He spent more or less all of the profits on his previous release on this new game. From the videos I've watched about the dev, he is somewhat of a perfectionist. It would likely rankle him to release a subpar experience on a given platform, which given the economic incentives, it would likely be.
Unreal maybe, but it also wasn't as viable a choice 8 years ago.
Source: I use unity professionally at my day job.
MINOR SPOILERS: Some of the later puzzle mechanics rely on playing with lighting and perspective in a way that seems to me would be difficult to implement in a stock engine.
I seem to be downvoted to death - the comments weren't meant in a disrespectful way - just wondered why the gargantuan effort with a DIY engine. The game itself looks really interesting.
Given that Jonathan Blow is in charge of the project, it's almost certainly because existing engines were inadequate for the task.
From what I can tell, Rust is the most ambitious Unity project out there. The Witness is substantially more technically ambitious than Rust. Additionally -from what I read on the Rust devblogs-, the Rust guys spend a large amount of time working around Unity's limitations, missing features, and parts of Unity that are just buggy. (Granted, the buggy stuff generally eventually gets fixed, but until it gets fixed (if it ever does), you have to work around it.)
FYI - I have only used Linux as my main laptop for the past 15 years. But he's right. I had really hoped that Steam Linux would be based on Android or something and build that friendly layer.. I'm still not sure why they went with Debian.
You appear to be completely unaware of the economics of game development. That coupled with the arrogant tone of your comments really makes you sound foolish.
Yes, I am arrogant by the way, I learned not to be afraid of standing tall with my own point of view.
Incorrect. The context is one of an indie developer trying to bring a product to market. This is similar to being a one-man startup (before hiring / contracting more staff). It is absolutely essential to make good decisions about scoping the project, marketing the project, and selling the project.
It is very common for technically-minded people to get into a pattern of addressing technical objections point-by-point, without realising that the technical objections are not the problem per-se, but just a symptom of a more important concern or problem.
If the goal is Linux advocacy, then nit-picking the tech objections isn't really that effective, unless you want to leave a rebuttal for future searchers. What would be more effective would be to try to understand why Linux was not right for this developer, on this project, at that time.
> Yes, I am arrogant by the way, I learned not to be afraid of standing tall with my own point of view.
Standing up for your beliefs is not the same as arrogance. Not realizing this is a sign of arrogance, and arrogance is not a good thing. It makes it harder to develop an accurate model of the world, and thereby makes it harder take effective actions.
EDIT> Part of the hostile reaction you're getting is because of the way you dismissed Jon Blow for not being able to install Linux. This is one of the best game developers out there. It's like calling some Nobel Laureate stupid because she decided not to use some particular niche app that you care about.
There are some Linux sales figures kicking around [1] and they generally show that the Linux share is under 3%.
[1] https://www.gamingonlinux.com/articles/sales-statistics-for-...
The Witness has has a peak of less than 5,000 players at one time so far.
The top two games (and basically the only active "pro" games at the moment), which together have about one million players playing at any given time are both carefully tuned to run on toasters according to Valve's hardware survey.
I like Valve's current micro transaction model. You can buy cosmetic items, but that's it. I can't speak for DOTA2, but in CS:GO they don't impact the game negatively like they did with TF2. Trading from free drops also allows kids and poor people to participate in the economy, which allows them to make more money for Valve than begging mom to buy a $60 game would.
I tested this with Quake III (a game from 1999!). My 2011 iMac could not render the game at 120fps, it managed somewhere between 50 and 70. I bootcamp into Windows, and I get a rock solid 120fps.
Depending on region and target audience, Macs show up between 1.5% and 3.5% of total player base in tracking (see Steam and Unity hardware surveys, and I also see those numbers in our own games), and these have been the same for 10 years. For comparison, Linux is between 0.5% and 1.0%. Most Macs are laptops with Intel GPU which are simply too weak for most current games. The Mac OpenGL drivers have a legendary reputation for being crap since they are controlled by Apple, not the GPU makers (although, native Intel GL drivers are not much better). Sometimes the reason is simply that the additional required customers support for Mac costs money and thus the development effort can't be justified.
I'm a daily Mac user and like it pretty much, but for games I also have a proper PC, attached to the TV. Most games I play there would run at most with 5fps on my MacBookPro.
I have high hopes for the beginning for the convergence (that is already happening with RemixOS) of ChromeOS and android during google i/o.
Frankly, Ubuntu 15.04 and 15.whatever_the_last_release_was have been really, really, really bad. I say this as a gamer, Linux power user, and sysadmin who's been using various flavors of Linux for ~fifteen years and who's quite happy managing his Gentoo Linux boxes, but really happy when he finds a largely-hands-off distro, because that means he can recommend a distro to his friends.
What's more, if you are using Ubuntu 15.latest, Steam doesn't work out of the box. It won't start. You get no error messages or anything. You have to move a bundled copy of -IIRC- glibc out of the way before Steam will work. Reports on the interwebs indicate that the moved file comes back when the Steam client updates itself, so this isn't a permanent fix!
Maybe Valve can't be arsed to target anything other than SteamOS, but it's not a good sign when the company that's -arguably- doing the most good for Linux gaming out of anyone out there doesn't ship a client that works on the latest version of one of the more (if not the most) popular Linux distros.
I'mma have to make a note to go back and remind myself about how dynamic linking and .so versioning works on Linux. I had thought that the tools to make this shit work correctly were widespread and relatively easy to use, but the state of Linux gaming leads me to question my memory. :/
I'm not sure that's going to incentivize anyone, especially if they're making plenty of money on the platforms they chose to release on.
Spending your money on existing Linux games in the same genre (and encouraging your Linux friends to as well) is more likely to incentivize devs. At least they can see that a Linux port sells, and they can find sales stats to assess whether the market for a particular genre is large enough to be worth the investment.
Inaction (i.e. waiting only) is not easily measured by game developers without spending money or doing additional work on market research or surveys. Waiting by itself incentivizes nothing.
Fundamentally, though, some kinds of customer aren't worth the trouble. I'm not going to sell Lamborghinis to people who don't really care about flashy cars. So, why would I expect to sell AAA polished games to people who won't install a gamer OS, out of principle?
I'll let you pull your own estimate out of the air for how much extra development time/money supporting a decent cross-section of popular Linux distributions will take. Oh, and don't forget to add in time for testing it, too.
This may change if Valve can make the Steam Box happen, but right now it sure doesn't sound like a sound business decision to bother looking for those last few extra bucks from Tux. Honestly I'm surprised anyone releases anything for Linux or for Macs now that I look at the numbers; I suspect we have the rise of cross-platform general-purpose game engines (which this game doesn't rely on) to thank for that.
"Oh here's my game on linux, please address any questions or problems to those shitty hardware vendors"
No, he's just going to release it on a platform where he can guarantee compatibility and functionality with minimum effort on his part.