Linux and Mac versions of Frontier games.

[Fost]

I wanted to continue this here, since it's bad form to pollute the other thread.

continuing from these posts:
http://forums.frontier.co.uk/showpost.php?p=866&postcount=52
http://forums.frontier.co.uk/showpost.php?p=868&postcount=53
http://forums.frontier.co.uk/showpost.php?p=879&postcount=55

I just want to say: I do really want to see Mac and Lin having more and more support, but I think arguing (al least in the case of Linux) that there is any financial gain to be had is not going to help. Right now I'm a small indie developer, and the only reason my company has failed to deliver a Mac and a Linux version of our game is that we ran out of time - which I've always been ashamed of, because we wanted to support Mac and Linux from the start, and we feel like we're letting everyone down. Now, certainly for shareware, Mac has a good market, but the Linux market is pretty much non existent as far as I can tell. Other developers I've talked to quote sub 5% of their sales coming from Linux users, and yet a lot of support issues - by that, I don't mean people keep mailing in with problems, I mean patches have to be made to support a wider range of distros and configs.

Now, please don't get me wrong - I still have the goal of supporting Linux in the future dear to my heart, but I won't be doing it with the expectation of any great financial gain. I think most other developers have access to similar sales figures, so trying to stress that there's a massive audience out there is going to fall on deaf ears and do the cause no good whatsoever.

In 2006, Linux is a household name, everywhere you turn now.

Certainly not in the UK in my experience. People may be using linux based devices without even knowing it - that doesn't mean they know a thing about it. the only people I know who are even aware what linux is are in the IT industry to some extent - sysadmins etc. When you say 'household name' I expect you to mean that if they took at survey out in the Sun newspaper, "We asked a hundred of our readers if they knew what Linux was!", then a household name would get more than 20%. I would be unsurprised if the above survey was run (Unlikely, since the people who write that newspaper probably don't even know what Linux is) that the result would be 0%. Sad, but that is what I believe is true right now.

I implore you to look at ataricommunity.com, Neverwinter Nights 2 forum, specifically the multiplatform support thread.. The clamor to have NWN2 ported to Linux and the Mac have made this thread one of the top 5 longest threads on their site..

I have seen those forums, and specifically in the case of NWN - that was again a labour of love for two of the programmers who worked on it in their spare time - it was not a business decision (as far as I could tell from interviews etc), but this I think is the crux of what I'm saying. Linux users do have the biggest advantage of community, and a very vocal one at that. Sometimes I think that gives them a false impression of the size of their market. All I'm saying is, the goal here is to try and convince more developers to release on Mac and Lin. Now, Mac's you may have a point financially - although like I mentioned before, mainstream developers I know said they were expecting everyone to dual boot on intel boxes in the future (another worrying potential trend amongst game devs there.) Linux just doesn't seem on paper to offer any financial reason to support it as far as I can tell. Only reason I want to support it in the future is for the hell of it!

Elite 4 will certainly come out for Windows, so this post, and posts like it, might not speak to you, and shouldn't concern you.

You read me wrong -it does concern me, because I believe developers should be aiming for better cross-platform support on all games. The only point I'm trying to make, is that Linux users wading in and saying there's a massive market out there is perhaps not the best tactic, since it's not hard to find sales figures that show otherwise. Like you say - community is the strongpoint for linux. Perhaps a better tactic is just to say - we may be small, but we will spread the word better than any other group. What if all linux users created fan websites, blogs etc for an upcoming game? - but always had a banner that just said they hoped it would release on linux. Maybe it wouldn't work, but at least the Linux community would be vastly more visible to the developer.

And guess what operating system the majority of us are playing on..

Are you saying you think most people in the world who are still playing Elite are playing it on Linux?
I personally find that unlikely in the extreme, even if it would be cool. What makes you think it's true? It would be cool though...

 

[VictorT]

*sigh*

This thread should [ultimately] only be for people who would like to voice their support for a linux version.. At least as long as one hasn't really been announced.

Alas, if you insist on a discussion, I'll indulge you this once.

Right now I'm a small indie developer, and the only reason my company has failed to deliver a Mac and a Linux version of our game is that we ran out of time -

So you see? even among indy devels you have a sense that there are indeed linux gamers out there. It's certainly wise for this discussion to exist for Frontier Developments to see, since time is still on their side as far as making a decision on what platforms to release on. Let's both hope their decisions are for the marketplace's better.

and yet a lot of support issues - by that, I don't mean people keep mailing in with problems, I mean patches have to be made to support a wider range of distros and configs.

