Print

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

[Abalieno]

And finally, you don't really specify what you would like development to look like, practically

I think I've been explicit enough.

Better relationship with mods, or more coordinated effort. This means giving them access to preliminary releases, as well integrating the stuff that you can find in DFhack, for example. The stuff that FIXES things, not the stuff that arbitrarily changes the game. Toady should work with them, talk with them and so on.

In the "Text Will Be Text - dfhack plugin" for example you can read:

I had a discussion with lxnt, baughn and Toady, and later on another with Taffer, Baughn and Toady. Baughn would like to update the graphics significantly, but cant work without source access. Toady understands this, but wont give this access. (which is perfectly within his right to do)

So a graphic engine update is most likely not going to happen any time soon.

So what I ask Toady isn't about going open source, but at least to reconsider these positions.

There was a time when he made significant progress in this direction. Then he stopped. I'm simply asking he goes back opening up the UI layer of the game (since it looks like Toady just won't do it by himself).

That was one aspect. The other aspect is that after two years of radical rewrites I hope he puts that kind of strategy on a similar year-or-two delay, so that the next couple of years are more focused on those features I defined "small things that have significant impact compared to significant things that make little difference."

[Putnam]

http://www.bay12games.com/dwarves/dev_single.html

There's a reason this version is 6% of the entire game and 15% of the game so far.

[GavJ]

Most of the OP is meandering hot air fluff (2d nostalgia? Really? And complaining about different colored stones? ...), but I resonate with two main points strongly.

I understand these two things aren't fun to do, but regardless, prior to any other features being developed IMHO,

1) The game needs Needs NEEDS to make its GUI modular and separate from all underlying code. The game needs to send information about each tile and what exists there in an information format API to the generic, abstract GUI, so that we can all go and make whatever graphics we want without getting in Toady's way. This is #1 priority, IMO. (He can keep using ASCII and not worry about it after that!!)

2) Less of a priority, but still I think more important than any given feature... As close as possible to every single constant number and string in the code should be moved to the RAWs. Basically anything other than math functions and abstract frameworks. Every building, every reaction, every script of code for one-off things should be tagged to be able to be used elsewhere optionally. Everything from gravitational constants to internal scalar values in the worldgen code. Because every minute Toady spends coding things into RAWs unlocks hours of time that the community modders can spend making millions of diverse features to keep players entertained. It's like investing in a highway system instead of investing in horse drawn carriage subsidies. It's just always going to be stupidly more efficient at contributing to game epic-ness than direct, solo, lone wolf coding by Toady while other people sit around twiddling thumbs or banging their heads against the wall finding absurd roundabout workarounds.

(Basically, if at any point in the code it says "X = [some specific number other than pi]" or "Z = [some specific string other than Toady's name]" and they aren't pointers to the RAWs, IMO it's doing it wrong.)

And then when there isn't much else to put in the raws, and the GUI is out there available to be made 50x more epic, THEN it becomes more efficient to go back to adding new features.

[miauw62]

I feel that Toady releases whenever the game is ready for release. Updates may have been frequent back in the day, but DF grows ever more complex, and the same is probably true for the code. Activating the world required him to make changes almost everywhere in the code (as one could easily deduce from the devlogs).

And after the long cycle, there will be a shorter cycle, as usual. One that polishes the game and adds small features that don't require massive rewrites.

Honestly, I feel that Toady rushed DF2014 a little bit towards the very end so our hype didn't implode.

E:
However I agree with GavJ. Especially point 1. The UI needs an overhaul NOW.
Toady can add more things to the raws when he feels like it as far as I'm concerned, but the UI is really important and overhauling it will make future development easier.

E2: And about the open source thing, I think that the DF source will almost certainly be released when Toady dies. The DF community has enough inertia for a permanent open source project, and I'm sure that Toady would want his opus magnus to live on instead of fading into obscurity.

I guess there's also a chance that the source will be released after we hit version 1.

[Lightman]

1) The game needs Needs NEEDS to make its GUI modular and separate from all underlying code. [...]

2) Less of a priority, but still I think more important than any given feature... As close as possible to every single constant number and string in the code should be moved to the RAWs. [..]

