Bay 12 Games Forum

Please login or register.

Login with username, password and session length
Advanced search  
Pages: 1 ... 19 20 [21]

Author Topic: Third party interfaces and "Losing control of the project"  (Read 135932 times)

Kindjie

  • Bay Watcher
    • View Profile
Re: Third party interfaces and "Losing control of the project"
« Reply #300 on: August 03, 2008, 01:39:56 pm »

I don't see why Toady needs to open up an API for the interface to improve...? I also completely understand why he wouldn't want to do it.

The following three things are all that's really needed to make the game more friendly:
  • Better tile support - any ingame object should be easily represented with a unique tile.
  • Better mouse support - make everything that affords clicking clickable.
  • Introduce the minor interface changes suggested here - context sensitive rollover help and the sorts of things isitanos listed.

These don't change the game in any fundamental way but would make the jump in for new players a lot smoother. Maybe a few releases from now the game will be more stable in its feature set, and Toady can look at adding more things mouse users expect - like buttons that look/act like buttons, etc.

Adding 3D seems silly. Who's going to create all the assets? What would it add to the game? Extra interface overhead for camera control... The game just wasn't design with true 3D in mind. Similarly for isometric. You rotate the grid to get... what? A counterintuitive keyboard to cursor movement relationship?

Honestly, I can see why people are asking for an open API, but why not just create a list of considered features you'd like to see? The hacked 3D party utilities are still going to be there.

EDIT: And one more thing - what would be the scope of the API? How much of the internals will Toady have to reveal? From the features being requested, it seems like quite a bit. What would stop someone from taking everything he's made and creating a competing game, asking for donations, etc? How much time should he spend creating this API instead of adding features? Supporting it? In the DF wiki, which interfaces will be covered for new players? And I'm sorry for presuming, but I seriously doubt most of the people asking for this are any better qualified to build interfaces than Toady. Would their results be a large enough improvement to justify all the risks?
« Last Edit: August 03, 2008, 01:48:42 pm by Kindjie »
Logged

Frobozz

  • Bay Watcher
    • View Profile
Re: Third party interfaces and "Losing control of the project"
« Reply #301 on: August 03, 2008, 02:34:31 pm »

Quote from: Kindjie
The game just wasn't design with true 3D in mind.
Wasn't designed with it in mind? If you think about it, Dungeon Keeper didn't exactly take advantage of the benefits 3D has to offer. In fact, most games consist of 2D gameplay with 3D graphics. It isn't uncommon at all. Thus I don't see how this influences anything.

Quote from: Kindjie
what would be the scope of the API? How much of the internals will Toady have to reveal? From the features being requested, it seems like quite a bit. What would stop someone from taking everything he's made and creating a competing game, asking for donations, etc?
I've already given examples on what the scope could consist of and still hide the details. If you can be bothered to read them, just scroll back to near the beginning. The example I gave consists of no more ability than if you were to send keystrokes to the game as it stands now.

Quote from: Kindjie
I seriously doubt most of the people asking for this are any better qualified to build interfaces than Toady.
How so? Toady himself has already admitted he probably can't compete with someone else's interface. How would you consider it qualified? And in case you hadn't noticed, there are quite a number of programmers in this community. I'm sure a good number of them are capable of building an interface at least as good as what exists now.
« Last Edit: August 03, 2008, 02:36:49 pm by Frobozz »
Logged

SirPenguin

  • Bay Watcher
  • NEVER A DULL MOMENT IN MID-WORLD
    • View Profile
Re: Third party interfaces and "Losing control of the project"
« Reply #302 on: August 03, 2008, 05:22:45 pm »

And I'm sorry for presuming, but I seriously doubt most of the people asking for this are any better qualified to build interfaces than Toady. Would their results be a large enough improvement to justify all the risks?

Uh, why do you think that? Toady is a math professor turned programmer. He himself has stated he's self taught. Then you have plenty of CoSci majors here (at least one including me) who probably have a better idea on how to get an GUI working than him.

He is an AMAZING programmer all things considered, but better, he's an amazing planner. He and his brother both. That's why DF is so successful, he has a clear and well documented path. And all the math portions of the game, such as pathfinding and world gen, are all fantastic. But that's under the hood. Hell, all of math is under the hood.

Toady is great, but he's not great at everything. No one is. The alternative to hiring 'staff' is to let the community fiddle and tinker. That's all we're (I'm) saying.
Logged

Kindjie

  • Bay Watcher
    • View Profile
Re: Third party interfaces and "Losing control of the project"
« Reply #303 on: August 04, 2008, 05:42:22 pm »

Wasn't designed with it in mind? If you think about it, Dungeon Keeper didn't exactly take advantage of the benefits 3D has to offer. In fact, most games consist of 2D gameplay with 3D graphics. It isn't uncommon at all. Thus I don't see how this influences anything.

Of course. In fact I'm playing Evil Genius right now which is similar in that regard. The difference is that those projects had comparatively large budgets with dozens of people working on them. You need coders to implement an animation system. Riggers, texture artists, modelers, and so forth. Dungeon Keeper let you take over a minion in first person view, which (besides marketing) was the reason it was 2.5D (3D w/2d sprites). Other games, like Evil Genius, save memory by storing skeletal animation data instead of pages and pages of 2D animation.

Dwarf fortress has at least two things that make 3D more complicated: 1) multiple z-levels, and 2) hundreds of tiny objects that are easily obscured by each other and walls. That's assuming a team could find and coordinate enough artists to generate the required content.

Like I said in my original post, you'd spend a lot of resources on something that will only add interface overhead.



I've already given examples on what the scope could consist of and still hide the details. If you can be bothered to read them, just scroll back to near the beginning. The example I gave consists of no more ability than if you were to send keystrokes to the game as it stands now.

I did, and I think most people will be underwhelmed. I'd expect a similar thread to this one, actually. (EDIT: I haven't tried it, but just using Win32's PostMessage() should give you this already.) And if you actually wanted access to his data then you'll run into all the issues people have already brought up. Furthermore, you're almost certainly going to run into performance issues because clients will be forced to compromise with their algorithms or ask Toady for more features.



How so? Toady himself has already admitted he probably can't compete with someone else's interface. How would you consider it qualified? And in case you hadn't noticed, there are quite a number of programmers in this community. I'm sure a good number of them are capable of building an interface at least as good as what exists now.

By qualified I mean has studied/experience in design. If you think implementing an interface is the same as designing one then you're wrong. Any programmer can implement an interface; it's trivial in most modern IDEs. It's quite another thing to design a good one. Just look at all the crap software interfaces that are out there!

In fact, a good way of coming up with a bad interface is to presume you know what the user wants. A good way of coming up with a good interface is to iteratively develop a series of prototypes while gathering user feedback. Which is pretty much what Toady's been doing and why I suggest we create a list of things we'd like to see instead of whining for something he's already considered.

I'd just like to add that I don't strictly disagree with third party interfaces; I just think time and resources could be spent more effectively elsewhere.
« Last Edit: August 04, 2008, 07:41:48 pm by Kindjie »
Logged
Pages: 1 ... 19 20 [21]