1,516 karma · joined August 8, 2010
However, those dollars you seem anxious I don't give up still didn't really cover the time investment of dealing with Linux - not just the support requests, but getting the build environment setup and performing the testing and all of that.
For reference, the game in question did ~75% of units sold on Windows, ~25% on Mac, and some fraction of a percent on Linux. If I hadn't of released on Ubuntu I would have probably lost less $1000, gross.
If I sell on Windows (I do) and Mac (I do) then I have to support a certain range of OS versions and ongoing OS releases - even if that means (for example) I have to figure out how to 'notarise' a Mac executable so that a user doesn't have a big scary Security Warning pop-up. Not ideal, but fine. The challenge with Linux is that I would have to communicate against expectations - that I would have to make it clear that when I 'support Linux' it looks different to the support for Windows or Mac. I do genuinely think that 99% of Linux people get this, it's just the 1% that's maybe less forgiving of different standards.
For me, just personally and selfishly, passing the buck to Photon or Wine is an easier sell for my business.
But that was five years ago. I'm pretty sure Photon supports my Windows builds on Linux better than I was ever able to do with the native executables, and at least there's that.
Obviously necessary for Korean/Japanese/other non-latin alphabet languages, but my Spanish localiser does a fantastic job in this too. For example: Hellion -> Infernodrilo, Grondomut -> Cienomello, Gastroquin -> Explosapo. No, they're not direct translations (if that were even possible) but they fit with the visual design of the creatures.
There's a lot of trust you have to have with the people who make sure a creative product retains meaning and expression in another region and language, but also is adapted where it needs to. It's also a lot of fun to see something I made enjoyed by people I wouldn't ordinarily be able to communicate with. Irreplaceable indeed!
[0] - https://store.steampowered.com/app/654960/The_Eldritch_Zooke...
I think Factorio itself actually runs better on this laptop than that Factorio blog post does.
From what I've seen on Twitter, it looks like it really delivers, even in its early stage: https://twitter.com/IonizedGames/status/1397636481913610241
But if you're a developer who isn't making more than $50 million (like me!) you end up paying more than the enormous corporations. Unless you're on the App Store, where it's the opposite. It's funny how the focus is different here.
I have to perform a few double-takes as I read the site to try and parse the meaning out. I did duck into the samples themselves, and it's much the same there as well.
I guess everyone has different tolerances for what may technically be a 'temporary issue.' Yes, I have no doubt that all this engineering will lead to a better, more stable Unity, but for right now I actually have to somehow write real software and ship something to paying customers.
Here's just one concrete example: real-time point-light shadows in the Universal Render Pipeline. URP is the new 'default' way of rendering, the old renderer is being phased out, this is the new way of doing things - as you say; technical debt as they re-engineer. But how horrid has this process of eliminating technical debt been? You've got Unity proudly announcing that URP was 'production ready' in February 2019, but you still can't have a point-light cast real-time shadows. Currently that (let's face it, really basic) feature is planned for Unity 2020.2, which is still in alpha, but even if you download the alpha you can't access that feature yet, and they've announced they're not going to backport it, so bad luck if you're using 2019 or even 2020.1 for your game.
It's actually making me a bit sad to go into detail on what watching the URP fiasco, so I'll spare any other examples involving post-processing stacks, or 'camera stacking', or Ambient Occlusion, or...
Sadly, I wonder how much the last few years of very obviously aiming at Going Public has led the company and Unity as a software to this state. When the forums are full of "what is going on with Unity" threads and the company blog is full of "we've bought this cloud company" and "we're investing in AR for automative engineering", something isn't matching up.
The plus-side to all of this is I know my game runs well on lower spec hardware. There's a confidence that I have that it'll perform decently on 99% of computers and laptops/macbooks. If something runs slowly then I know I can't fudge things in the release build by assuming/hoping it'll run okay on the majority of computres, but I also know that if I do need to optimise something then it's not premature, unnecessary optimisation, which helps me use my time properly.