PDA

View Full Version : Is it that demanding also to release a game for Linux?



infestor
December 25th, 2009, 10:17 PM
i dont know the reasons why (a huge proportion of) game companies dont release also a linux installable version of their games (they do for mac os for instance). is it bacause:

-they just dont bother because of <2% market share of linux
-directx is a must, they just cant only use opengl

Zzl1xndd
December 25th, 2009, 10:47 PM
There are a number of Factors involved, however I would assume that it is the same reason that many of the Developers do not Make games for Mac.

Mostly a matter return on investment.

However many independent developers do make Linux Ports.

You might wanna check out http://lgn.linux-hardcore.com/ if you are looking for Linux Games.

pwnst*r
December 25th, 2009, 10:51 PM
how many of you still play quakelive after complaining the entire time that there wasn't a linux version?

Zzl1xndd
December 25th, 2009, 10:57 PM
how many of you still play quakelive after complaining the entire time that there wasn't a linux version?

Actually I Do, I also play Enemy Territory Quake Wars.

Dark Aspect
December 25th, 2009, 11:01 PM
-directx is a must, they just cant only use opengl

It would be as much trouble to port a Windows games to Linux that it is to port a game to Xbox/Playstation. Most consoles games are based on opengl so I don't really buy that as an answer.

pwnst*r
December 25th, 2009, 11:10 PM
Actually I Do, I also play Enemy Territory Quake Wars.

my point is, lots cry about it, but then either stop playing or forget about it. and that's with a totally free game...

ctrlmd
December 25th, 2009, 11:11 PM
it's not just the games also the applications
maybe it's business reason

Bölvağur
December 25th, 2009, 11:16 PM
I'm not sure if I understand the point of this thread... at all.
Is it just to get people talking by saying something very random? Or do you actually think the following:



-they just dont bother because of <2% market share of linux


Are you sure it is because there are few people or is it because they think the cost of porting is more than that investment will return?




-directx is a must, they just cant only use opengl
Why is directx a must?
I am not really sure it is compared to what I've been hearing them self be saying. But they do prefer it because of few things (Im more interested in hearing the explanation than me saying stuff that has already been said million times).



I'd be interested in hearing if you think you have any idea what you are talking about.

The Secret
December 25th, 2009, 11:19 PM
i dont know the reasons why (a huge proportion of) game companies dont release also a linux installable version of their games (they do for mac os for instance). is it bacause:

-they just dont bother because of <2% market share of linux
-directx is a must, they just cant only use opengl

The answer is in your question.

Zzl1xndd
December 25th, 2009, 11:30 PM
i dont know the reasons why (a huge proportion of) game companies dont release also a linux installable version of their games (they do for mac os for instance). is it bacause:

-they just dont bother because of <2% market share of linux
-directx is a must, they just cant only use opengl


The answer is in your question.



Do you have something to back that up? Considering the PSP, PS3, OSX and Nintendo's Wii all use OpenGL and not DirectX.

gsmanners
December 25th, 2009, 11:34 PM
The big game companies (like EA) have been brainwashed into believing that SecuROM actually prevents piracy. You really think people that gullible are going to be immune to the marketing power of a behemoth like Microsoft?

Mr. Picklesworth
December 26th, 2009, 12:06 AM
DirectX plays a big part here; Microsoft's strategy is of course to create a software ecosystem proprietary to Windows that developers are brought up to depend on. (And if you look at the "deals" they are giving to students these days, it's getting pretty ominous). With DirectX they've gone a step further than usual by having the actual hardware vendors build to its specs, too.

That's why I am excited about even the iPhone's developer community; these are all developers working with OpenGL ES. Thanks to Apple's ridiculous marketing ("oooh, technical words! We said API!"), the name is out there, too. It is looking more and more profitable for developers to know and support that technology.

phrostbyte
December 26th, 2009, 12:08 AM
Linux ports will happen when the following equation will hold true the vast majority of the time:

(Money made from a Linux port) - (Cost of producing a Linux port) > $0

Dark Aspect
December 26th, 2009, 12:19 AM
Do you have something to back that up? Considering the PSP, PS3, OSX and Nintendo's Wii all use OpenGL and not DirectX.

