Print

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

[Dellnak]

Greetings everyone.

To begin with, I did some searching and I know several of my thoughts have been discussed before, but I hope to provide more "why" than "what" by explaining what I love the most in DF and how that could be expanded. I should also mention that I think DF is a wonderful game as is and I am not sure if all my suggestions would be good for it, as they stand on the edge of a dangerous ground. I don't in no way mean to tell Toady how he should develop his game, but I want to offer a view of what makes DF so good a game for me.

I love designing and building systems, and DF is a wonderful playground for that. The most fun I've had in it was building a water network for my fortress, with several well reservoirs, a waterfall, a drowning chamber, mudding of underground forest sites, with good modularity for easy separate expanding and maintenance so that modifying one or building a new system will be risk-free for dwarves and disturb other systems only slightly. All water sections could be separately emptied practically infinite time, as the evaporation chambers were quite big, and there were lot of doors for easy access to pipes and for limiting a possible flooding. This is probably quite basic stuff, but it was very fun to figure out how to do that and actually see the whole thing working perfectly, with zero dwarves getting dangerously wet anywhere in the process.

The point was not to actually make the fortress super-optimal. I don't think my king will appreciate his personal well, because he likes beer. Also as a newbie I picked an easy site and haven't needed to use the drowning chamber even once. That is not the point. The point is learning how to build a system that _could_ be useful in a very wide sense, either currently, possibly in future or in personal dreams. The  purpose of most games is to have fun, and I think systems are very fun. I think I am not the only one, which is why I though it might be useful for these thoughts to exist outside my own head.


I was studying thermodynamics the other day and couldn't help to think how awesome it would be to be actually able to build your own engines or even power plants. (I know this is not fantasy-like, and I'll address that later in the post.) It's not that important what is the source of heat (coal, nuclear, something else), but the whole design of the plant (essentially the transfer of the heat and a way to convert heat to power).

Of course if there are many uses for bare generation of heat itself if temperature got a greater part, such as central heating or cooling for the fortress of sub-optimal natural temperature, some plants requiring specific temperature range, food preservation, drink quality (who the hell likes warm beer?), etc. The transfer of heat is probably quite demanding for the CPU, but I think it could be doable with some approximations. The conduction checks interval could be longer, maybe even dependent on the temperature, so the very hot could still burn stuff fast but for small fluctuations or even steady states the temperature can be updated less often.

This is quite closely linked to steam, which is at least slightly fantasy-like. There seems to be many steam-related topics already so I only briefly describe what aspects a steam system in my opinion should have. Transformation to liquid at the specific temperature would be very nice. Some form of pressure is a must as is conduction. Also the basics of thermodynamics such as cooling when expanding and heating when compressed (non-isothermically) would be nice [I'm not sure if the dwarves want to produce liquid nitrogen, but in that case Joule-Kelvin effect would be needed also]. I think the system can be quite simple so that there is no need to model the air in the fortress and outside as a separate gas and play with partial pressures, I'm happy with just gases other than air modeled (although I think vacuum-chambering would be a nice execution method).

Gases and liquids would also benefit greatly from pipes which would make moving them around much easier, but I think this is a recognized enough thing already so I'll not expand the subject.

Electricity would be a way of making power more useful, by allowing the storage of it in an abstract form and the transfer of it more easily and for longer distances. It would also allow some new technologies, such as artificial light.

[EDIT]
In my opinion electricity didn't need to be universally conducting through all the materials, because that would be unavoidably bad for the FPS. We could have some special conducting materials so that it's the decision of a player with a supercomputer to build a fortress out of copper and install a frying switch to a unnecessarily central location, if he so wishes.
[/EDIT]

There are also many purely mechanical things that could make life more interesting. Floating platforms would be cool, for example for water-powered lifts, or reversely for pushing water to a higher level. This would require modifications to water pressure system, but I don't think that's an entirely bad thing. If gases are implemented pressure system has to be remade in any case.

Floating platforms could be generalized to any movable constructs and water could be made to apply force to other stuff than itself, too. So in addition to the lift I could build a movable wall, put pressurized water to one side of it and watch the wall move and possibly squish enemies on the other side. With that made it would be cruel not to make some purely mechanical way of moving (pushing and pulling) things around. Build a hook to a movable wall, link the hook with a rope of length x to a windmill, and see the wall being moved.

