Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Torak]

Also, why can't you people whom are so eager to make a new interface project just wait for Toady to remake it on his own? It's not that far off in the future, maybe a year or so off.

[Doppel]

I don't see whats wrong with being eager to push this project forward, isn't that a sign of positivism and that people care for this?

[Torak]

I don't see whats wrong with being eager to push this project forward, isn't that a sign of positivism and that people care for this?

That's not my point, they ask as if Toady's long since decided to abandon any type of interface change. A good time to complain about/change the interface would be after toady decides to remake it himself.

[Frobozz]

I've donated a small amount to Toady over this last year and since I'm paying his rent I'm pretty much allowed to just walk into his house and crash on his couch whenever I want?

I never implied you could waltz into his house and lounge around on his furniture. What I was implying is that for Toady to continue receiving donations he has to 1) continue development of Bay 12 products (he may receive donations due to projects other than Dwarf Fortress) and 2) make them available freely (who'd pay for something then turn around and donate?). This is just common sense. Not a threat, not something that entitles the donor to something, etc. Just basic logic.

should he choose to do so it would most likely increase FPS since the program could be threaded, and it would also create the possibility for third party GUIs to take over the rendering side.

I'm a bit skeptical about how much of an improvement you'd see with frame rate considering DF is currently using OpenGL to hardware accelerate the graphics. But at the same time I doubt you'd lose any FPS since the game could be processing a cycle while the graphics are being rendered in another thread.

It's not that far off in the future, maybe a year or so off.

A good time to complain about/change the interface would be after toady decides to remake it himself.

First off, I seriously doubt its only a year or so away. Maybe changes the size of which he is currently making. But a completely remade interface is more likely closer to version 1.0 than 0.3. Also I fail to see how suggesting possibilities for third party interfaces is complaining.

[Torak]

First off, I seriously doubt its only a year or so away. Maybe changes the size of which he is currently making. But a completely remade interface is more likely closer to version 1.0 than 0.3. Also I fail to see how suggesting possibilities for third party interfaces is complaining.

I say year or so in speculative terms. I know as well as anyone else, which isn't that much. Toady said he's going to do the presentation arc long before 1.0, it was actually next on the agenda for a while until he decided not to constrain himself to one arc. I also never said that you were complaining, that's why I put 'complain about/change the interface'.

[FlexibleDogma]

...
I'm a bit skeptical about how much of an improvement you'd see with frame rate considering DF is currently using OpenGL to hardware accelerate the graphics. But at the same time I doubt you'd lose any FPS since the game could be processing a cycle while the graphics are being rendered in another thread.
...

Off the bat I'm sure you're correct, I was just thinking that starting to split the program up so it becomes threaded could be a big step.  Yes, the starting chunks would be "UI" and "Everything else", but it's a start.

@Torak: Hrmm, the quote tags got borked in that message, I didn't say that.

EDIT: I type English good.  I fix to better English.

[Omega2]

I don't see whats wrong with being eager to push this project forward, isn't that a sign of positivism and that people care for this?

That's not my point, they ask as if Toady's long since decided to abandon any type of interface change. A good time to complain about/change the interface would be after toady decides to remake it himself.

Wait, but you just did the same thing here:

http://www.bay12games.com/forum/index.php?topic=21870.0

Besides, that reasoning can be applied to every single suggestion/complaint in this forum: "wait until the game is further developed". It doesn't stop modders asking for new creature tags, or for normal players to ask for alternatives to wood in bed construction. Why should it stop us from asking for more interface support?

[Blacken]

I don't see whats wrong with being eager to push this project forward, isn't that a sign of positivism and that people care for this?

It's not your project to push forward.

There are also some pretty good technical reasons why a client/server model isn't exactly ideal. For one, it's pretty slow; IPC is not fast no matter the operating system; it's surely fast enough for non-immediate apps (ones that aren't updating > 100 number of times per second), but I'd bet heavily on a performance drop. While this is partially alleviated because of multi-core computing--though, for a game, you have to have some way to handle mutual-exclusion locking between processes, which is decidedly nontrivial to do in a decently-performing way. That abortion called FreeCiv can get away with it without problems (haha, FreeCiv not having problems, I must be an idiot!) because the game's turn-based and doesn't have to do immediate updates; Dwarf Fortress does. (Yes, I know about memory mapped files and synch objects. They're still very problematic--for example, how do you make the server yield for an update cleanly? What about when a third-party client chokes and dies?)

Another problem: it's not cross-platform. Windows and OS X don't use anything remotely near the same mechanisms for IPC. WINE doesn't support the Windows or UNIX IPC methods worth a crap, either. Yes, you can do shared memory for both, but it doesn't make a lot of sense to start letting the Windows and OS X codebases diverge a lot for little benefit (yes, this is "little benefit," you're being feature-champions and feature-champions totally suck).

While it'd be nice, I guess, to see it, I don't see a good reason for Toady to do it (either for the health of his project or for technical reasons). It's his project, he calls the shots, end of story.

I'm a bit skeptical about how much of an improvement you'd see with frame rate considering DF is currently using OpenGL to hardware accelerate the graphics. But at the same time I doubt you'd lose any FPS since the game could be processing a cycle while the graphics are being rendered in another thread.

No, there'd be quite a bit of improvement, especially for multi-core computers. Once you figure out how to do IPC in a way that doesn't completely suck ass, that is, and works on both Windows and OS X. That performance improvement can be achieved in a way that won't help sink Toady's project just by invoking threads within the app rather than hoping IPC works out for you.