Hm.... Thanks for pointing that out, its not like I said that 3 post ago. (http://ubuntuforums.org/showpost.php?p=8558596&postcount=5)

sandyd
December 26th, 2009, 12:30 AM
i dont know the reasons why (a huge proportion of) game companies dont release also a linux installable version of their games (they do for mac os for instance). is it bacause:

-they just dont bother because of <2% market share of linux
-directx is a must, they just cant only use opengl
one thing comes to mind. - quake 4.

Dark Aspect
December 26th, 2009, 12:32 AM
one thing comes to mind. - quake 4.

Quake 4 never used Direct X and was created by long time supporters of Linux so I fail to see a point.

Zzl1xndd
December 26th, 2009, 12:55 AM
Hm.... Thanks for pointing that out, its not like I said that 3 post ago. (http://ubuntuforums.org/showpost.php?p=8558596&postcount=5)

Sorry Dark Aspect, Didn't notice you had already said that.

jflaker
December 26th, 2009, 01:12 AM
It is all about the mighty dollar! If there is no perceived profit to be made in developing a Linux port, most companies will not port to that platform....It was only in the last few years that MAC really went mainstream and now you can get even Microsoft apps for the MAC.

Give it time and you will start seeing more stuff for Linux

pwnst*r
December 26th, 2009, 01:32 AM
It is all about the mighty dollar!

when that's how you make a living, why yes, yes it is.

JDShu
December 26th, 2009, 01:39 AM
Linux ports will happen when the following equation will hold true the vast majority of the time:

(Money made from a Linux port) - (Cost of producing a Linux port) > $0

I'm picking at details, but technically its:

(Money made from a Linux port) - (Cost of producing a Linux port) >= $0

blueshiftoverwatch
December 26th, 2009, 04:47 AM
Microsoft's strategy is of course to create a software ecosystem proprietary to Windows that developers are brought up to depend on. (And if you look at the "deals" they are giving to students these days, it's getting pretty ominous)
I can attest to that. The school I'm attending has a deal with Microsoft where we're given access to a website where we can legally download tons of commercial Microsoft products for free. The only reason I'm running Windows 7 is because I got a free serial code. Most of the other kids think I'm weird for running Linux as my primary OS.

SSTwinrova
December 26th, 2009, 06:19 AM
I'm picking at details, but technically its:

(Money made from a Linux port) - (Cost of producing a Linux port) >= $0

No, I would argue that it's strictly >. If I'm a developer looking to make a profit (and that's it, no other influences on my decisions), I'm not going to spend the additional effort/time for something that would net me $0. Of course, IRL there are a lot of other factors in play that impact the decision, but for the sake of this limited scope, > $0 is a requirement.

phrostbyte
December 26th, 2009, 06:25 AM
No, I would argue that it's strictly >. If I'm a developer looking to make a profit (and that's it, no other influences on my decisions), I'm not going to spend the additional effort/time for something that would net me $0. Of course, IRL there are a lot of other factors in play that impact the decision, but for the sake of this limited scope, > $0 is a requirement.

It's worth noting that games, this equation really holds true. Especially when "cost of porting" is low due to the game engine being written with cross platform in mind. Eg: one reason I think World of Goo was so successful was because of it's cross platform support. It certainly got some free marketing here at UF it wouldn't have gotten otherwise.

Also things like Wine actually inhibit Linux ports from happening because game devs can just point at Wine if you want to get a game working on Linux. Blizzard is an example of this. :) Wine's game support is one of it's major strengths at this point.

jbrown96
December 26th, 2009, 06:52 AM
Games are not written in basic opengl. They produce an engine that uses some underlying toolkit. On Windows, that's an engine running on top of DirectX. On the game consoles, they use a toolkit developed by the console maker (nintendo, sony, etc.), which is built on opengl.
It's not really the same opengl that linux uses at all. There's all kinds of customizations that would make porting a nightmare.

