So pointing out objective facts means I'm ignore objective facts? Okay cools. Or maybe there's a conspiracy that Microsoft used nano machines to rig the survey or something. I don't see you posting a link to anything objective or useful.


Are you just willfully ignoring what I've written or what? Windows Polaris indented to replace Windows S and the 32Bit will be replaced by some kind of virtual emulator. considering how good Microsoft actually is at emulation I don't think it will be an issue. Enterprise and pro versions won't be affected either. As for ignoring reality are you going to address the Trine 2 and Wine 2.7 and lest not forget remastersys. If Linux is so good at backwards compatibility then explain why I can't run those programs anymore. Why is Linux so bad at keep the old stuff working?

Let me repeat it one last time, because further repetition would be time wasting. Steam survey is not a fact that describes Linux gaming market or sales potential. You can't estimate anything from it. That's besides the point of non public methodology of its collection. So if you are going to bring it in any Linux market conversation, don't get surprised when it will be dismissed as irrelevant.

You can run virtual machines for many historic OSes, that's not backwards compatibility, that's virtualization. Let me know when you'll be able to run The Neverhood on Windows (without ScummVM), and then we'll continue this discussion about how Windows is good at backwards compatibility. Until then, it's pretty pointless.

So the amount of people using something has absolutely no bearing on it's sales potential? Okay here's an idea. It's 2001 and instead of releasing our games on the PS2 lets release it on the dreamcast instead because everyone own ones of those right. We'll make it big by releasing it on something hardly anyone is using.

I'm not surprised at all at this point. You could offer the Linux community a piece of software that would solve all their problems... for free and they'd sooner throw it away because it's closed source or not compatible with opensuse or something.

Microsoft wasn't too clear about it so we'll have to wait and see. Though doubt it will be a virtual machine. Not that it matters as long as it works.

Despite the Neverhood being a lot older then the programs I listed (1996 vs 2013 no that's not silly at all) I've accepted your challenge and see if I can get it working. Even if Scummvm is needed; from my own personal tests even emulators like dosbox run better on Windows then any of the

Not much, if most of those who use it for example will never buy your game, no matter on what platform. That assuming the number is even correct to begin with (which it is not with Steam survey).

I already said above, if you want to estimate sales potential, you need to analyze examples of existing games, and their percentage of sales per OS (or rather downloads per OS, since most stores including Steam sell all versions at once). Such analysis will be more to the point. Especially if you can find games similar to yours in genre and other aspects. You can find such info from various developers who published it, or from Humble Bundle. Neither Steam nor GOG publish this information.

I like linux, and i would use linux if it were an option for me at the moment. That said, it's far from being worth developing for at the moment. They'd basically have to write a wrapper for the windows/linux differences, and I can't imagine anyone justifying it at this time.

ScummVM will run it just fine, but Windows itself won't run the binary as is, because it's Windows 95 application with partial 16-bit code. MS removed support for such applications for good, so no matter what you'll try - it won't work. It will work in Wine on Linux, because Linux supports 16-bit instructions on 64-bit OS. It's just an example which demonstrates the point. Of course running it in ScummVM is the easiest way.

Right on the money my friend. But how would a wrapper solve the dependency issues. Say you make a game release it finish it and then move on to something else. How do you make sure said game continues to work in the future on Linux with dependencies constantly changing between distros and Kernel revisions?

They justified it for TW2 just fine, when Linux gaming market was smaller than today. And it was reported as success (TW2 port). So what changed in regards to TW3?

Basically, you don't use libraries that could end up deprecated, but have fun with that, but that's also a windows issue, too. I'm wondering when people will finally start writing OSALs to go along with the HALs.
They were expecting a bigger future for linux? Maybe they were also looking at the fact that TW3 is alot harder to port, given all the graphics junk that comes with it. There's a lot less portable code just to get some extra performance out.

I suppose TW3 can be harder to port, but there are porting companies that do the port with their own resources (Feral for instance), so CDPR wouldn't need to spend money on it (just share resulting profits from the Linux port). So why they didn't try that?

The time devs are spending coding is money spent. There's no such thing as a free lunch. Don't get me wrong, i'd love to see TW3 on linux, too, as that would really drive the legitimacy of Linux up. But, as it stands, it costs alot to port and support.

That didn't deter CDPR in the past, since they were clearly planning to do it. So I doubt the cost was the issue. More likely just bad planning.

You still need those library files to prevent deprecation


I had to look over this twice. How could so much be wrong with just two sentience?

1. Do you have anything other then buzz words or an examples of how the backwards compatibly is better is Linux Ditsros?

2. Of course they used closed source drivers, every body does. Do you honestly think hardware makers are gonig to share their source code with you just because you want them to and think they should?

3. Serious question but do you know what open and closed source mean?

A simple example. Take some laptop that's relatively old, and try running recent Windows on it. Most likely it will go bust, because there are no drivers for your hardware. Tough luck looking for them, they are closed, and manufacturer never cared to update them for newer Windows. Their idea was "buy new laptop, pay for new Windows too".

More often than not, Linux will run just fine on such computer, because drivers are open and they still support that hardware. Such cases are commonplace. So again, what has better backwards compatibility support when drivers are concerned? Surely not Windows as far as I can tell.

So let's close this discussion about backwards compat. Windows simply loses here.

Not quite so in my experience most programs come with the dll files needed to work. However they don't always do that. In the case of deprecation you have options like this
https://forums.mydigitallife.net/threads/update-v56-visual-c-runtime-installer-by-burfadel.72953/page-9#post-1385452

You have to be a member of mydigitallife but it's safe and free. From there you can download an AOI VC++ installer. In just a few second it will solves almost all dll related problems you might have. If Linux has something like this. I might be willing to try it.
The latest version of Mint still has issues running on my computer. Linux works well on old stuff I'll give you that but new stuff just forget it. It doesn't months it takes years before everything is working right.

But the question remains. What is open source? No examples or hypotheticals. What is open source. If I say that I have a USB stick with source code on and it's open, what will you see if you look in it?