Elite:Dangerous for Linux?

Sorry, gonna need a source on that one, I suspect we will find out all software has bugs though.
Fair enough:
http://arstechnica.com/security/2016/09/linux-kernel-security-needs-fixing/
http://www.zdnet.com/article/major-linux-security-hole-gapes-open/
http://arstechnica.com/security/201...f-millions-of-pcs-servers-and-android-phones/
http://arstechnica.com/security/201...-escalation-bug-ever-is-under-active-exploit/
http://arstechnica.com/security/201...ow-that-hacking-desktop-linux-is-now-a-thing/
http://arstechnica.com/security/201...scores-serious-concerns-about-linux-security/

Of course: it's difficult to truly compare the two OS's. What do you compare? "Discovered exploits"? That just tells you who finds more bugs.
But the fact is that MS has been throwing a *lot* of money for a long time into a lot of developers dedicated to security in their OS.

So I fully admit that my statement is opinion.

Well, I'm pleased you have a platform that works for you. I *don't* have a platform that works for me with ED. I use Linux for everything other than Elite. The only reason I have windows, and need a second platform (dual booted or not) is for ED and I'd prefer not to have to deal with the randomness I find that I encounter with windows.

I understand.

But you are an example of why FD *doesn't* need a LINUX client. You purchased the product anyway. I suspect that a great majority of LINUX users who would buy ED are in your boat... they already baught it. I don't think the same is true for PS4.

Of course, I'm speculating. FD should really study the matter and decide based on the data.

This strongly depends on what you mean by "free", the word is loaded. Some refer to the cost of the OS, which they consider to be Free of Charge, others refer to the freedom of Open Source, whereby one organization cannot obfuscate and hide code, or what can be done with a particular program. To go one step further, Elite is a highly specialized product that runs on (multiple) commodity platforms, to compare the financial cost of the game to the financial cost of the OS you use, isn't a fair comparison for either.
Why not?

As has been pointed out: you can simply factor the cost of the OS into the games you play on it and consider it the "opaque loading platform for the opaque games". You can still run LINUX as your primary OS.

I'm kinda disappointed that we have degenerated into OS bashing though, I just don't see a need for it. Windows is not a good OS for me, but it is for you. All I want is to coexist in game. I come from an age where seeing a game get ported to multiple platforms was the norm, to see ED make it to Linux would be perfect for me, but I don't see how this impacts or detracts from your experience on windows.
I'm sorry that you feel that I'm trying to bash LINUX. It's not my preferred desktop OS, but I'm not trying to bash it at all.

LINUX is a great OS. I have quite a few things running on one version if it or another; though my desktop is not one of them (but more power to the people who are running it as a desktop OS).

From my perspective, I think FD have demonstrated the ability to manage multiple platforms already and will continue to do so, I'm just at a loss to understand why some posters here seem to think that a future Linux port somehow detracts from their current game experience.
Limited resources + new constraints imposed by the new environment.

(generalize) *All* ports detract from the other versions. There are, of course, exception arguments to be made (when a port's sales bring in more resources than the port itself needs, etc).

If this were a LINUX client, porting to windows would detract from it as well.

And I'm not saying "it shouldn't be done", so much as I'm saying "here are the costs of doing it and I would be surprised if the return justifies them in this case".
 
Last edited:
This argument of "as a consumer..." starts with a soft of false premise.

Not really, it starts from the premise of which would I prefer, pay for an OS and pay for Elite, or get an OS for free and pay for Elite. It's no more complicated than that.

In reality I have paid for two operating systems (neither of which is Linux) and have paid for Elite once which I can play on both.
 
Last edited:
I would support this aswell.

Besides the point most windows software including Adobe runs on Linux just fine.

I even managed to get ED running smoothly under Ubuntu, however Joystick Support doesen't seem to work for me there and as i simply can't play ED without my extra bits of controllers its meh.
If there was a native Linux Version id instantly use it - because i could use my PCs Ressources soo much more efficiently and would not have either to dual boot or XEN-out a Windows VM just for Elite.

As for the Security Matter. Linux has a far superior Security design than windows has. In Win10 they even adopted alot from the Linux world on that side.
In the end the weakest Link in the chain for a secure system is still the User in front of the Screen.