The business equation is not
(Money made from a Linux port) - (Cost of producing a Linux port) > $0 Businesses don't make decisions just because it will bring in some profit; they develop a product A because it will get more profit than product B.
Game L is a linux port of existing Game A for some platform (windows/console)
Game B is a new windows/console game
(revenue from sales of Game L) - (cost to do Game L) = profit for Game L
(revenue from sales of Game B) - (cost to do Game B) = profit from Game B
Unless Profit from L > Profit from B the Linux port will not get made.
Those developers that spend time doing the port could be generating more revenue on another project; it's not like they sit idly.

There's really no incentive to support Linux. Develop costs are much higher (there are not many Linux game developers so labor costs are higher because of scarcity), and Linux has practically no marketshare.

Development costs are largely tied to the tools that developers can use. There aren't really any good tools for Linux game development, so it's much more costly.

Gizenshya
December 26th, 2009, 07:10 AM
I'm picking at details, but technically its:

(Money made from a Linux port) - (Cost of producing a Linux port) >= $0

No, I would say using only > would be more accurate. Overall effenciency would be 0% or it was even, and there would be no company incentive to do it, rather than do nothing. The exception would be to keep human capital within the company during a downturn. But that could even be done at a loss, because human caital lost to a competitor costs more in the long run than paying nothing to keep him loyal and onboard.

But, that doesn't take into consideration important factors. One is the time value of money. The movement from idea to sale of a game isn't instant. Therefore, any money made would at least have to be enough to offset inflation. But I'll leave that out of the equation. Just pretend it is taken into consideration in the cost section. This is usually accounted for in the cost, because future inflation rate is generally known within a reasonable margin of error.

Risk and profit margin. If the risk is too high (too few knowns, or too much variation in estimates), the project cannot go on. Simply turning a positive profit is not enough to warrant a game company to make an investment. These companies are often either owned by holding companies that are publicly traded, owned by companies that are publicly traded, are publicly traded on their own, or may one day be bought by a company that is one of the above. In other words, in one way or another, the influence of the market drives the products. Since there is an incessant drive in the free market to increase profit margin, effeciency, and other performance measures, any investment, after all risk and whatnot is taken into consideration, must be at least able to return some minimal estimated return percentage. This percentage is usually determined by marketing position in the industry. If the mid of the field is estimated to average 8% next (year, or whatever), and you are shooting for average, then 8% is your goal.

In this case, a more accurate general model would be:

(Money made from a Linux port) - ((Cost of producing a Linux port)*1.08 )>= $0

If true, invest, if not, don't.

And there are other factors to consider. One is, with so many users using both OS's, one false assumption in the equation I just posted would be that the introduction of a Linux port would not affect sales of any existing versions of a game. In fact, it would certainly reduce non-Linux-port sales of the game (necessarily, though perhaps minimally). This would also need to be taken into consideration. Though it might not be enough to affect other port sales tremendously... I strongly suspect that the vast majority of profits that would make the Linux port worthwhile would be directly taken from existing port sales. This fact (?) alone would be enough to kill most Linux port investment ventures before they even get off the ground.

There are many other factors, and all signs point to no. The sad fact is that with many, including myself, who like video games, it is worth it to have windows on standby just for that purpose. We buy the games anyway (basically, I agree with pwnst*r).

murderslastcrow
December 26th, 2009, 07:32 AM
Well, the key is to use development tools that allow you to port to every system very easily- to start with cross-compatibility in mind. I'm surprised so many games made for PS3/XBOX 360/Wii follow this mindset, but when it comes to OSes it's completely ignored.

So, for the companies who don't care if it's for anything but Windows at first don't bother to learn to use anything else but DirectX-only technology, it is hard. They have to port graphics frameworks and parts of their game engines manually.

If you use cross-platform compatible development methods, it's usually just a matter of exporting the binaries and packaging them correctly. Not really that hard at all.

