Print

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

[Siber]

Just a word on the multithreading note, from my spectating on the computer industry, it seems like raw processing power hasn't really been going up for a long time, but rather the focus is on adding new cores. With that in mind, it follows that if DF wants to take advantage of future computing advances to support it's ever increasing complexity, multi-threading is the only way it'll be able to do that.

Saying it'd be easy, though... that's a whole other kettle of fish.

More in line with the topic, I'd say that the paragraphs of description are probably not going to be a huge boon to dwarf mode, but that they sound like something that can really benefit adventure mode down the line. Hell, you could be told to search for someone with a certain appearance, or maybe there can be cases of mistaken identity, or so forth. Dwarf mode isn't the only thing that this game aims to make good.

Still, I think it's really easy to get off track here. I don't think Threetoe is really asking what problems you have now with the game, or what needs to be done to bring it into line with your perfect vision of the future. I think that instead of asking what turns you off in hour five of playing, he's looking for what turns most people off in minute five. Endgame fun, eventual framerate crawl, unrealized features... none of these are issues that are going to smack a new player in the face and make them give up. They might eventually drive that player away in frustration, but only after they've given the game a real fair shake. They're not problems that stop someone from even having a chance at getting hooked.

I've asked people who were turned off in those first five minutes, and the answer is always the UI being unintuitive, not knowing what to do or how to do it, and not being able to figure it out. These aren't dumb people, these aren't people that are driven away by complexity. This is why I suggested tutorials and graphics tiles before. ASCII graphics may be suitable for those that are already familiar with them, but for the uninitiated, you have to learn what everything is. With a good tile set, you don't have to learn that one symbol is a dwarf and one is a door and one is a bed, because they look like dwarves and doors and beds already.

[Capntastic]

Just a word on the multithreading note, from my spectating on the computer industry, it seems like raw processing power hasn't really been going up for a long time, but rather the focus is on adding new cores.

False.
A 3ghz core made today is far more powerful than one from a few years ago.   Clockspeed isn't the only factor.

[Sowelu]

..........?

Okay, it's been like ten years since I had any clue at ALL about hardware, but I don't understand that.  How are two chips with the same clockspeed and number of cores different in terms of performance?  Is it because some have better optimizations, because I thought those had to be compiled in...or is it because they have a cache on the chip itself or something...or better pipelining, I didn't think that was that big of an improvement?

[G-Flex]

..........?

Okay, it's been like ten years since I had any clue at ALL about hardware, but I don't understand that.  How are two chips with the same clockspeed and number of cores different in terms of performance?

The clock speed isn't a measure of how many of a given operation is done per second. It's just a measure of the amount of clock cycles used to time it. Obviously, the same chip will perform much better if it's timed faster, but two 3GHz CPUs might still process things at different rates, since one might be able to do more in a single clock cycle than the other.

[father_alexander]

From the start, the difficulty of learning or rather understanding, an ingame tutorial would SO help geting new players, something like, "ok this is the embark screen, this and this do this and this" and then telling you simply what each one does, always, and i mean ALWAYS relying on a manual or the wiki for the more deep knowledge, then a manual would be of LOTS of help, not much , just a text something simple, like the wiki.

Second the graphics, while i dont mind the ascii, i do mind understanding whats going on , a decent solution to this would be making the graphic easier to apply, and maybe even attaching a simple graphic set to it, nothing big, just you know, something really simple, that would also make it a lot easier to control dwarves, and even feel attached to them!

After a while either the lack of something new or the difficulty to get a good map! see unless you start close to gobos you will eventually find it hard to get sieges or ambushes (though this will probably get solved in the next update) and its also solved wen you learn that you should always use science* , as for a good map, well , let me say i HATE aquifers , they have this tendency to appear in the most AWESOME maps and just ruin, them, and i am really obbsessive so sometimes i really need a perfect map.

*testing and trying shit while killing living things in the process

Some improvement in the interface would rock too, but its not thaaaat needed (from my point of view at least) if you do the tutorial , still some decent mouse interface would rock, also some things are really boring, like having to make 200 bedrooms, or a whole dining room, or even giving the correct jobs to everyone.