Ok, let's get one thing straight. This is exactly what I mean by "uninformed" discussion.

There is only one Linux. There are many distributions of this Linux thing; Redhat and Fedora and Debian and SuSE and Mandrake, et al, but this is irrelevant, since any Linux binary will run on any distribution of Linux. Linux is Linux. The distributions are distributing Linux, not writing it. Linus and friends [and literally anyone talented enough] are writing Linux. Period.

If you make a program for Linux, be it an application, utility, or game, you need to only compile it once [per processor architecture]. If the target machine the program is being installed on has the same CPU and has installed the required libraries it needs to link against, then the program will run..

And to be clear. Every reasonable distribution of Linux packages SDL, Xorg, OpenAL, and any other library you'd need to write or play a game. Even the unreasonable distributions can have those libraries installed by hand if need be, and when your game/application/utility is satisfied, guess what. It'll execute.

SDL, Xorg, OpenAL, KDE, Gnome, OpenOffice.org, the GIMP, Firefox, Thunderbird, PERL, Python, Apache, Wesnoth, ad nauseum.... these aren't written by or for specific distributions, hell they're not even written for specific operating systems, these are all independent projects that do not rely on any one unique factor. Redhat, Fedora, SuSE, Mandrake, Debian, Gentoo; the distributions go to the homepages of each of these application and game projects, download the latest, pack it nice, and redistribute it. Anyone who wants can go to these applications homepages, download each individual application, and run it. That 's the whole idea of the Linux-From-Scratch distribution [if it could be called a distribution, it's mostly some howtos and you do all the footwork], for example.

So having to support multiple distributions and multiple configs is literally a strawman argument. The requirements for Elite 4 should be:

Linux 2.6.12 or greater kernel.
SDL 1.2 or greater [SDL is just an example]
OpenAL 1.1 or greater. [OpenAL is also just an example]
[and hell, maybe qt or gtk, or both if they wanna be extravagant]
x86 or x86_64 2.0GHz or better [and if they're really brave Linux/PowerPC]
1GB of ram or better
2 GB of free drive space

NOT

Gentoo,Debian, RedHat, but not Fedora, SuSE, or Mandrake Linux
ThisPackage.rpm, ThatPackage.rpm, SomeObscurePackage.rpm
x86 or x86_64 2.0GHz or better
1GB of ram or better
2 GB of free drive space

Simply making it run on Linux, any distribution, and letting all the little hacker children try to get it to work on their new FooBar1000 distro that 5 people in the world use, is enough. If they want to come ask Frontier Developments for support, it's entirely up to Frontier if they want to provide support if it's somehow distribution specific.

One version of Elite 4 [client and server? ] for Linux, and patches can apply to it alone.

let's burn some more strawmen,

in the case of NWN - that was again a labour of love for two of the programmers who worked on it in their spare time - it was not a business decision (as far as I could tell from interviews etc), but this I think is the crux of what I'm saying.

This too is not fact. Without Atari's specific go ahead, Bioware would never have been able to let their developers make a linux client for any reason.

This is why the multiplatform thread exists on Atari's pages and not Obsidian's. Obsidian is perfectly willing to port NWN2 to linux, and has told Atari as much. But without the publisher's go ahead, they can't.

because I believe developers should be aiming for better cross-platform support on all games. The only point I'm trying to make, is that Linux users wading in and saying there's a massive market out there is perhaps not the best tactic, since it's not hard to find sales figures that show otherwise. Like you say - community is the strongpoint for linux. Perhaps a better tactic is just to say - we may be small, but we will spread the word better than any other group.

Fine, in the UK, we may be small, but we will spread the word better than any other group.

..which i believe i said in my last post.

You are claiming that you want better cross-platform support on all games? let's end this discussion on that point, since we essentially agree. If you believe my approach isn't up to your snuff, approach it in your own way and ask for multiplatform support as I am, instead of attacking me and exclaiming that it'll never happen, and without one word from Frontier themselves. Lets avoid the self-fulfilling prophecy there, shall we?

Thanks.

Are you saying you think most people in the world who are still playing Elite are playing it on Linux?
I personally find that unlikely in the extreme, even if it would be cool. What makes you think it's true? It would be cool though

Check out alt.fan.elite.. give it an ask..

 

[Fost]

This thread should [ultimately] only be for people who would like to voice their support for a linux version.. At least as long as one hasn't really been announced.

