ExtremeTech's "Why Gaming Sucks on LInux"

[joshuapurcell]

The major problem with gaming on Linux is the fact that OpenGL isn't used near as much as DirectX from how I see it. I would hope that Vista's changes will make more developers move towards OpenGL, but that will only work so far. OpenGL needs to be seen as being a much easier to use feature-filled replacement for DirectX, and Vista's mistakes will only go so far in that. If I had to pick the most important area to help Linux gaming it would be adding features and useability to OpenGL, but that's out of my reach at the moment.

[niniendowarrior]

Well the thought's been percolating in my head since GDC '03 when I went there to give a talk on the current state of affairs for Linux Game Development and Gaming on Linux (Albeit a clumsy one- NEVER change up your planned talk at the last moment, even if you think Icculus is going to be interested in joining you as a guest speaker... ). I'd walked about, talked with all the Indie bunch about Linux prospects, talked with the hardware vendors, talked with the middleware vendors. Lots of interest, even from Sony, but NOTHING ever came of it. VIA still owes me one of the Glory reference platform machines with Linux drivers for the latest Chrome series... (Heh... )

So now I can blame the current state of Linux gaming to you. Just kidding. I guess you thought things were going to change at that day, huh? We may need another trip to GDC! Though at that time you should announce some large coup like some huge title on Linux.

Indeed. I'm hoping that the title I've recently seen that's coming for PS3 ends up matching the looks it's got- if so, I'm hoping that I or someone else can manage to talk them into a reasonable deal on the Linux segment of the space so that someone can actually make money selling it. I do know that it's the first major title other than Doom3/Quake4 engine titles that is OpenGL, etc. based in some 2-3 years running now. I'm gunning for pipe dreams I might be able to manage to pull off somehow...

That's awesome stuff Svartalf. Let's hope that things work out.

[Svartalf]

I think it's an interesting prospect although the work is clearly equally hard or even more difficult. I do hope to at least stimulate the thought.

Well the thought's been percolating in my head since GDC '03 when I went there to give a talk on the current state of affairs for Linux Game Development and Gaming on Linux (Albeit a clumsy one- NEVER change up your planned talk at the last moment, even if you think Icculus is going to be interested in joining you as a guest speaker... ). I'd walked about, talked with all the Indie bunch about Linux prospects, talked with the hardware vendors, talked with the middleware vendors. Lots of interest, even from Sony, but NOTHING ever came of it. VIA still owes me one of the Glory reference platform machines with Linux drivers for the latest Chrome series... (Heh... )

Well, you have a point and I would think that if somehow the choice of what to port is a bit conservative, you might snatch some nifty titles and add a tiny bit of bonus content like what Konami and Silicon Knights did for MGS Twin Snakes. Again, in the hopes of starting something.

Pipe dream, everyone. It's okay to dream...

Indeed. I'm hoping that the title I've recently seen that's coming for PS3 ends up matching the looks it's got- if so, I'm hoping that I or someone else can manage to talk them into a reasonable deal on the Linux segment of the space so that someone can actually make money selling it. I do know that it's the first major title other than Doom3/Quake4 engine titles that is OpenGL, etc. based in some 2-3 years running now. I'm gunning for pipe dreams I might be able to manage to pull off somehow...

[niniendowarrior]

LoZ:TP is probably a bad example for this (Lucifer really would be ice-skating to work if that one got ported to anything NOT Nintendo controlled... ), but I would be inclined to agree with the sentiment. The only problem is that until the current generation of consoles, you typically had a lot of programming close to the metal going on in the game- they would be major undertakings that nobody in their right mind would take on unless the game was based on an engine with interchangeable rendering backends, etc. PS3 titles are going to be mostly portable. Wii ones will probably be. PS2 ones are going to be evil, as will GameCube ones, unless it was done for all console types out of the gate (Anything with an X-Box/X-Box 360 version is a likely candidate as it'll be largely a Windows to Linux port at that point.). I probably wouldn't mind too much if something Viewtiful showed up (1 or 2...) or Burnout Revenge showed up on Linux.