First off, I seriously doubt its only a year or so away. Maybe changes the size of which he is currently making. But a completely remade interface is more likely closer to version 1.0 than 0.3. Also I fail to see how suggesting possibilities for third party interfaces is complaining.

This is an idea that sucks technically and sucks from a community/social standpoint. The bunch advocating this are being, in large part, whining jackasses (threatening a dev with "no donations" doesn't work, by the way, and would ensure that those of us who actually respect a dev's work will donate more; I'm just waiting on a project to clear before I send Toady some greenbacks). I can't for the life of me understand why you want Toady to do something self-destructive like this.


(Disclaimer: I'm an open-source dev, I write closed-source software, I can't make a water tower in DF without flooding everything, etc. etc.)

[Torak]

Wait, but you just did the same thing here:

http://www.bay12games.com/forum/index.php?topic=21870.0

Excuse you? In what way does asking for multiple level designations have to do with anything other than trying to prove a point of yours which has nothing to do with the discussion?

[Blacken]

What Torak asked for is an improvement to the game (hell, just save Z-level with the designation start/end and iterate). What you're asking for is for a radical restructuring of the game that is likely to diffuse the focus of the community.

Sing it with me: one of these things is not like the other!

[Doppel]

That's not my point, they ask as if Toady's long since decided to abandon any type of interface change. A good time to complain about/change the interface would be after toady decides to remake it himself.

I get your point, it just sounded as if "people that can't wait" and wished for a better/different interface somehow didn't want the best from DF and Toady. Almost as sort of a paranoia that they are out to get Toady or something. I most likely am reading way to much into it though, i'm not saying that is what you ment, so hope you don't take offence.

It's not your project to push forward.

What i ment more was people just wanting the best of DF and also wishing it the best, not as in forcing to lay their personal likes and dislikes on Toady's project. I'm seriously thinking way to many people here have ungrounded skepticism about the good intend of other people.

[isitanos]

um... actually the OP was asking for open source for the UI.  that's where this whole argument came from...  you know from the OP, who started the thread and asked for open source...

Right. I was, and I'm still suggesting that the client be made open-source, because I see it as the most effortless way for Toady to publish his API: by example. But others are right to point out that opening the source code is not necessary for 3rd-party interface clients to happen. But then Toady has to produce documentation, unless he just wants to let people hack the .exe and/or listen to sockets traffic to see how things work out between client and server. Writing documentation begins to look like the "collaborating with people" he wants to avoid.

By the way, Toady has published the source of some stuff, such as Kobold's Quest, if I'm not mistaken. So it's not as if he was some kind of radical anti-open-source guy. He doesn't want people to steal DF, but I'm not sure he treasures that much his interface code... the real jewel is in the game logic.

About the technical aspect of things, I mentioned sockets as an example. If it hurts gameplay under Wine or MacOSX, and there's no way to find a cross-platform library that abstracts the sockets functionality, it may not be the best choice. There are alternatives, such as the server being a DLL that's called by the client (or the other way around).

[Omega2]

I was pointing out Torak just did what he told us not to do. So that reasoning doesn't work, it can't be applied to this discussion. You can't tell people to wait until Toady works on the interface before they can suggest things about it, because that would imply everybody would have to wait until the game is finished before being able to make any suggestions.

All that, regardless of the content of the suggestion. If Toady reads this and doesn't find the idea interesting, we just wasted time and keystrokes. If he does fish any good concepts from it, then it's good enough. We are tossing ideas out, ideas he can simply ignore, not grabbing him by the collar and demanding he gives us full access to the code until midnight or the cute little rabbit dies.

See what we are coming up with here? Suggestions in good faith, not demands in the name of past donations or whatever.

[Torak]

I was pointing out Torak just did what he told us not to do. So that reasoning doesn't work, it can't be applied to this discussion. You can't tell people to wait until Toady works on the interface before they can suggest things about it, because that would imply everybody would have to wait until the game is finished before being able to make any suggestions.

You're completely wrong, and I find your reasoning insulting in the least. I said not to worry about the interface because toady will remake it to his specifications. He has not said anything about multiple level designation, so it's free game. Don't try and mix my words around to make it seem like I am a anti free-thought junkie.

[Jifodus]

I'm a bit skeptical about how much of an improvement you'd see with frame rate considering DF is currently using OpenGL to hardware accelerate the graphics. But at the same time I doubt you'd lose any FPS since the game could be processing a cycle while the graphics are being rendered in another thread.

No, there'd be quite a bit of improvement, especially for multi-core computers. Once you figure out how to do IPC in a way that doesn't completely suck ass, that is, and works on both Windows and OS X. That performance improvement can be achieved in a way that won't help sink Toady's project just by invoking threads within the app rather than hoping IPC works out for you.

You'd only free up maybe 10ms of CPU time (read the Kobold Quest source code, it has DF's rendering method), which on most computers is at most 1 more frame. Also, you'd probably lose 20-50ms just by sychronizing the data accesses.

Right. I was, and I'm still suggesting that the client be made open-source, because I see it as the most effortless way for Toady to publish his API: by example. But others are right to point out that opening the source code is not necessary for 3rd-party interface clients to happen. But then Toady has to produce documentation, unless he just wants to let people hack the .exe and/or listen to sockets traffic to see how things work out between client and server. Writing documentation begins to look like the "collaborating with people" he wants to avoid.

Toady doesn't have to release any code, or provide any interface for 3rd party clients, no matter how many times people whine or scream. No matter what alternatives you provide, Toady does not have to provide anything.