[Davion]

Now explain to me why teeth, beard layers and the hyper-detailed physical properties of the haft of the sword of the third '@' from the left are critical to any of those?  Many of these details are probably going to have to turn into init options so that computers don't get choked on all the calculations anyway, why not pick the ones that are headed that way, put in the placeholder code now, add some fun stuff to lure back the people on the non-simulationist side of the fence, and then go back and add the detailed stuff later?

Yeah, doing twice the amount of work for items that are already coming up in the near future sounds productive.

Spend months putting in placeholder code and then a year down the line gut all of that and make the "real version", all to gain a couple hundred more downloads and keep peoples interest for a couple of months before they figure out how to abuse it and get bored again. Then when he starts gutting the placeholder stuff you will probably return and complain about him gutting it and redoing it when he could be spending his time implementing other stuff you'd rather see and use the same argument: "Why don't you just keep the placeholder stuff and make more placeholder stuff bloo bloo blah blah mad libs teeth and beard simulation"

[Aquillion]

You all have many valid complaints, as should be expected.  It seems like the thing that would keep the most players from giving up is a good tutorial, in addition to other fixes.  I’d be interested in hearing how you think that should look.  For instance, one long tutorial, or several guides aimed at different aspects of the game.  What subjects are the most confusing?  Should the tutorial map fit in the world itself as a mission from the Mountainhomes for example?

I think a tutorial fit into the world itself would be the most fun for players -- basically a scripted "campaign mode" that is really a tutorial.  Give the players sequential requests to build and use certain structures (a farm with farming instructions, a dining hall, some bedrooms, and so forth.)

Oh, also.  The "Start now" setup should probably be improved; it's pretty un-optimal, I think.  Give players some of every seed, say, which is just common sense when starting out.  No animals aside from dogs -- they're not useful enough to a starting fort, and players get them automatically from immigrants anyway.  A good common-sense skill-assignment with one or two miners, one or two farmers, a chef, a fisherdwarf, a brewer, someone with decent social skills and so on -- players shouldn't have to read the wiki to determine what a good starting setup is; when they hit "start now" the game should provide something decent for them.

Oh, also!  Another option would be to make a 'scenario builder' that players could use to make their own scripted scenarios -- where they could set certain events to happen at certain times, or require that the player produce / do certain things first.  This could be used for all sorts of things, but it could also be used to let users create and update their own in-game tutorials.  You could then link to the best of those on the download page.

[G-Flex]

Yeah, doing twice the amount of work for items that are already coming up in the near future sounds productive.

Spend months putting in placeholder code and then a year down the line gut all of that and make the "real version", all to gain a couple hundred more downloads and keep peoples interest for a couple of months before they figure out how to abuse it and get bored again. Then when he starts gutting the placeholder stuff you will probably return and complain about him gutting it and redoing it when he could be spending his time implementing other stuff you'd rather see and use the same argument: "Why don't you just keep the placeholder stuff and make more placeholder stuff bloo bloo blah blah mad libs teeth and beard simulation"

Tell me about it.

I know Toady has to rely on donations, but I hope he doesn't ever end up making too many concessions in order to drum up downloads or a bigger player base.

Granted, I don't honestly think he will, but I've seen it from commercial developers so much that it seems like it's damn near the default.


I think some people just don't understand how much more work it would be to do all this piecemeal placeholder junk they seem to be suggesting, rather than actually designing the whole system (whichever we're talking about) at once. Take bodies, for instance. Yeah, beards aren't the most important things in the world, but did adding them really take much time out of development? If you look at how things have been going, you can tell the answer is no: He was already working on new body definitions and how they work, and the less-important bits were just a small part of all that.


I think one sort of user-help the game needs is a sort of live help system. Right now, when you view a flux stone or piece of ore, it says what reactions it's a part of. I can see it helping if the game did that for just about EVERYTHING: Telling you what uses something has when viewing it if you want, and telling you what's necessary for any given job at a shop.


Although, personally, the game-affecting issues I had mentioned previously have more bearing on what bugs me, although that's from the perspective of someone who had relatively little trouble getting into the game at first (although I had the wiki and a graphics pack available).