Well, I know how bad an example LoZ is but it's a thought of getting some brains to think about the idea (Hey, LoZ did appear on CDi! *snicker* *snicker* ). I know what you mean. I do think that people would be up for a little Viewtifulness and Burnout stuff on Linux. I think it's an interesting prospect although the work is clearly equally hard or even more difficult. I do hope to at least stimulate the thought.

But, the story I told you about the Windows to Linux porting is going to likely be even worse for console titles to Linux- because they're going to be gunning for numbers commensurate with THEIR market share.

Well, you have a point and I would think that if somehow the choice of what to port is a bit conservative, you might snatch some nifty titles and add a tiny bit of bonus content like what Konami and Silicon Knights did for MGS Twin Snakes. Again, in the hopes of starting something.

Pipe dream, everyone. It's okay to dream...

[Svartalf]

You know, this was just a thought, but if Linux porting houses decided to do exclusive console games ported to Linux. You'd at least have a game that Windows users don't have, and it would be a sort of exclusive kind of thing. The problem becomes a matter of translating it properly for the PC from consoles. Wishful thinking and highly unlikely example might be The Legend of Zelda: Twilight Princess Linux port. This might make things a bit more interesting in the Linux arena, could it?

LoZ:TP is probably a bad example for this (Lucifer really would be ice-skating to work if that one got ported to anything NOT Nintendo controlled... ), but I would be inclined to agree with the sentiment. The only problem is that until the current generation of consoles, you typically had a lot of programming close to the metal going on in the game- they would be major undertakings that nobody in their right mind would take on unless the game was based on an engine with interchangeable rendering backends, etc. PS3 titles are going to be mostly portable. Wii ones will probably be. PS2 ones are going to be evil, as will GameCube ones, unless it was done for all console types out of the gate (Anything with an X-Box/X-Box 360 version is a likely candidate as it'll be largely a Windows to Linux port at that point.). I probably wouldn't mind too much if something Viewtiful showed up (1 or 2...) or Burnout Revenge showed up on Linux.

But, the story I told you about the Windows to Linux porting is going to likely be even worse for console titles to Linux- because they're going to be gunning for numbers commensurate with THEIR market share.

[Svartalf]

So.. those titles were as good as it gets, and nobody with an interest in Linux gaming has $100K? What about Shuttleworth or somesuch?

We've got a few more that are more to speak of coming our way. I don't know if Mark would be interested in that aspect of things yet- he's still trying to win the baseline battle for the desktop. While Games ARE important to that thing, ease of installation, use, and office software are a little more pressing. I guess I can see if I can get him and one of the players (or several...) to see what we could accomplish under a little better conditions...

[Svartalf]

I like the idea of hitting back with a LINUX only game. Imagine a game that rates highly and makes windows users think again that Linux gaming is below mediocre. A game that is relatively new with technological advancement and original. It's a good thought at the moment, but I guess for me, just a dream. I'd love to try and make it a reality and become the next ahem, loki. But without the corruption / embezzlement etc...

Heh... Add just plain flat stupid business decisions on top of the other two and you'd have Loki ("Stupid" would be things like having Q3:A for all intents and purposes and delaying the official release for fifty thousand special edition tins, held up in customs for three or four weeks... This is why the Linux sales for Q3:A never did well- everyone couldn't wait, bought the Windows version and "patched" it to be the Linux one...).

Unless you're flush with cash, the only way you're going to accomplish that is to sit down and just do it. Be forewarned- it's NOT simple or easy. I know, I've been at it as a spare-time vocation for the last four or so years now. I've gotten a little somewhere with it all, but it's still not there with all that effort.

I'm very impressed with the Nintendo wii. I was once a die hard Sega fan. For a gamer brought up on Sega mega drive and Sega Saturn. Then to Windows gaming for 7+ years. Now Linux for the past three years. There is a future, but not as bright as I first thought. Consoles are becoming far more successful with games than PC's ever have.

