A Radical Idea For More Linux Game Ports

[dimko]

Porting is one thing, maintaining forever (since API/ABI compatibility in Linux is a joke, whereas in Windows you compile once and then run for years, or even decades) is another.

No, RedHat and Ubuntu don't have resources for that.

Basically the guy is asking game studios to open source their games for selected RH/Ubuntu developers - nah, it's not gonna happen.

1) statically linked stuff?
2) We had steam for a while. Not once game was broken by my system upgrade.

[shanefagan]

Its like this guy is MMO fanboy that doesn't like dualboot. While thats a nice wish nothing will explode unless Linux will get some AAA(non MMO) game exclusive(even temporary, for 1 year) and this stuff likely costs alot.
Current course is nice, alittle by little step by step Valve pushed this idea around the web and its progressing as we see not so popular but still new games like Witcher series.



And Riot with Blizzard aren't indie in the slightest , they have tons on top of tons of money. Porting is not developing, they could easily port their games if they would ever care.

Actually im not an MMO fanboy I just know where the users are at. Blizzard games and LoL are both pretty much one of the last big hurdles. LoL being the biggest game in the world right now and yes I know its on Adobe air but im sure Riot know of the problems with Air too and they have a lot of development they don't talk about so id presume they are more than likely working on a new client to update the graphics and stability of the game.

My idea has nothing to do with exclusivity, its more about bringing games that wont get ported otherwise to the platform that would be blockers for some users. So as for cost the idea is that they pay the development costs only to port the game to Linux and its only a loan also so you can definitely expect some if not all of it back.

[Azrael5]

Answer is so SIMPLE: start and play for every kind of game. Linux has to implement a system to make games instant playable after having implemented directX. so why porting?

[blackiwid]

Actually im not an MMO fanboy I just know where the users are at. Blizzard games and LoL are both pretty much one of the last big hurdles. LoL being the biggest game in the world right now and yes I know its on Adobe air but im sure Riot know of the problems with Air too and they have a lot of development they don't talk about so id presume they are more than likely working on a new client to update the graphics and stability of the game.

My idea has nothing to do with exclusivity, its more about bringing games that wont get ported otherwise to the platform that would be blockers for some users. So as for cost the idea is that they pay the development costs only to port the game to Linux and its only a loan also so you can definitely expect some if not all of it back.

Do I understand your idea right, a Linuxcompany lets say canonical pays Blizzard to port Hearthstone or maybe Riot, for the porting, and then the linux-client-based income can canonical take for the risk and investment?

Why would Blizzard or Riot do that, many of this users would if there is no linux client, dualboot or use a special windows machine to start the game instaed so they would loose income?

This company never would know how much real new users they get through this instead of users that would have played it under windows. So they just have to hope that they get enough real new users through this. Or do you mean its a gift for this company they just get money and pay nothing back, so that they have even a bigger advantage over the competition than before when they are genre-leaders already?

It makes no real sense give the companies that made linux ports already and maybe had not much success yet, some money so they dont feel as loosers because they invested into linux and quit doing that. Reward the good one, dont give the bad companies money to reward them for ignoring or boycotting linux. It just makes no sense no matter how you spin that.

[Hamish Wilson]

Really? Then why not a single Loki ported game works correctly and fully in all modern Linux distros? Not to mention a totally broken Loki installer which refuses to run with modern glibc and intentionally broken `tail` utility?

I have ran plenty of older Linux titles, including Loki games, on modern distros.

[laykun]

As a windows game developer I can say it's not just the cost of porting the software to other platforms, it's the support costs that are a worry, a continuing cost that nobody would want to pay for. Considering the fractured nature of linux it just makes games harder to support unless you target only Ubuntu or only Redhat. This is the purpose of having the SteamOS distribution I believe.

I would prefer to see Valve fund the development of Wine to be honest, and push graphics vendors to release high performance open-source drivers with good support for gallium, and in turn give the wine developers a good reason to use gallium-nine instead of their horrible GL->DX translation layer.

[Hamish Wilson]

But the thing that *always* works, and only keeps on getting better is old windows games with wine on linux. I can run games on my linux box that do not run on windows. That's a fact.

And on the other hand, I have encountered several older Windows games that do not work in WINE. I think it is time everyone stopped using sweeping statements.

[Hamish Wilson]

As a windows game developer I can say it's not just the cost of porting the software to other platforms, it's the support costs that are a worry ... I would prefer to see Valve fund the development of Wine to be honest

If the problem is not the porting but the support costs, how would using WINE even be helpful?

[laykun]

If the problem is not the porting but the support costs, how would using WINE even be helpful?

Using WINE takes the support costs away from the developer as they are not directly involved in the development of wine and do not provide official support the linux platform. There would be no costs as the porting costs are on the wine developers side and support is handled by the community through portals like WineHQ or some sort of Steam service if Valve did some funding.

[Daktyl198]

If the problem is not the porting but the support costs, how would using WINE even be helpful?

It's quite reasonable, given the rest of his post. Most of the performance hit (probably a good 80%) of running games in Wine is the DX > OGL translation and execution. If NVidia tomorrow were to release a DX 9-11 supporting driver for Linux, and AMD did the same for their blob driver, then Wine would have "native" paths to take and wouldn't have to translate anything in the graphics department.

From that point on, all they would have to focus on (for Linux at least) is the other generic Windows >> Linux translations, most of which they have down pretty well so far. Then, running games "ported" to Linux via a Wine-container (like what PoL does) would be usable and very close to native Windows speeds (average 80-90%, I'd assume).

Thus, porting costs = negligent (somebody to write the wine container script and package the game), and supporting costs even less so (somebody to fix said script/packaging every year or so if it breaks).

The problem then comes down to people working on the graphics drivers... ignoring the FOSS drivers (who work their asses off), the blob driver people I feel are lazier and probably don't support DX on Linux purely out of the added work they'd have to do to maintain it :/