REQUEST: Linux version

That english is better than mine, and I am english

Is the Dolphin source open? If it is, a Linux version is simply porting the core and writing OpenGL plugins. Not exactly all in a day's work, but if Sixtyforce can be ported from 1964, then there's no reason Dolphin can't be ported to Linux, OSX, BSD, etc.

Edit. Did some back reading. Not open source! Wow.

I just kind of ASSUMED that it was, because emulators are probably one of the best types of projects to open source.

My pleas will probably fall on deaf ears, but I implore the authors to register a project with SourceForge. Open source == community development and cross platform ports. There's no reason NOT to open source it except for vanity, and we're above that here, aren't we?

</endrant>

Kethinov said:

Is the Dolphin source open? If it is, a Linux version is simply porting the core and writing OpenGL plugins. Not exactly all in a day's work, but if Sixtyforce can be ported from 1964, then there's no reason Dolphin can't be ported to Linux, OSX, BSD, etc.

Edit. Did some back reading. Not open source! Wow.

I just kind of ASSUMED that it was, because emulators are probably one of the best types of projects to open source.

My pleas will probably fall on deaf ears, but I implore the authors to register a project with SourceForge. Open source == community development and cross platform ports. There's no reason NOT to open source it except for vanity, and we're above that here, aren't we?

</endrant>

Click to expand...

I'd say one of the big reasons it isn't being made open source is because there are a lot of idiot code monkeys that want to nick it and make it their own - with a project this significant, I don't blame the authors for making it their own work entirely. Plus, I'll bet it's really satisfying to be able to say "I was one of just a handful who created the next-gen gc emulator". Correct me if I'm wrong

Plus, I'll bet it's really satisfying to be able to say "I was one of just a handful who created the next-gen gc emulator".

Click to expand...

In other words, vanity. Such vanity hinders development. If the project was open source, other "code monkeys" might contribute, or fork the project, or whatever. But the point is more people would be working on it and therefore the emulator would be playable sooner.

It seems the authors feel that their vanity is more important than perfecting the emu sooner.

Not surprising, abeit a bit petty. Lots of game related projects remain closed source due to the vanity of the creator(s). I guess gamers are just more likely to be vain? Subspace/Continuum, for example, is a perfect example of a game which should be open sourced, to fix rampant security problems and to allow for a porting of the client to Linux and OS X. Instead, the most recent version of the windows only client, was modified purposely so that it could no longer be ran in WINE, and all servers were forced to deny older clients. The developer was basically saying "don't tell me what to do. I'm in charge here, not you" after repeated requests for open sourcing the project or at least porting it.

Granted, that story is an extreme. But as you can see there is a clear relationship between vanity and unproductivity. Now if the authors of Dolphin want to continue to keep the source closed to indulge their pride and whatnot, then whatever. That's their right. But it only benefits them. Whereas open sourcing it would benefit the greater good.

Enough said. I'm not here to start a flame war.

Kethinov said:

In other words, vanity.

Click to expand...

It's a matter of opinion. I don't see it as vanity, not wanting to get your hard work destroyed by some. I do see the benefits of open source, but I don't think the authors deserve to be called "vain" because they don't see it your way. 'Nuff said.

Closing the source of a console emulator is not preventing their work from being "destroyed". Entirely the opposite. You said you recognize the benefits of an open source project. Surely you can see how opening Dolphin would possibly help the project, or other projects such as Dolwin? Much like how 1964 has helped other N64 emulators.

There are only two reasons to close the source of any software project. 1. Intent to sell. 2. Desire to take pride in being one of a small group of elite coders with "special" source code.

Clearly, this set of circumstances falls into the second category. They have no intent to sell, that I know of.

Some people on this forum would have us believe that asking for the source code behind Dolphin is just as offensive as asking for ROMs or disk images. I find that very belief offensive. I see open source as democracy and closed source as fascism.

Yes, I realize that vanity may have a negative connotation, especially in the context I've used it. But it's entirely accurate. By the very definition, it is taking pride in one's accomplishments. Single handedly coding a GameCube emulator is indeed something to take great pride in! But there are other ways to indulge one's vanity. I'd be pretty pleased with myself if I had coded the Dolphin core, open sourced it, and watched as people flocked to make improvements. Just as Linus Torvalds was pleased when he started the Linux kernel. Only 2% of the current kernel is actually his code now. But because he started all, he's still the famous one.

I know I come off as somewhat persistent, but open source philosophy is something I greatly believe in. I implore the developers to reconsider the decision to keep the source closed.

If you don't respect our decision to keep the thing closed for now, then don't use the emulator. By keeping it we are able to personally have a full understanding of the entire code base and that's practically necessary for a project like this (you need to have written a highend emulator to really understand this). This wouldn't be the possible if many more people were adding their own little tweaks all the time. Besides, there aren't really many people who are able to make useful contributions to a project like this anyway.