Speaking of graphics, I think one problem is how inefficient they are. The game could easily use 24-bit color instead of the limited palette it has now, and could use arbitrarily-sized tilesets. Right now, by default, it basically uses an ASCII tileset some of which isn't even used for anything. I know this is all obvious and everything, but I thought I'd throw it in instead of just yelling "TILESET SUPPORT ".

[tourettedog]

Yeah, doing twice the amount of work for items that are already coming up in the near future sounds productive.

Spend months putting in placeholder code and then a year down the line gut all of that and make the "real version"...

There'd be no more gutting involved than there is for the temperature init option right now.  Hell, he could have set most of this up through empty functions a while ago; and for all I know that's what he did.

You write something like this:

function calculate_weapon_damage(weapon_data, victim):
  if weapon_data has super_complex_materials_data then:
    // calculate it based on tensile strength and victim tissue information
  else:
    return dieroll(2d6+1)
end

And until you're ready to start implementing all that fun stuff you just never attach materials data to any weapon data.  Once you decide to start dealing with the tensile/shear strength of someone's armor versus the weapon being used, you go and fill in the code in the comment, write up and link your material data to the weapons and armor, and go from there.

There's some functional equivalent of this code implemented already, as evidenced by the fact that you can turn off things like the economy, weather, and temperature effects; all of which are pretty pervasive.

Meanwhile, huge amounts of detail is being added to stuff that could easily be abstracted out in a way that code for it could be dropped in later, while major gameplay issues are left unaddressed for months.

I know Toady has to rely on donations, but I hope he doesn't ever end up making too many concessions in order to drum up downloads or a bigger player base.

Unless the "I love playing Dwarven-dress-up" brigade is ready to stand up and support Toady all by themselves, they should accept the fact that he might actually have to turn his attention to immediate gameplay concerns to retain sufficient fans to support himself, rather than laying ever-more-elaborate foundations for a hugely ambitious and complex vision that may or may not ever actually be realizable.

[G-Flex]

Unless the "I love playing Dwarven-dress-up" brigade

If you're going to act like that, people aren't even going to read the rest of your posts, much less take them seriously. Realize that you're being a dismissive jerk not only to the people who like where the game is headed, but also to the person who decided to do it that way in the first place.

[tourettedog]

If you're going to act like that, people aren't even going to read the rest of your posts, much less take them seriously. Realize that you're being a dismissive jerk not only to the people who like where the game is headed, but also to the person who decided to do it that way in the first place.

So "myopic myopic can't read the design docs you're so short-sighted you enemy of art and I pity your ignorace" condescension is cool, but I finally take a dig at you and suddenly that's a foul?  Seriously, for someone so concerned with being "dismissive", have you actually re-read any of your own posts?

But whatever.  If I really did hurt your feelings, then I apologize.  I thought that it was pretty obvious that that line was hyperbole, and I assumed that it could be read in the context of your 'oh god please not too many concessions to the gameplay-oriented plebs' comment as a response in similar vein.

Now would you like to respond to any of the other points?  Specifically the one where I said the thing that turns me off about DF is that it has major gameplay issues (not only bugs but lack of content) that have languished for months, and that they're probably driving away players and potential sponsors?  And that there are small things that could be done now to improve gameplay that would not necessarily involve sacrificing long-term plans?

[Davion]

You write something like this:

function calculate_weapon_damage(weapon_data, victim):
  if weapon_data has super_complex_materials_data then:
    // calculate it based on tensile strength and victim tissue information
  else:
    return dieroll(2d6+1)
end

And making placeholder junk to implement sending out war parties to sack goblin forts, megafauna that can be captured and sent along with war parties, actual diplomacy options, being able to embargo a city and have them cave to demands, and slave trading/exchanging prisoners of war could probably be implemented in about five minutes right? Probably just a couple of dice rolls or something. 