I'm not sure about firearms. They could be semi-logically be prevented fluff-wise by making gunpowder impossible to make for some reason, but I don't think they would necessarily be a bad thing.

Some form of remote communication might be nice. There are small uses for this, such as a lighthouse dwarf living thousands of tiles away from his family staying in touch, mayor making last-minute import requests to Mountainhome and you getting early warnings about immigrants, ambushes, etc. You could also build measurement devices, which would directly display their measurements in some screen for the player (water levels of different reservoirs, outside temperature, amount of food on stockpiles #7, #39 and #114, etc). The larger uses would require new features and I'm not sure if they would be useful. For example, if the dwarves had artillery, a spotter could report where the enemy is, so that the cannons could shoot to a correct position.

Maybe the "communications" should be more like remote controlling of things. Levers could have some remote device built on them and the general of the military could have in his office a machine with which he could pull any lever with a remote device. Of course the levers could be skipped altogether, linking the original target straightly via a remote device. The point of this would not to reduce walking times, but to make building new systems necessary! The different methods of communication would each have their limits. You can build a cable to a remote location, but it's slow, possibly expensive, cable is easy to sabotage, etc. Or you could build a transmitter and a receiver, but then the antennas have limited ranges and don't work well through stone. Or you could use smoke signals, but boy it is not easy for the programming dwarf to build an algorithm for the receiving computer that understands those.

Now about the level of technology. Judging from the forum posters' obsession to year 1400, I assume it's an official statement for the level of technology. Still, I think there should be no fundamental reason why some of the more advanced technologies couldn't exist.

I don't think it is necessary to stick with some time period. Actually I think it would be wonderful to be able to begin the game in stone-age with more advanced things like windmills and water pumps didn't exist, and then actually devote some smart (new ability! could also affect judging of intent, appraising, etc) dwarves to research, which in time could result in new technologies being discovered. Some of them could maybe also be bought from Mountainhome or even some other friendly civilizations. The existence of fantasy races does not conflict with advanced technology in any way. Allowing advanced technologies would however be very dangerous. DF could possibly not feel like a DF anymore. For a long time I didn't even build windmills or water wheels, because I felt they are not dwarfish. There could also be problems with why isn't real-world technology X in the game even though Y is and IRL X was discovered before Y. Some of these should be easy to avoid by remembering that the technology trees of societies are not guaranteed to progress similarly and dwarves have different priorities than humans, but I understand there would be many potentially unrealistic advancements. 

I apologize for the relatively bad structure and repeating of some old subjects in the post. I don't claim these are the most needed things in DF, but as I've been fantasizing about these for a long time, I thought I'd write them down, and post them here.

[bjlong]

Toady has repeatedly stated that he wasn't willing to work on technologies that are post-1400's. This unfortunately stops most of your suggestions, with the dubious exception of cannons used as seige equipment.

As for different ages and researching stuffs, we've had discussions about those in the past. If you can come up with a system that allows people to research without feeling metagamey to the detractors, I'm sure we'd all love to hear it. But know that such a system will take plenty of time to make work on Toady's part, so it'll be a ways off.

[Dellnak]

Toady has repeatedly stated that he wasn't willing to work on technologies that are post-1400's. This unfortunately stops most of your suggestions, with the dubious exception of cannons used as seige equipment.

Well, I guess I have to accept that. I want to emphasize I think DF will probably be the best that way, but I can't stop thinking how fun more complex systems would be to build.

However, I do think that using floating platforms in water pipes as elevators would not be post-1400, as wouldn't using walls movable by water or pushing/pulling.

In addition I'm not sure if the criteria is when the actual technology was invented or when it could have been invented. What I mean is that given that some dwarf colonies have easy access to essentially unlimited supply of heat (and a way to produce not easily melted machinery) and water, steam power (maybe ineffective, but in principle) would probably be very plausible for them. Fluff-wise water wheel should be trivial to modify to a steam wheel. Pour some water to magma, shut the pouring hole and watch the steam escape via another hole, which has the steam wheel. There you have steam power, made with all the equipment you already have.

I suspect for real humans the easy source of volcanoy amount of heat was a problem, because the net energy for getting the water to the top of a volcano and using it in such an ineffective machine would have been very negative (being lousy miners they did not mine channels to underground magma).