Methuselah
December 26th, 2009, 07:36 AM
Games can sell on linux but the linux audience is a bit more discriminating and wary of things like intrusive DRM etc.
2D boy had quite good success with World of Goo on linux.
In fact, in their pay-what-you-want promotion linux users picked the highest prices on average:
http://2dboy.com/2009/10/26/pay-what-you-want-birthday-sale-wrap-up/
So, I think that indie games especially, can find a niche on Linux.

Methuselah
December 26th, 2009, 07:40 AM
BTW, a game that intends to be ported should generally abstract away details of input handling, sound, graphics etc.
One has to believe that is common software engineering practice so it shouldn't be a huge technical hurdle to convert a game from DirectX to Opengl/SDL especially since the APIs are largely feature equivalent.
I think it's more a matter of will because they don't consider the Linux market to be worth it.
And frankly, it will never be worth it since many linux gamers keep windows around for games anyway.

-grubby
December 26th, 2009, 08:26 AM
Linux ports will happen when the following equation will hold true the vast majority of the time:

(Money made from a Linux port) - (Cost of producing a Linux port) > $0


You win a cheese grater:

http://mattjokes.files.wordpress.com/2008/03/345px-cheese_grater.jpg

Enjoy.

earthpigg
December 26th, 2009, 09:15 AM
-they just dont bother because of <2% market share of linux

the 2% market share is not that relevant.

what is relevant is what % of that 2% is predisposed to even considering spending money on a computer game?

and what % of the ~90% Windows desktop market share is predisposed to spending money on computer games?

the 2% and 90% and 8% (OS X) are needed to answer those questions, but the X% market share is only the first question.

if 1 in 50 windows installs resided on a computer wherein the owner was willing to purchase computer games, and half of linux users where willing to purchase computer game... well, suddenly Linux has a very significant market for computer games.

but that isn't the case.

if you took 100 random Linux computers and 100 random Windows computers at random from places in the world where people actually have money to waste on computer games, i would be willing to bet a significant amount of money that more of those windows computers would be under the control of someone willing to purchase computer game software.

people willing to spend $0 on computer game software get exactly zero votes and exactly zero influence. their opinion and what they enjoy is simply... completely irrelevent. no one relevant cares. at all. software pirates unwilling to purchase, and people not interested in purchasing or playing video games are equally irrelevant.

people willing to spend $50 on the same get 50 votes and 50 influence.



the best hope a computer game company has at present is to release the game and hope to dominate the niche market. i purchased World of Goo, for example.

pwnst*r
December 26th, 2009, 03:43 PM
i purchased World of Goo, for example.

you got to name your price, so that's not a significant purchase. i'm betting *nix users wouldn't pay $50 or $60 for modern warfare 2 on linux.


big difference.

Zzl1xndd
December 26th, 2009, 04:51 PM
you got to name your price, so that's not a significant purchase. i'm betting *nix users wouldn't pay $50 or $60 for modern warfare 2 on linux.


big difference.

I guess there is no way for us to know, however it has been documented by a few independent developers that Linux users are More likely to buy and will spend more when the options are made available.

However, as with 2D Boy's World Of Goo, this may simply be a case of these developers being a big Fish in a Small Pond. But that isn't necessarily a Bad thing, I have bought about every Linux game I have come across.

However this wasn't the case a Year ago for me, it was only when a Realized that there are a lot of games I could get that I started to buy them, and That might be a bigger issue.

JDShu
December 26th, 2009, 05:36 PM
No, I would argue that it's strictly >. If I'm a developer looking to make a profit (and that's it, no other influences on my decisions), I'm not going to spend the additional effort/time for something that would net me $0. Of course, IRL there are a lot of other factors in play that impact the decision, but for the sake of this limited scope, > $0 is a requirement.