Right now OpenGL is less suited than Direct3D for emulating the gamecube graphics (OpenGL has less per-primitive overhead so I would have chosen it instead if the shader language situation wasn't so abysmal). D3D has HLSL which is practically necessary while the standard OpenGL shader language GLSL still doesn't work on nvidia cards and is very early and glitchy, and isn't as well suited. It's not a simple matter of just porting the current plugin to OpenGL.

Besides, if we were to make a linux port, all the usual problems with linux development that makes me stay away from it crops up: What UI toolkit should be used? What audio server? etc... whatever choice we would make, haters from the other side would crop up. That's totally unlike the situation on Windows where the only choices for these things are Win32 and DirectSound, and everyone is happy with those since they are good and work on everyone's computers without configuration.
I'm not interested in hobby Linux development until these things are standardized properly, it's just not fun to deal with crap like this. The coherent environment of Windows is a much better framework for "for fun" hobby development.

Hmmm, in the Dolphin Helpfile stands, that there most likely will NOT be a linux port by the authors auf Dolphin. And there is no need for one atm. Have a look at dolwin, this is the open source projekt emu ;D

ector said:

If you don't respect our decision to keep the thing closed for now, then don't use the emulator.

Click to expand...

No problem. I don't respect the decision and I won't be using it. Or anyone else's GC emu for that matter! hehe

This is because I don't own an actual GameCube and therefore have no games to rip. I will eventually join the scene though. Closed source or not.

As for your complaints about OSS and Linux, I'm not asking you to port it. Only open source it. I may not be an emu programmer, but I am a programmer and I know how much a chore it can be to port software. By open sourcing, you can let others deal with porting the core and dealing with OpenGL plugins.

One thing you said puzzles me. You ask "what toolkit to use" and "what sound server". Well, toolkit is irrelevent seeing as how I'd say most people play full screen anyway. But personally I'd use GTK2. Lot's of KDE users are used to using GTK apps, like GAIM, the GIMP, and XMMS. As for the sound server, it's my understanding that ALSA is the de facto standard here. (Correct me if I totally missed what you were talking about.) But why make that decision at all? Couldn't the plugins decide? Much like there are OpenGL AND DirectX graphics plugins for 1964?

Also, about plugins, isn't it helpful for plugin authors to have the source available to them? What if Rice wanted to write an OpenGL plugin for Dolphin?

Finally, you claim that keeping it closed keeps your understanding of all the code intact. That, I agree, is important with such a complex project. But if it were open source, it doesn't mean that everyone has a say on what gets added to *your* project. It only means other people can *submit* improvements, or fork the project into their own emulator if they wish. I see both prospects as helpful to GC emu as a whole.

Open Source

I think you're forgetting something ector. You don't HAVE to accept patches from people "you don't trust". You don't HAVE to port it to linux, others will for you. You don't HAVE to do anything other than open the source and let the people who are interested in porting your software to other platforms do all the work for you. Not only would you get possible bug fix patches (which you could of course deny) you would also get people doing you a favor by making your project more popular and making a lot of OS X and linux users happy. Are you really so concerned with having complete control over the source that you don't want to let others choose their own platform? Many of us use macs and linux for philosophical and practical reasons. I'm a developer, and a serious one at that, so i don't consider windows a practical platform.

Windows is not a better gaming platform. it's just the act of ignorant developers who don't know that there are other options like linux/mac.

My god guys you are irritating. Just leave it alone - a group of people intelligent enough to write code for something like this deserve the right to make an educated and intelligent descision. Now pretty please, an author's made his statement, let it be.

omnivector said:

I'm a developer, and a serious one at that, so i don't consider windows a practical platform.

Click to expand...

This line tempts me to write you off as a slashdot troll, you know.

Fact is, Windows is an extremely competent platform for serious media apps and games, much more so than linux at the moment and probably about equally as good as MacOSX. The threading model is (IMHO) way more practical and easy to program than posix threads, there's Direct3D that unlike OpenGL doesn't force you to write specific code for each card to reach the most recent features (the shader situation in OpenGL is just a goddamn mess), etc. Saying that Windows is no good is just ridiculous. Sure, Microsoft's security record isn't the best, but they're goddamn excellent at software and hardware compatibility and consistency compared to linux. Can't really compare to MacOS due to the "fixed" platform.

The plugin interfaces are still not fixed, the graphics plugin interface is due a big reworking and the sound plugin interface hasn't even been defined yet (the code is all in the core), so we can't open those interfaces yet. They will be opened when finished.

The main thing I have against opening the entire source is that it's a lot of extra work and it's no fun reading through sent-in patches and carefully merging them into the codebase. It also creates an obligation for us to actually look at patches. We don't want that since this isn't a really serious project and we often take long breaks. Dolphin is a project 100% for fun and 0% for profit.