[PMantix]

I asked the wife what would have made it easier for her to stay with the game when I first showed her (she played her own fort for a while, enjoyed it, but then gave up on it..   and now only gives me the look when she hears "Beyond Quality!", heh)

Here are her suggestions as an 'outsider' who hadn't been sucked in:

1. A quick help. She was completely overwhelmed by the massive amount of things that can be done. What she would have liked to see was the ability to click a word or object and have some kind of helpful hint pop up to briefly explain the word. For example, she was making her starting 7 and wasn't familiar with what a thresher was or did. I wasn't around to explain it or show her the wiki, so she just skipped it..   what would have made the difference was a quick pop up text bubble stating what the job is, inputs/outputs, and associated buildings/areas. Same goes for objects..  "this is a throne. It is needed as seating adjacent to tables in dining rooms. It can be designated as an office for administrative purposes or as a throne room for nobles." That way you can be quickly informed about what something is, instead of the default "what the hell is that" or "how the hell do I make this"

2. Direction. I think the reason she quit was because she just didn't know what to do anymore. She liked gathering wood, and building beds, and digging out bedrooms and making every little dwarf their own private room. But once she had done that she asked me, "ok, now what?" And I, of course, said, "well, you can do basically whatever you want. Build a grand dining hall. Make some crafts. Train some warriors and beat the hell out of some goblins. Trade with the damn tree huggers. Anything really." So she then goes and figured out how to gather, cut, and encrust some gems on things she had made. But without any real direction to go, or some concrete objective dictated by the game, she just lost interest. I think the nobles demands are the right kind of idea to give this kind of motivation, they are just implemented in such a way that makes you want to kill them instead of fulfill their request. I think it really boils down to the fact that nobles are being complete selfish pricks and only asking for things that they want when they should actually be demanding things like, "Too many lives were lost last season to the siege. Build defensive marksdwarf towers at the main entrance to protect our hoard!" Easier said than done I know..   but giving an objective that isn't based on the fact that your duke likes pigtail socks would be a massive improvement in my opinion. Bonus points for the demands being based on the history of the fort (siege aftermath (like in the example above), food stockpiles, work availability, lack of beds/wells/soldiers, etc) and it gives useful objectives to guide new players when they aren't sure what they should be doing.



Those were her two big ideas for what could have made DF work for her, fleshed out by yours truely.

[tourettedog]

And making placeholder junk to implement sending out war parties to sack goblin forts, megafauna that can be captured and sent along with war parties, actual diplomacy options, being able to embargo a city and have them cave to demands, and slave trading/exchanging prisoners of war could probably be implemented in about five minutes right? Probably just a couple of dice rolls or something. 

Excellent point.  Because that totally wasn't just a 6-line pseudocode example written to indicate how he could do it without setting up a hack he'd have to pull out later. 

And anyway, he clearly hasn't already done something similar with an init option for sieges.  Or temperature.  Or cave-ins.

[Aquillion]

Let's stop talking about placeholders.  Yes, yes, we know some people want it and others don't.  Toady is equipped to know exactly how much of a bother it would be.  Comparatively speaking, we aren't.  Therefore, there's nothing else for us to discuss there.  That discussion is totally useless to Toady and Threetoe; it doesn't answer their question at all, because Toady is going to already know the instant he looks at the suggestion what the issues are (or, at least, after thinking about it for a bit, he'll know better than we could.)  Some people saying that they want it is good, but beyond that nothing you people are saying at this point is any real use to them.

1. A quick help. She was completely overwhelmed by the massive amount of things that can be done. What she would have liked to see was the ability to click a word or object and have some kind of helpful hint pop up to briefly explain the word. For example, she was making her starting 7 and wasn't familiar with what a thresher was or did. I wasn't around to explain it or show her the wiki, so she just skipped it..   what would have made the difference was a quick pop up text bubble stating what the job is, inputs/outputs, and associated buildings/areas. Same goes for objects..  "this is a throne. It is needed as seating adjacent to tables in dining rooms. It can be designated as an office for administrative purposes or as a throne room for nobles." That way you can be quickly informed about what something is, instead of the default "what the hell is that" or "how the hell do I make this"

This is a good idea.  The '?' help would be much more useful if it was context-sensitive and brought up information about whatever your cursor is over at the moment.  It would especially be nice to be able to get information about workshop tasks, too, learning what each task requires and produces.