You're forgetting that that developing cost is included in the equation. This includes money paid to the developer or the "wages" you give yourself. Hence you will invest in developing something if you are able to break even. As I said, picking at details.

gsmanners
December 26th, 2009, 07:08 PM
Actually, a lot of game companies don't even develop their own engines. If you use the Quake 4 engine, for example, there's nothing stopping you from making a quick Linux port (aside from some technical problems like sound) with just a few developers. The main thing is you have to convince the shareholders that there's money in it, and good luck plowing through all the pro-MS FUD you hear from them.

Danimoth
December 26th, 2009, 09:03 PM
I think earthpigg summed it up nicely.

I understand why companies don't benefit from making linux ports and hope this will change once the market share grows. Until the, wine/dual boot/another computer with windows or a console will be fine. I am lucky in that wine plays all windows games i currently play.

Btw, i purchased World of Goo (not in the pay_what_you_want sale) because 1) it was a sparkling diamond of a game which is kinda rare lately and 2) because it had a native linux port.
I will also buy Heroes of Newerth(~= dota) which also has a native port.

earthpigg
December 26th, 2009, 10:49 PM
you got to name your price, so that's not a significant purchase. i'm betting *nix users wouldn't pay $50 or $60 for modern warfare 2 on linux.

no, i paid $20 shortly after the linux port was released.

i agree with you that there would be proportionally fewer Linux/Unix users willing to pay $60 for MW2 on Linux, but i myself likely would... well, if i didn't it would be because MW2 for the PC has no dedicated servers. that's a half-arsed port.



while we're on the subject, i have a question:
FreeBSD reportedly allows you to run native linux apps with an extremely low performance hit. i've heard it said that a binary of Firefox for Linux runs faster on FreeBSD than it does than Linux in the same DE. no, i cant source that nor do i remember where i heard it, but my question is...

is the performance hit still trivial for playing fancy games designed for Linux with 3d effects and the like on FreeBSD, or does it start to become painful?

Frak
December 26th, 2009, 10:51 PM
Do you have something to back that up? Considering the PSP, PS3, OSX and Nintendo's Wii all use OpenGL and not DirectX.
3rd party studios do the ports, such as Ryan Gordon.

phrostbyte
December 26th, 2009, 11:14 PM
no, i paid $20 shortly after the linux port was released.

i agree with you that there would be proportionally fewer Linux/Unix users willing to pay $60 for MW2 on Linux, but i myself likely would... well, if i didn't it would be because MW2 for the PC has no dedicated servers. that's a half-arsed port.



while we're on the subject, i have a question:
FreeBSD reportedly allows you to run native linux apps with an extremely low performance hit. i've heard it said that a binary of Firefox for Linux runs faster on FreeBSD than it does than Linux in the same DE. no, i cant source that nor do i remember where i heard it, but my question is...

is the performance hit still trivial for playing fancy games designed for Linux with 3d effects and the like on FreeBSD, or does it start to become painful?

I think the issue would more with FreeBSD's video drivers then anything. The only company I know of with half decent FreeBSD video drivers is Nvidia.

Techsnap
December 26th, 2009, 11:48 PM
1) Cost of porting the game over to an OS with barely any marketshare, they're going to get very little back and probably not enough to make any real profits.

2) People will just complain that it's proprietary [Don't comment against this, because they WILL]

3) The game will only work with lib and not with olderversionlib which is only provided by distribution with kernel version and gcc etc. It's not going to be practical for them to release it saying it will work on any Linux distro because there is no stable API.

4) Graphics drivers (opensource and proprietary) seem to be very temperamental on Linux and X11 is rubbish for the most part so they'll just have to spend more money with tech support going back to point 1.

phrostbyte
December 26th, 2009, 11:57 PM
1) Cost of porting the game over to an OS with barely any marketshare, they're going to get very little back and probably not enough to make any real profits.

2) People will just complain that it's proprietary [Don't comment against this, because they WILL]

3) The game will only work with lib and not with olderversionlib which is only provided by distribution with kernel version and gcc etc. It's not going to be practical for them to release it saying it will work on any Linux distro because there is no stable API.