As with the Mac & PS4 Versions ( seeing that the PS4 uses already a BSD Kernel ) its certainly possible to port and also has not the closed-behind-locked-doors philosophy Mac has with its OpenGL drivers - since current Linux drivers are top notch nowadays especially now AMD's thanks to the opensourcing of their driverstack.

Cheers!
 
Last edited:
Besides the point most windows software including Adobe runs on Linux just fine.

Erm... as near as I can determine, Adobe CC does not run on Linux just fine.

I've found lots of discussions asking "can you do it", but no answers saying "yes, and here's how." A couple of people seemed to have a brief bit of success running a very old version of Photoshop in wine/Mint, but that was about it.
 
Last edited:
Of course, if we were really serious about running Elite non-natively, we could just run Windows in a VM and use PCI passthrough.

I'll have to try this out - it seems like a reasonable middle ground without a proper Linux port on the horizon.

- - - Updated - - -

People will spend $300 on a case with a side window and neon/LED case lights, but can't bring themselves to pay for an OS. ;)
Honestly, I have never met a Linux user like this. In my experience, the vast, vast majority of the go-faster-stripes crowd are intense gamers who wouldn't consider Linux for even a fraction of a second because, well, it's not, for them, a proper gamer's operating system.
 
Oh crikey, actual debate! cool! (also thank you!)

I'm gonna skip the 0day links because, as you rightly point out, all we'll prove is that they exist.

But you are an example of why FD *doesn't* need a LINUX client. You purchased the product anyway.

Yes, very true. FD does not need a Linux client to further their share price. But, that doesn't mean I can't advocate for one.

I suspect that a great majority of LINUX users who would buy ED are in your boat... they already bought it. I don't think the same is true for PS4.
I think you are probably correct here too, but it was a choice between not having the content I want to enjoy or joining in, so I bit the bullet (and don't regret it one iota). I also think that the PS4 port is a promising sign that FD may have some resources available post launch to fund other platforms.


As has been pointed out: you can simply factor the cost of the OS into the games you play on it and consider it the "opaque loading platform for the opaque games". You can still run LINUX as your primary OS.
Indeed, I can still run Linux as my primary OS. But I can't access it while I am in game, because I've booted into Windows to access the content I want. Windows at the consumer level often has its costs hidden by buying a brand name machine, the costs of Linux are hidden from the user because large companies are subsidising the development in the server room (although canonical are asking for donations from private users downloading Ubuntu and have done for a while now).

Operating systems have essentially become commodities, especially in the so called cloud computing age, the OS underneath is largely irrelevant as the content (in this case Elite) is what is being paid for. The content could just as easily be a database or a CRM system running on Azure or AWS.

Also, I think that to factor cost of the OS into the equation is unfair to windows. By and large most users do not pay full price for it anyway and I can download many different Linux distros for no cost. So the equation always comes out worse for windows, but I also agree that the cost of windows is largely negligible compared to the hardware and the game. Which is why I call Elite "content" and I am willing to pay a premium for it.

Limited resources + new constraints imposed by the new environment.

(generalize) *All* ports detract from the other versions. There are, of course, exception arguments to be made (when a port's sales bring in more resources than the port itself needs, etc).

If this were a LINUX client, porting to windows would detract from it as well.

I think that this is based on the assumption that a team will not grow to match the workload but there is nothing that suggests to me this is the case. Of course, I fall foul of speculation here as well, as neither of us actually know how FD manages its teams, I think that they have the project management skills to manage it though.

And I'm not saying "it shouldn't be done", so much as I'm saying "here are the costs of doing it and I would be surprised if the return justifies them in this case".
And I dearly hope that you are (pleasantly) surprised!

The short of it is, I know getting a Linux client is a long shot, but fundamentally, I disagree that more ports automatically means fewer resources for core platforms, so I'm willing to ask publicly for Linux, even if I am ultimately unsuccessful.

o7
 
To read this thread you would think that rolling out a game on Linux is like moving mountains. It isn't. Surprisingly most developers who publish for linux do so without trumpets playing or the heavens opening.
 