Yup, you're totally right. I should say, that is me too, we're on the same side, I'm just suggesting that focussing on the community strengths (which you did also do very well!) is the better thing to do in my opinion. Please don't take my questioning of some of your points as being an anti -linux or anything as that is totally not the case. I just don't believe there is a (sizable) market for linux games still, but the good news is there are other reasons to support it, and I feel if linux gaming is to get anywhere then these are the things that need to be concentrated on (which to your credit, you were doing).

So you see? even among indy devels you have a sense that there are indeed linux gamers out there.

No that's just it, I don't. I have a sense the market is pretty much non-existant, and I have sales figures which fellow developers have been kind enough to show me that say if I was purely a business man, it would be a waste of time. Luckily, I'm not just a business man, and I really hope to support linux in the future, but that's more of decision of the heart. Attacking this comment seems pointless though - it's the reality(although if someone has some amazing sales figures to prove me wrong there, I'd be more than happy to be made a fool of! ), so ignoring it or trying to claim there's something there that isn't is just putting our heads in the sand. All I'm saying is, Linux users always come across as evangelists - rightly so, they've got a great point to make, I just think we could make it better by accepting some of the realities.

Ok, let's get one thing straight. This is exactly what I mean by "uninformed" discussion.

Aplogies, I appear to be winding you up - that was not my intention You may be correct, although I am 'informed', perhaps not by the right people though. Again, I can only base this for myself on other shareware authors private comments who seem to think Linux versions have caused equal or more post-release work than their windows counterparts, despite accounting for far less of the sales. I actually did get this impression too from the NWN (1) linux forum, which pretty much seemed to be discussion about how to get it to work. You're right here though, this is one thing that without having been through the experience, I'm not really qualified to talk about it - sorry.

This too is not fact. Without Atari's specific go ahead, Bioware would never have been able to let their developers make a linux client for any reason.

Ok, I have no facts to give you - however, it is my opinion, having worked in the games industry (including Infogrammes/Atari) that nobody in management would say 'Hey, a linux version is a great idea' It seems to me to have been lead by two fantastic programmers - who went out of their way to make it happen. Yeah, Atari green-lit it, but if two top programmers on a game tell you they want to make a linux version in their spare time, then why wouldn't you let them! That's probably where we need to give our support - programmers who want to support linux, because they are the ones leading the charge, not I believe, anybody in management at any games company anywhere.

instead of attacking me and exclaiming that it'll never happen, and without one word from Frontier themselves. Lets avoid the self-fulfilling prophecy there, shall we?

Please! I really must apologise, you've got the wrong end of the stick- I'm not attacking you, and I'm sorry you even got that impression. I didn't claim it would never happen. I just wanted to tell you that the impression I get is game dev companies (at least, the people in them that make the major financial decisions) generally believe there's no worthwhile market for linux games -hey I could be wrong, hell, I'd LOVE to be wrong, but I think it's better to deal with that reality head on - because there's other stuff we could be emphasising that could be a benefit if developers embrace the linux community. You are doing that better than I would be able to, and perhaps I should have commended you on that from the start rather than just singling out one part of what you said.

I know linux users have to put up with a lot, but sometimes you might want to avoid being too defensive

Check out alt.fan.elite.. give it an ask..

Yeah, good idea, I'll stick a post on, and btw, I've read every post on there for about the past 3 and a half years if not longer. Lately though, it's not the most active of places, and I'm not sure it would be a very accurate survey, but it'll be an interesting thing to read.

 

[VictorT]

My final reply to this..

...what we have here is; a classic chicken and egg conundrum.

I just don't believe there is a (sizable) market for linux games still, but the good news is there are other reasons to support it, and I feel if linux gaming is to get anywhere then these are the things that need to be concentrated on (which to your credit, you were doing).

Thank you. I appreciate that you recognize that I am not trying to blow smoke. Still, it is apparent that you haven't been a big part of the Linux game community, since you seem to believe there isn't [a sizable] one. We agree that there should be, I happen to know that there is, I highly recommend following http://linuxgames.com and http://happypenguin.org for a month and decide again if you don't see a community there.

if I was purely a business man, it would be a waste of time. Luckily, I'm not just a business man, and I really hope to support linux in the future, but that's more of decision of the heart.

Chicken, meet Egg. Egg, Chicken. Becase there is no sales does not mean there is no demand.. There is no formula to calculate the number of gamers playing on Windows because they have to, instead of because they want to. If you were a businessman I'd recommend you develop for Linux [and the Mac] purely to have first mover's advantage. When you support the Linux community, the Linux community supports you [assuming your software is worth it ].