4) Graphics drivers (opensource and proprietary) seem to be very temperamental on Linux and X11 is rubbish for the most part so they'll just have to spend more money with tech support going back to point 1.

If you code against OpenGL/OpenAL you really won't have many "stable API" issues. In fact one of the common complaints about OpenGL is it isn't developed fast enough. :P

Queue29
December 27th, 2009, 01:13 AM
Let's say it takes a year and a half to develop a mediocre game. How many operating systems are going to have to be targeted during that time to make the game hit a large enough audience?

Well, you'll want to hit the big distros: Fedora, Debian/Ubuntu, OpenSuse, Sabayon/Gentoo, and Mandriva. (and probably others) Now, how many OS versions will each of those distros go through? Fedora will go through 3, Ubuntu 3, OpenSuse 4-5, Sabayon/Gentoo 2-3, Mandriva ?? . Every single iteration will have different drivers, different kernels, constantly changing API's to target, and varying supported packages. All of which need to be accounted for during the development processs.

And then there's Windows/ OSX, where API's do NOT change, documentation is fantastic, OS's get major upgrades every 2-4 years, and there's > 97% market share held between the two of them, so a WAY bigger paying audiance to target.

If you're looking to make money, which platform would a sane person develop for?

Why would a sane, money making person even bother porting to linux?

gsmanners
December 27th, 2009, 01:50 AM
And then there's Windows/ OSX, where API's do NOT change, documentation is fantastic, ...

You obviously aren't a developer.

Queue29
December 27th, 2009, 01:52 AM
You obviously aren't a developer.

Actually I'm an employed developer.

Granted, the OS I develop against is RTEMS, not Windows/OSX/*nix.

earthpigg
December 27th, 2009, 02:17 AM
1) Cost of porting the game over to an OS with barely any marketshare, they're going to get very little back and probably not enough to make any real profits.

release enough details about the innards of the game to allow the community to port it for you. this worked with World of Goo. as i understand it, 2dboy spent $0.00 on the technical aspects of the port, and the blogosphere and digg-type sites did the marketing for them.


2) People will just complain that it's proprietary [Don't comment against this, because they WILL]

those people's opinions are irrelevant, as they aren't potential customers. they are willing to spend $0.00, so they get exactly zero vote. the game company has zero reason to even acknowledge these people exist.


3) The game will only work with lib and not with olderversionlib which is only provided by distribution with kernel version and gcc etc. It's not going to be practical for them to release it saying it will work on any Linux distro because there is no stable API.

include the exact version you want with the game, toss it all in /opt.

look at all these version numbers...


[chris: ~]$ ls /opt/dropbox | grep lib
_dbus_glib_bindings.so
_hashlib.so
libasound.so.2
libbz2.so.1.0
libcrypto.so.0.9.8
libdbus-1.so.3
libdbus-glib-1.so.2
libdirect-0.9.so.25
libdirectfb-0.9.so.25
libexpat.so.1
libfusion-0.9.so.25
libncurses.so.5
libpng12.so.0
library.zip
_librsync.so
libSDL-1.2.so.0
libsqlite3.so.0
libssl.so.0.9.8
libstdc++.so.6
libwx_baseud-2.8.so.0
libwx_baseud_net-2.8.so.0
libwx_baseud_xml-2.8.so.0
libwx_gtk2ud_adv-2.8.so.0
libwx_gtk2ud_aui-2.8.so.0
libwx_gtk2ud_core-2.8.so.0
libwx_gtk2ud_html-2.8.so.0
libwx_gtk2ud_qa-2.8.so.0
libwx_gtk2ud_richtext-2.8.so.0
libwx_gtk2ud_stc-2.8.so.0
libwx_gtk2ud_xrc-2.8.so.0
libz.so.1
zlib.so
[chris: ~]$


potential video gamer customers don't care about bloat. a video game is bloat by definition.