Of course, if we were really serious about running Elite non-natively, we could just run Windows in a VM and use PCI passthrough.

I'd want more CPU cores before I did this, but yeah, I think it is the best solution right now. I'm just holding on to see if WINE delivers, but it isn't looking promising.
 
To read this thread you would think that rolling out a game on Linux is like moving mountains. It isn't. Surprisingly most developers who publish for linux do so without trumpets playing or the heavens opening.

it's quite odd to hear opposing opinions on not letting existing work that's already been done go to waste. I can see the opinion of just letting that other port die. That's quite valid. I dont see the point in trying to argue an anti-linux stand on why that existing mac code shouldn't be moved to linux. It wouldn't hurt windows. Any damage done in terms of increasing time of development by porting has already been done. You're not going to be negatively impacted if someone plays the game on one of the other supported platforms.

but meh. this thread is just devolving into a OS war. One which is funny because they're old tired and debunked arguments.
 
Yes, very true. FD does not need a Linux client to further their share price. But, that doesn't mean I can't advocate for one.
I fully support your right to ask for what you want.

I won't be on the forums whining if they do make one either. I will think it's cool.
(*note: The exception to this, and it's already true with consoles, is when I see a "well: this would be cool but the other platform won't support it"; in which case I reserve the right to be grouchy that I don't have the cool thing)

But I can see why it wouldn't be in their best interests to do so, and I don't suspect they will.

I think you are probably correct here too, but it was a choice between not having the content I want to enjoy or joining in, so I bit the bullet (and don't regret it one iota). I also think that the PS4 port is a promising sign that FD may have some resources available post launch to fund other platforms.
Agreed


Indeed, I can still run Linux as my primary OS. But I can't access it while I am in game, because I've booted into Windows to access the content I want. Windows at the consumer level often has its costs hidden by buying a brand name machine, the costs of Linux are hidden from the user because large companies are subsidising the development in the server room (although canonical are asking for donations from private users downloading Ubuntu and have done for a while now).
I can understand the frustration. I have a similar problem playing in VR. I love VR, but let's face it: ED is more down-time than play time; and I can't easily multi-task.

Operating systems have essentially become commodities, especially in the so called cloud computing age, the OS underneath is largely irrelevant as the content (in this case Elite) is what is being paid for. The content could just as easily be a database or a CRM system running on Azure or AWS.
I see a lot more nuance than is described there. I certainly am familiar with SaaS; but the underlying infrastructure matters in so many ways.

"it's stored data; doesn't matter what DB", well, except capabilities and performance.
"OK, well: it's SQL, but doesn't matter what SQL engine", well... that's not entirely true either. Yea, they all take SQL commands, but there are still many, many differences in how they are managed and how the do replication and the like.
"OK. Well, it's Oracle, but it doesn't matter which OS"... and so on and so forth.

Certainly: in the case of ED, this matters. The really good example of this is why we lost OSX support. The OS was not capable of offering the needed service (T&L?) to the application.

Also, I think that to factor cost of the OS into the equation is unfair to windows. By and large most users do not pay full price for it anyway and I can download many different Linux distros for no cost. So the equation always comes out worse for windows, but I also agree that the cost of windows is largely negligible compared to the hardware and the game. Which is why I call Elite "content" and I am willing to pay a premium for it.
I agree that the $ comes out worse for windows... I just think that, for most people who are playing this game, the difference is negligible.

I think that this is based on the assumption that a team will not grow to match the workload but there is nothing that suggests to me this is the case. Of course, I fall foul of speculation here as well, as neither of us actually know how FD manages its teams, I think that they have the project management skills to manage it though.
Agreed. There is the case to be made that more platforms actually equals more resources for the game in general. That's exactly the sort of data that FD should be crunching in making this chose.

And I dearly hope that you are (pleasantly) surprised!
It would be nice to have a surprise be pleasant for once :p

The short of it is, I know getting a Linux client is a long shot, but fundamentally, I disagree that more ports automatically means fewer resources for core platforms, so I'm willing to ask publicly for Linux, even if I am ultimately unsuccessful.
 
