make sense to invest in a Linux client because very few of their customers will be able to use it.

(Daren Chandisingh [Vision]) #6

OK, potential user base. I get that. It’s a pretty small number, and this is where a proprietary engine has a big impact on the equation I expect. With a different one, like Unity perhaps, compiling for a Linux system might not be so big a deal. (Disclaimer: I’ve no idea how good Unity support for Linux actually is, just postulating.)

(David Gibbs (Red Rider)) #7

As a professional software developer, I can assure you that the R&D costs are trivial compared to the QA efforts.

Zwift currently supports 4 major platforms (Mac, PC, iOS, & Android). Each of those platforms has multiple variations (Win 7, 8, 10, IOS 8-12 + ATV, etc).

Adding a new platform is gong to be a major undertaking, especially since Linux has so many variations (distro, hardware, modifications, etc).

Trying to test & support all those variations is going to cost a lot of money.

Even IBM (who, I just read, is buying Redhat) only supports a small subset of Linux distros on some of its enterprise applications (which is the space I work in).

(Daren Chandisingh [Vision]) #8

Heheh, you don’t need to tell me about QA or testing varied platforms. More than 30 years in this game now. I moved away from desktop and embedded programming, but even Web stuff is bad enough. /o\

But in theory that’s one benefit of building on top of a platform like Unity, isn’t it? The platform acts as an interface/buffer and you just use the API to code stuff up. The platform’s mainly responsible for driver and peripheral support, not the stuff built on top of it. Same with stuff like OpenGL or DirectX.

I do think a platform like that – assuming it supported ANT+ for example – would be easier to develop for and test, compared to a bespoke in-house one. Certainly in my current world, using third-party frameworks and other open-source components takes a lot of the headaches away from me, and I don’t have to keep reinventing wheels.

(David Gibbs (Red Rider)) #9

I get your point … but there’s still underlying native code that has to be verified on every platform.

In the end, it’s a matter of where Zwift wants to focus their efforts … new features, functionality, and bug fixes, for 99% of their user base or adding a new client platform for 1% of their user base.

(Daren Chandisingh [Vision]) #10

Oh yeah. I’m not expecting them to change tack and re-implement Zwift in Unity now. That was a technology choice to be made 5 or more years ago. In the early days after those initial discussions with Eric or venture capitalists, that would have been the time for Jon to say “OK, my engine has been great up to now, but let’s do this in something else now we have the funding for it.”

Maybe. Certainly it’s probably far too late in the game to change things now. I’m more discussing what might have been a better approach with the benefit of hindsight. Maybe easier to support more platforms. Maybe easier to develop mapping tools or other systems. Maybe easier to support better graphics. Maybe easier to support more jerseys in games.

Maybe we’d even have pedals! :smiley:

It’s all hypothetical and they are where they are, for better or worse.