In reality, the current generation of consoles may be a goad for many Linux titles. OpenGL ES 2.X is the base rendering API for the PS3. It's one of the primary middleware targets on the Wii. This is an easy, easy port back to OpenGL 2.X (and vendors are working on an OpenGL ES 2.X driver interface for Linux and Windows, removing the actual need for any port...)- and you're about to see a resurgence of OpenGL/OpenAL/etc. "PC" games pop up as a result of the PS3 and the Wii even being around. I know this for a fact as a result of my work for my current day-job client. This makes for more prospects for everyone interested in Linux gaming.

PC's would be lucky to cover a 6th of the console market in terms of profit and sales. Linux games almost feels like a hobby than a fully fledged industry. I'm glad it exists though. Gaming on Linux has something special about it, I'm trying to put my finger on it, but so far finding it hard to point out. Linux feels very fluid to me. A much more mature environment and less driven by money.

It might be that. It's definitely more of a hobby than an industry for many (I won't place Michael Simms and Ryan Gordon in that set- for them, it IS an industy, fledgling as it is right now... ) I know I can't support myself with it as a form of making a living right now.

Maybe this is the problem since games developers want large sums of money. I don't see this as being a problem though if it's done right. If a game was as good as say WoW, yet cost much less to play, it may just hit off. WoW might be a bad example of a game I have in mind.

That is precisely the problem. The studios or publishers want what is an insane amount of money to make a port happen. They expect turn arounds like they see on the Windows side of things, not realizing that the scale of the Linux market for titles, at best, is 1/5th the size of the Windows one. They ask at least 5-10 times more than they probably ought to be asking for the access and far too much up-front on royalties.

[Synergy6]

So.. those titles were as good as it gets, and nobody with an interest in Linux gaming has $100K? What about Shuttleworth or somesuch?

[Svartalf]

Overall, I do not agree with the statement that people will not do a Linux title, for whatever reason. However, I can but hope that the examples provided were not the shining lights of Linux porting.

Well, you try doing it effectively on a shoestring.

Each title that comes out costs money to even get access to the privilege to have a shot at porting it. Typically, that amount is anywhere from $10-50k for each title ported. This doesn't even cover the royalties owed for a production run to sell the title so published- which, oftentimes, part of the total for a production run is owed up-front at the time you press the title. This is IF the studio and their main publisher even agree to at least give you a shot- a large number of them won't even give you the time of day on this; this having absolutely nothing to do with Loki's flameout- they did this even back in Loki's day.

In order to get something like Quake 4, you'd have to pony up 50-75k, and would owe at least around that much for the up-front production royalties.

And that doesn't cover the labor for the porting effort in any of that.

Do you have about 100-250k burning a hole in your pocket? I can probably hook you up with several prospects for you on some titles that are about to come out for the PS3 that might have a sympathetic ear for a version in the Linux space.

[b15hop]

I like the idea of hitting back with a LINUX only game. Imagine a game that rates highly and makes windows users think again that Linux gaming is below mediocre. A game that is relatively new with technological advancement and original. It's a good thought at the moment, but I guess for me, just a dream. I'd love to try and make it a reality and become the next ahem, loki. But without the corruption / embezzlement etc...

I'm very impressed with the Nintendo wii. I was once a die hard Sega fan. For a gamer brought up on Sega mega drive and Sega Saturn. Then to Windows gaming for 7+ years. Now Linux for the past three years. There is a future, but not as bright as I first thought. Consoles are becoming far more successful with games than PC's ever have.

PC's would be lucky to cover a 6th of the console market in terms of profit and sales. Linux games almost feels like a hobby than a fully fledged industry. I'm glad it exists though. Gaming on Linux has something special about it, I'm trying to put my finger on it, but so far finding it hard to point out. Linux feels very fluid to me. A much more mature environment and less driven by money.

Maybe this is the problem since games developers want large sums of money. I don't see this as being a problem though if it's done right. If a game was as good as say WoW, yet cost much less to play, it may just hit off. WoW might be a bad example of a game I have in mind.