Last edited:
I'd want more CPU cores before I did this, but yeah, I think it is the best solution right now. I'm just holding on to see if WINE delivers, but it isn't looking promising.
On a modern bare metal hypervisor you'll lose only 1-3% in performance and have a 10% overhead for memory to run the VM. Why would you need more cores to compensate for a 3% loss?

- - - Updated - - -

The short of it is, I know getting a Linux client is a long shot, but fundamentally, I disagree that more ports automatically means fewer resources for core platforms, so I'm willing to ask publicly for Linux, even if I am ultimately unsuccessful.
I respect that. The performance advantage of Linux would be a huge boost to people who want more bang for their hardware. Here's an example of the same game on both Windows and Linux platforms.

[video=youtube_share;2pdEftFFG_I]https://youtu.be/2pdEftFFG_I[/video]
 
I respect that. The performance advantage of Linux would be a huge boost to people who want more bang for their hardware. Here's an example of the same game on both Windows and Linux platforms.
That's actually not my quote you are responding to.

That said: The video above
1) is atypical (Valve found a 20% performance gain)
2) represents the better optimizations in OpenGL over DX10 (FD is considering dropping DX10 support).

If there's an inherent speed advantage to the Linux kernel over W10 (or vice versa); it's going to be very minor. Both kernels are well optimized. API and driver optimizations can be significant; but the winner yesterday may not be the winner tomorrow.
 
1) is atypical (Valve found a 20% performance gain)
2) represents the better optimizations in OpenGL over DX10 (FD is considering dropping DX10 support).

If there's an inherent speed advantage to the Linux kernel over W10 (or vice versa); it's going to be very minor. Both kernels are well optimized. API and driver optimizations can be significant; but the winner yesterday may not be the winner tomorrow.
Atypical, yes, but a real world comparison nonetheless showing a 50% advantage. Both kernels are optimized, but I can tweak a Linux kernel minutely AND any Open Source API supporting it. Linux has always outperformed Windows in this regard as MS has to support a myriad of business needs with each OS release. The MS kernel is bloated with things the average person (and certainly gamers) do not need and cannot remove. DX makes gaming possible on Windows, but is as closed as Apples garden. Any standard completely under MS control is problematic long term.

This is the future and all it will take is one killer app to push it.

https://www.khronos.org/vulkan/
 
Atypical, yes, but a real world comparison nonetheless showing a 50% advantage. Both kernels are optimized, but I can tweak a Linux kernel minutely AND any Open Source API supporting it. Linux has always outperformed Windows in this regard as MS has to support a myriad of business needs with each OS release. The MS kernel is bloated with things the average person (and certainly gamers) do not need and cannot remove. DX makes gaming possible on Windows, but is as closed as Apples garden. Any standard completely under MS control is problematic long term.
Since you refuse to accept that an atypical example is useless.

Look: windows is faster:
https://www.youtube.com/watch?v=89G9qHrjS4A

And again: Windows is faster:
https://www.youtube.com/watch?v=ldEXGNtq1DA&spfreload=10

Oh look: Windows is faster:
https://www.youtube.com/watch?v=w9S5dZN--xE

The fact is that there's no significant performance difference between the Windows and Linux kernels.
 
Since you refuse to accept that an atypical example is useless.

Look: windows is faster:
https://www.youtube.com/watch?v=89G9qHrjS4A

And again: Windows is faster:
https://www.youtube.com/watch?v=ldEXGNtq1DA&spfreload=10

Oh look: Windows is faster:
https://www.youtube.com/watch?v=w9S5dZN--xE

The fact is that there's no significant performance difference between the Windows and Linux kernels.
:)
Sorry i just have to interfere when i see something too wrong... i just have to...
You cannot compare DX12 and OpenGL
Compare it with Vulcan ...and when it matures a bit that is....
thats all continue. Dont mind me ;-)
 
Last edited:
:)
Sorry i just have to interfere when i see something too wrong... i just have to...
You cannot compare DX12 and OpenGL
Compare it with Vulcan ...and when it matures a bit that is....
thats all continue. Dont mind me ;-)

Great thing about Vulkan too is that it's largely platform agnostic. I'm assuming DirectX 12 is largely Windows 10/Xbox One exclusive. That writes it out completely for me, so DirectX 12 capabilities are irrelevant to me. Assuming Frontier are sane, they'll be supporting the Windows 7 OS for quite some time yet.