For point #2: Some of us don't use mods, so this really isn't a priority for everyone.  There are other features I'd much rather see.  Still, I'm sure Toady is aware of the modding community.

For point #1: It would be interesting if the UI could be decoupled but it's probably not an easy task.  Even though the UI is not a problem for me, I know it's a stumbling block for many people (especially new players) and it does need improvement.  Perhaps it can move to something customizable, over time.

[Xnidus]

Pse... I love the work of Toady, I like very much the "city-builders" & simulation games and DF in this context is the best game that never I see. The simulation of a "town" with their resources and their people is very great, and now, with a living world... become incredible.

But I have my own opinion about the "real needs" of this game, but I think that is a big trouble that must be solved URGENTLY. DF is complex, and every new release become the game more and more complex. And more complexity increase the resources that the game needs for work, every additional bit of simulation increase more and more the limits of the simulation: You can create a complete fantasy world, but... How you can play in this world if you really need a CERN supercomputer for it? Or, how you can simulate a very long living-epic fortress if the FPS become near and near in every release?

I don't know the problems of the programation or the source code (although I read some years ago that the game only can use one pc core), and maybe should be very difficult or near impossible to solve this... Today,that only the 40% of the game is ready, a typical fortress only have a ~20 years lifespan. How much can live a fortress in a 1.0 version of the game? I usually think that the FPS death kill more than 50% of the true real potential that the 0.34 version have, and force the player to many "weird" strategies for gain two or five FPS more, or disable the weather or temperate, or the invasions, and others sources of FUN... 

[miauw62]

Computers always get faster, but sadly most of that is in more cores rather than faster individual cores.

And there's indubitably much optimization to do, Toady has even said that there's lots he can do to speed up the calendar.

[Xnidus]

Computers always get faster, but sadly most of that is in more cores rather than faster individual cores.

And there's indubitably much optimization to do, Toady has even said that there's lots he can do to speed up the calendar.

Yeah, there is the question, that the game only can use one core and the velocity of the individual cores not evolved greatly in the last years... But, Can Toady can increase the general optimization of the game? Or better... Could the game in a future use more than one core? I like to believe that not is too late.

[Uristsonsonson]

Toady's already said he's not doing multithreading. It's up there with time travel and fortress sewage, IIRC.

[Hetairos]

I go take a shower and 30 replies get posted in the meantime 

Remember the learning curves thread?

[Zeranamu]

things

Um. What? No. Also, Toady does what Toady wants. Not what you want. I win!

[metime00]

http://www.bay12games.com/dwarves/dev_single.html

There's a reason this version is 6% of the entire game and 15% of the game so far.

That dev page is outdated, use the one linked from the website http://www.bay12games.com/dwarves/dev.html

And I believe it's been said somewhere, maybe the DF Talks, that version number is roughly correlated with percent complete of the game.

I think it lacks real awareness to see a new release that's entire purpose was setting the groundwork for the living world that will be used by most every feature from here on out, and then complaining that it's a trend that Toady does these huge rewrites. It's like if when he added 3D people got scared and said "he's just gonna add a 4th dimension later, remember when he didn't add dimensions willy nilly".

These reworkings of the core of the game aren't just for kicks, they're all deliberate and required to move the game in the direction that Dwarf Fortress always had since it became the main game of Bay12Games. Do you guys worry that he'll take another two years to give each world another world to have simulate and live while you play? This release cycle was all one time foundational stuff, the features added were mostly ancillary.

There's nothing about the development of version 0.40 that suggests that after this more frequent, smaller, and content-driven releases won't happen; he's already done all the heavy lifting in regards to the engine of the game!

[miauw62]

Not all the heavy lifting.

* miauw62 glances at UI code

[Sergarr]

Not all the heavy lifting.

* miauw62 glances at UI code

At least rewriting that doesn't require changing the way all interactions between people work.

Also UI is much easier to test.

[WaffleEggnog]

Nightly releases might be a good idea. That way we can preview things before they come out and see whats being worked on and possibly provide constructive critisism, as well as having the convinience of releases nearly every day to spice up the game.

I know DDA and tons of other roguelikes do it, and as far as I know it worked for them.