4) Graphics drivers (opensource and proprietary) seem to be very temperamental on Linux and X11 is rubbish for the most part so they'll just have to spend more money with tech support going back to point 1.

customer's responsibility and problem, just as it is on windows.... you don't call EA Games when you can't get any 3d stuff to work. that is your fault and problem, windows or Linux. list one of the requirements as "functional video card and drivers".


alternative to all of that: make the game work with WINE, and include the exact version of WINE you want your game to run on (but don't make it replace WINE, of course.). this is how google does some of their stuff.

jflaker
December 27th, 2009, 02:24 AM
Games are not written in basic opengl. They produce an engine that uses some underlying toolkit. On Windows, that's an engine running on top of DirectX. On the game consoles, they use a toolkit developed by the console maker (nintendo, sony, etc.), which is built on opengl.


ALL the more reason NOT to marry to a platform as your applications core, game or otherwise. Doing so limits your ability to support other platforms when something new or old becomes popular..Granted, it makes development quick and easy, but it prevents flexibility.

Frak
December 27th, 2009, 02:27 AM
You obviously aren't a developer.
You obviously aren't a developer.

Frak
December 27th, 2009, 02:30 AM
release enough details about the innards of the game to allow the community to port it for you.

That won't fly with any large game publisher. They have company secrets embedded in the source. Like hell they'll release that.

ad_267
December 27th, 2009, 02:36 AM
alternative to all of that: make the game work with WINE, and include the exact version of WINE you want your game to run on (but don't make it replace WINE, of course.). this is how google does some of their stuff.

No, Wine is one of the reasons there aren't as many commercial games for Linux. Too many people are happy to play Windows games using Wine.


That won't fly with any large game publisher. They have company secrets embedded in the source. Like hell they'll release that.

You don't have to release the source code to everyone, just people who will port it for free. Just get them to sign a NDA. Ryan Gordon is porting Aquaria for free. Great game btw.

RiceMonster
December 27th, 2009, 02:36 AM
No, Wine is one of the reasons there aren't as many commercial games for Linux. Too many people are happy to play Windows games using Wine.

Don't kid yourself, companies wouldn't port them anyway.

Frak
December 27th, 2009, 02:50 AM
You don't have to release the source code to everyone, just people who will port it for free. Just get them to sign a NDA.

NDAs are risky. Even though they're there, there's still a risk of a leak.

Arthur_D
December 27th, 2009, 02:58 AM
Let me fix that for you:
People are risky. Because they're there, there's still a risk of a leak.

Frak
December 27th, 2009, 03:10 AM
Let me fix that for you:
People are risky. Because they're there, there's still a risk of a leak.
True, but employees are way less likely to leak information. Employees can lose their jobs and reputation (to the point of being shunned from the industry). With the ability to monitor your own employees, it's more difficult to lose information. With 3rd party porters, there's much less risk. A leak could be made, and it could be nearly impossible to distinguish the source from the studio itself or the 3rd party porter.

Arthur_D
December 27th, 2009, 03:17 AM
Yeah, you got a strong point there. But it's not that uncommon either that the employees themselves give away info. Not sure if it would be that easy to find out who it was, either.

starcannon
December 27th, 2009, 03:20 AM
i dont know the reasons why (a huge proportion of) game companies dont release also a linux installable version of their games (they do for mac os for instance). is it bacause:

-they just dont bother because of <2% market share of linux
-directx is a must, they just cant only use opengl
Mostly it is a non-issue for the companies, they will make a killing just off a win32 application, the incentive is not great enough. It has already been shown that there is money to be made by releasing GNU/Linux compatible versions of the games, but either there is a lot of misinformation, or there is not a large enough profit margin, or a combination of both, that most game companies don't do it; or, they start out with it in beta, and drop it before release.

Frak
December 27th, 2009, 04:02 AM
But it's not that uncommon either that the employees themselves give away info. Not sure if it would be that easy to find out who it was, either.

The company has various means to see if someone either transmitted the secrets in house, or left with it it. In house, the company can track use of resources, security logs, audit logs, connection logs, and failure logs to track back to the source. It's difficult to track a 3rd party.