In fact I have now gotten so bored of this discussion that I quit it [the discussion]. Deal with it, and have fun with the emulator instead of debating software philosophy.

As you may know i'm developping mupen64, a n64 emulator, as an open source project for MacOSX, Linux and windows. The multiplatform characteristic of my project was a challenge, i wanted to see how far optimizations can be pushed without using any platform specific code... Let's stop talking about my project i'm off topic
I just wanted to tell that opensourcing the core of my emulator has never ever helped developpement and i believe not much has been contributed to 1964. PJ64 hasn't got any improvement in the core itself too as far as i know after 1.4 source code release. Why ? Because to add something in an emulator core you should fully understand the whole source code, all modules are usually dependants.
Now it's true that the emulator can be ported, extra features can be added.... if it's an open source project. But it's maybe a bit early to think about this... Working hard on a port doesn't make much sense if the next version has a totally different source code. And extra features aren't necessarily needed for an emulator in such early state. Maybe the authors will make it open source when the code will be more mature if they want to, but it can take years coz again it's hobby developpement and things aren't constantly progressing.
Also there are a lot of other ways to help emulation community than open source project. I'd say open discussion between authors of the different emus can be much more productive than open source actually....

Last one thing, i'm a linux fan and no windows doesn't sucks on everything, yes it's pretty good for multimedia, yes all mature OS like linux, windows and OSX have their pros and cons, sometimes i'm programming on windows and i'm thinking, damn it's so hard on windows but it would have been so easy on linux. Some other times, it's easier on windows than linux.... Integrism is bad even in the computer world.

I agree that its probably too early to make it open source, but really, maybe you need something that works quite a bit better before you consider it for that. In any case I think I should just point out that the open source development model is not as random as you appear think. Many open source developers have a highly organised structure to their development models, with chief software designers retaining ultimate athority over what is and is not included in the codebase. People submit patches and enhancements as they see fit and they are either accepted or rejected by the project directors. You can't force bad code to be accepted into a project, because the job of the project directors is to prevent this from happening. Ofen projects of this nature are supported by a large number of testers, and these testers tend to report bugs back to the project directors, who either reject modifications on the basis of the seriousness of the bug (or otherwise) or kick it back to the developers to request changes.

Moreover as has been pointed out, if the author has no interest in porting their work, then it is simply a matter of allowing others to take up and complete that task, so that if the author is uncomfortable with the difficulties involved in writing specific optimisations for specific cards, he need not worry about it, because the task need not concern him at all.

How fast development progresses simply depends on how popular a project is. I'm not sure how popular an n64 emulator might be, or how much interest among developers it is likely to generate, but I am certain that a GameCube emulator is likely to generate a whole new level of support.

The only thing you have to loose - well there isn't anything I can think of really - except perhaps if you think one day people might pay for your emulator - which is a little unlikely given that most other emulators are free. The advantage is that perhaps, just perhaps someone will make a discovery that you yourself can use - since effectively any discovery that is made would be open source too.

But of course the author retains full control over this decision - and it may indeed be too soon to consider whether to do this or not. I just wanted to try to correct the misconception that open source development is an entirely random process. Indeed this is not the case - and while anybody could in theory alter the code - this does not mean that any such changes need to be accepted into the official project.

Kind regards,

GJ

Since this is not a debate on Linux vs Windows and only on open source, I won't comment further on ports or OS comparison.

ector said:

The main thing I have against opening the entire source is that it's a lot of extra work and it's no fun reading through sent-in patches and carefully merging them into the codebase. It also creates an obligation for us to actually look at patches. We don't want that since this isn't a really serious project and we often take long breaks. Dolphin is a project 100% for fun and 0% for profit.

Click to expand...

There is no obligation. You don't have to accept anything anyone sends you. You might even add another person to the team whose sole job is to judge what gets added and what doesn't.

You may find it interesting to know that Linus Torvalds doesn't actually code the Linux kernel barely at all anymore. All he does is say yes or no to additions.

The point I'm making is simple. The more the merrier. raid517 summed it up pretty well. You have nothing to lose and so much to gain in an open source project.

*Sigh*...

Please, if you have nothing useful to contribute, then say nothing. That utterance was entirely pointless.

Kethinov said:

Please, if you have nothing useful to contribute, then say nothing. That utterance was entirely pointless.

Click to expand...

Don't get frustrated - the authors have a right to make their own descision, and deserve to not get constantly harrassed if they don't see it the way of others. Moderators, how about a topic lock?

This is not harrassment. It is a discussion about software philosophy. And this is a forum. People come here to discuss stuff. If you're no longer insterested in the discussion, you don't have to post.