And speaking of the Linux community,

Linux users always come across as evangelists - rightly so, they've got a great point to make,

Dear sir, nobody is "evanglizing" Linux because they're trying to make any points great or poor.

Linux isn't written by a company that's going to send out Sales Engineers to Frontier Developers to "suggest" they use Linux, LLC's API. Linux is written by the people, for the people. If we in the Linux community are going to convince publishers and authors [like yourself] to develop for Linux, we must do it ourselves as a community. Even as many people will post to say "Gee, we think you should develop for MacOSX", at any time Apple can sit up, take notice, and send out sales reps to Frontier and get the bend of ears we would never be able to do as a community.

So consider retiring the word "evangelism" and using the word "democracy" instead. Linux gamers don't represent any corporate "brand" or idea, or ideal. We represent ourselves. We approach publishers and authors ourselves the only way we can [via forums, email, happening to know someone], or no one else will.

It's worked in our favor for IBM, HP, Intel, AMD, nVidia, ATI, Samsung, Hitachi, PeopleSoft, Oracle, RealNetworks, EMC, Mentor Graphics, Pro-E, Novell, Motorola, SAP, PalmSource, Dell, Sony, and so on and so on and so on..

Trying to get the game industry to make a few games for Linux?

...Of all things, that should make reasonable business sense.

, but if two top programmers on a game tell you they want to make a linux version in their spare time, then why wouldn't you let them!

Simple, because your upstream publisher won't greenlight it. Another business reality.

but sometimes you might want to avoid being too defensive

Yes, well I couldn't agree more. Lets close this discussion on that note. You have many fine points and many valid concerns. I thank you for engaging me and allowing me to fine hewn my message. And I will implore you to do some more research yourself; perhaps immerse yourself a little more in the Linux community maybe? Get involved with your local Linux Users Group, we're in every country with electricity on Earth [including Bhagdad], I know there's one in your area, they could use a voice like yours, outspoken and down to earth. If you feel they're not going in the right direction when you get there, feel free to guide them in discussions that you feel are more on track with the realities of the software development world, your experience would be worth much to everybody.

And if you absolutely positively feel you must have the last word on this matter, go to town.. I've said my piece. Thanks.

-m

 

[Fost]

...what we have here is; a classic chicken and egg

Yes, that's exactly it. There isn't a sizeable commercial market, but there's no reason there couldn't be if a few more people would support it.

I highly recommend following http://linuxgames.com and http://happypenguin.org for a month and decide again if you don't see a community there.

That's exactly what I mean. That's not helping, not one bit - Yeah, there's a fantastic community there, of pretty much freeware developers - they even have a tick box for whether or not the game's source is available. Look, I totally appreciate what you are saying there I really do, but pretend you are a business man, looking at that! You seem to think I have no interest and don't follow what's going on, but that's completely the opposite of what I've been doing, I think there could be better progress here, but I get frustrated when the Linux community shoots itself in the foot by pointing out things that will warn business men away. It's like trying to convince someone to work on Windows by showing them gamehippo.

I am sorry this might be winding you up, but everything else you are saying is spot on, and it really frustrates me seeing that comment as I think it's undoing some of the good work.

And if you absolutely positively feel you must have the last word on this matter, go to town..

Ha, that old gambit, your jedi powers won't work on me...


So, here's what I think will help: Forget the top end - Managers, business men, etc, they just aren't interested unless you can tell them their profits are going to be up 50% (and you can't in the case of Linux). They might have to greenlight something like you say, but the idea has to come from lower down - programmers who get interested in Linux and fancy the challenge of making their games work everywhere and for everyone (and that's the basic egalitarian idea that interests me actually).

So, let's pretend we are working at Frontier Dev? What should we be doing to make games cross-platform capable? There's a lot of stuff they are probably already doing just to get games on consoles. I'm guessing though, that their current cutting-edge engine is DX based right now for PC and Xbox360. Being able to s***(w_a_p is filtered?) renderers out is going to be important for potential Mac/Lin versions, but also for PS3 which uses GL. If you're wanting to support PS3, so you need GL support, then surely that's a good argument for being along the road to Mac and Lin ports?

I'm wondering though, what's the best way to author shaders that target GL and DX? Is there a way it can be done so you don't have to rewrite them? Anyone had any experience doing this with cg? In theory cg should work, but I've no idea if it's any use in practice. Maybe everyone just goes the hlsl/glsl route and takes the time hit?

 

[starman]

I wonder how that worked out?