If we wanted a true apples to apples, it would make more sense to do a Vulkan on Windows comparison to Vulkan on Linux, since we can't do DirectX 12 on Linux, Windows (other than 10 and Xbox One), Mac, and now PS4.

OS overhead isn't all that much of an issue for me. Where it counts most is likely in BOINC. For video games for me, it's largely irrelevant.
 
Last edited:
:)
Sorry i just have to interfere when i see something too wrong... i just have to...
You cannot compare DX12 and OpenGL
Compare it with Vulcan ...and when it matures a bit that is....
thats all continue. Dont mind me ;-)

I agree with you (mostly).

A previous poster had put up a single, cherry picked instance of program that ran faster on an Linux build than on a Win7 build as a supposed sign of things to come regarding what ED would be like on Linux. I felt that I was unable to convince him that his single example was not actually indicative of anything; so I resorted to counter examples.

Where I don't quite agree with you is on the "when it matures a bit". I agree that, when it does, the comparison should be re-made; but if we are discussing expected real-world performace, we must deal with what is available now and in the real world.

IBM Used to have an add for the PowerPC that showed how much faster the PowerPC chip was than an x86 by showing measured performance of current x86 chips and projected performance of future PowerPC chips.

And again: I'm in no way trying to disparage Linux. I would assert that any performance difference between a Windows and Linux PC, assuming the program in question was well optimized for each platform, would generally be minor (and major differences would be quirks that could go either way).

- - - Updated - - -

Great thing about Vulkan too is that it's largely platform agnostic. I'm assuming DirectX 12 is largely Windows 10/Xbox One exclusive. That writes it out completely for me, so DirectX 12 capabilities are irrelevant to me. Assuming Frontier are sane, they'll be supporting the Windows 7 OS for quite some time yet.
Maybe. They have already announced abandoning of DX10 and Win32.

If we wanted a true apples to apples, it would make more sense to do a Vulkan on Windows comparison to Vulkan on Linux, since we can't do DirectX 12 on Linux, Windows (other than 10 and Xbox One), Mac, and now PS4.
Depends on what we mean by "apples". If we want to compare real-world performance we should compare what real companies would really chose. Right now, for windows, that doesn't generally appear to be Vulcan; but rather DX11 or DX12

OS overhead isn't all that much of an issue for me. Where it counts most is likely in BOINC. For video games for me, it's largely irrelevant.
Agreed. It's not really an issue for most. OS-imposed limitation (See OSX's ED client) can be, however.
 
Last edited:
Great thing about Vulkan too is that it's largely platform agnostic. I'm assuming DirectX 12 is largely Windows 10/Xbox One exclusive. That writes it out completely for me, so DirectX 12 capabilities are irrelevant to me. Assuming Frontier are sane, they'll be supporting the Windows 7 OS for quite some time yet.

If we wanted a true apples to apples, it would make more sense to do a Vulkan on Windows comparison to Vulkan on Linux, since we can't do DirectX 12 on Linux, Windows (other than 10 and Xbox One), Mac, and now PS4.

OS overhead isn't all that much of an issue for me. Where it counts most is likely in BOINC. For video games for me, it's largely irrelevant.

Yes, from business perspective it is sensible to go where you can get most customers.
And since Vulcan can work on both Linux and Win10 it seems logical for game dev to go that way.
And as a fellow dual booter myself for few years now, i did noticed a spike in game availability in recent months or a year....
And i am not talking about Wine - i dont use that - i mean really native support.

And when it comes to average Granny, they dont care what OS they run. All they care is if browser works and they can have some pictures saved....and occasional web based game as a high point of their gaming experience :).
I am spreading Mint like a plague in my PC fixing zone of activity :-B
So yeah we are in for a interesting times i bet.

Also you can clearly see what is valve doing.
They went Vulcan way and coupled it with their own console ;-)
It cant be more obvious really... if only i was rich....i would become investor.
 
Last edited:
Back
Top Bottom