Frak
December 27th, 2009, 04:04 AM
Mostly it is a non-issue for the companies, they will make a killing just off a win32 application, the incentive is not great enough. It has already been shown that there is money to be made by releasing GNU/Linux compatible versions of the games, but either there is a lot of misinformation, or there is not a large enough profit margin, or a combination of both, that most game companies don't do it; or, they start out with it in beta, and drop it before release.
The classic quote is "If RedHat had to pull out of the Desktop market due to lack of paying users, what chance is there for us to pull a profit?"

Then they look at the marketshare statistics and deem that the port would cost more than the profits made.

earthpigg
December 27th, 2009, 10:41 AM
release enough details about the innards of the game to allow the community to port it for you.

That won't fly with any large game publisher. They have company secrets embedded in the source. Like hell they'll release that.

You don't have to release the source code to everyone, just people who will port it for free. Just get them to sign a NDA. Ryan Gordon is porting Aquaria for free. Great game btw.

i didn't say anything about full disclosure of source code that would require an NDA. release juuuust enough information to the public-at-large about how the binaries interact with the operating system to let them do it on their own.


No, Wine is one of the reasons there aren't as many commercial games for Linux. Too many people are happy to play Windows games using Wine.

i didn't say anything about requiring users to install or see or even be aware of the existence of WINE. roll it in with the binary, honor the license and release source code that includes any modifications made to WINE. gamer's dont care about bloat, we covered this.

Ji Ruo
December 27th, 2009, 11:00 AM
You win a cheese grater:

http://mattjokes.files.wordpress.com/2008/03/345px-cheese_grater.jpg

Enjoy.

Well hate to be technical but that's actually a picture of a cheese grater bla bla bla <formulas>

:p

Frak
December 27th, 2009, 07:05 PM
i didn't say anything about full disclosure of source code that would require an NDA. release juuuust enough information to the public-at-large about how the binaries interact with the operating system to let them do it on their own.



i didn't say anything about requiring users to install or see or even be aware of the existence of WINE. roll it in with the binary, honor the license and release source code that includes any modifications made to WINE. gamer's dont care about bloat, we covered this.
Why would you, though? As Phrostbyte said above, the Cost of producing the port minus the profits should be over $0, but as it stands, profits don't usually go over the $0 mark. See Loki Games.

ad_267
December 27th, 2009, 08:51 PM
Don't kid yourself, companies wouldn't port them anyway.

Maybe not big companies, but for smaller indie games I'm sure it's a factor. If we're going to get more Linux games we've got to start supporting the indie game developers that do support Linux. There's actually quite a few of them.


Why would you, though? As Phrostbyte said above, the Cost of producing the port minus the profits should be over $0, but as it stands, profits don't usually go over the $0 mark. See Loki Games.

The Loki story isn't that simple. See http://phoronix.com/forums/showthread.php?p=77586&highlight=loki#post77586.
Also, there definitely is some money in Linux games:
http://blog.wolfire.com/2008/12/why-you-should-support-mac-os-x-and-linux/
http://2dboy.com/2009/02/12/world-of-goo-linux-version-is-ready/

Linux Game Publishing also seems to be doing alright. I've just bought X3: Reunion from them. If you want more Linux games you should really be supporting LGP. The more copies of current titles they sell, the better quality the games they'll be able to get rights to later. The huge number of people playing Windows games using Wine don't help the cause at all.

jflaker
December 28th, 2009, 03:38 AM
Talking about proprietary code....Windows does it in the form of Dynamic Link Library (DLL) files. With the DLL the binaries of your API's can be hidden and your secrets kepts safe.

There are ways, if there is the will to allow development on top of your core.

Frak
December 28th, 2009, 03:48 AM
Talking about proprietary code....Windows does it in the form of Dynamic Link Library (DLL) files. With the DLL the binaries of your API's can be hidden and your secrets kepts safe.

There are ways, if there is the will to allow development on top of your core.
And .so doesn't count? A DLL is a shared resource, just like a .so.