As for different ages and researching stuffs, we've had discussions about those in the past. If you can come up with a system that allows people to research without feeling metagamey to the detractors, I'm sure we'd all love to hear it. But know that such a system will take plenty of time to make work on Toady's part, so it'll be a ways off.

I'm not sure if I understand the metagaming problem here. Is the problem that if there are several possible research paths, the dwarves could possibly not know which path leads to which actual technology even though the player does? To be honest, I fail to see how that is a real problem.

The technologies don't have to be anything revolutionary such as "play around with variables energy, mass and the square of light speed and see if you find any relations", more like "see if you can harness the kinetic energy of water/wind/water vapor somehow...ah yes nice wheel you have here, but could you please invent something that lets us transfer that rotation to a remote location....wow nice axle but how are we supposed to handle corners...." or "experiment with different amounts of stuff to mix with iron and see what kind of metal alloys you get". The point is that the dwarves approximately know what they are going to get, because the individual steps are small.

Each new discovery requires x timeunits of active researching. So if you have just built an "Armchair and a blackboard" workshop and designated one dwarf to have the research labor, then you can set up a job at the workshop to research water power. The dwarf does not need any raw materials or anything, otherwise it functions exactly as other workshops.

Say the first "level" of water power needs 100 timeunits of researching. After your researcher has done 60 timeunits, you build another workshop and designate another dwarf, after which each of them has to do only 20 timeunits and the 100 is complete. Of course the actual times must vary with skill and attributes, and there should be some limit for parallel researching, either hard cap of dwarfs per project or diminishing results per additional dwarf to a project. The parallel researching could be altogether prevented by making research projects workshop-specific like workshop jobs are now, but I think it would be nicer to be able to do that. The research projects should be tracked globally, but I don't think that would be a huge problem.

[Granite26]

Holy Wall of Text, Batman!!!

Formatting would help... Bold topic headers, etc...

As for different ages and researching stuffs, we've had discussions about those in the past. If you can come up with a system that allows people to research without feeling metagamey to the detractors, I'm sure we'd all love to hear it.

I'm not sure if I understand the metagaming problem here.

Each new discovery requires x timeunits of active researching.

Say the first "level" of water power needs 100 timeunits of researching.

There's your explanation...

[Dellnak]

Holy Wall of Text, Batman!!!

Formatting would help... Bold topic headers, etc...

I think for a text of only ~1600 words, there is no need for titles or any other hi-tech text stuff. I know (I really do) long texts can be intimidating to read, and I will consider adding bold headers next time I'll write a long post.

Each new discovery requires x timeunits of active researching.

Say the first "level" of water power needs 100 timeunits of researching.

There's your explanation...

I'm sorry, but I do not see it. The timeunits are obviously replaced by real time units in the actual game. Is it better if I say it takes a year of nonstop working for one dwarf to research a water wheel (6 months for two)? It's essentially the same thing, but I did not want to set any scale so I used abstract "timeunit" in the place of a real time unit.

[Granite26]

Having a researching working at a desk accumulating research points until BAM he 'discovers' a tech is gamey.

IRL, research is 25% persperation, 25% population density, 25% tech synergy and 25+% inspiration. 

Most modern technologies were around long before they were 'discovered' or widely used.  It took other inventions to make them feasible and a certain population level for them to be profitably applied.

This is especially true in the time period we're looking at.  Compare to da Vinci, who is known for a large number of inventions that weren't feasible at the time.  (Or really the scientific advancements of the entire renaissance).  Most research models in games are based on the Manhattan project, not the simple gradual improvement that's more historically accurate.

[Dellnak]

Having a researching working at a desk accumulating research points until BAM he 'discovers' a tech is gamey.
[...]
Most research models in games are based on the Manhattan project, not the simple gradual improvement that's more historically accurate.

Well, that's how initial innovations work. You either think and/or experiment and get no useful results, until one day you come up something that works. After the initial prototype exists there is of course much room for improvement, but the first working piece of a new technology really comes as BAM.

[Granite26]

Having a researching working at a desk accumulating research points until BAM he 'discovers' a tech is gamey.
[...]
Most research models in games are based on the Manhattan project, not the simple gradual improvement that's more historically accurate.

Well, that's how initial innovations work. You either think and/or experiment and get no useful results, until one day you come up something that works. After the initial prototype exists there is of course much room for improvement, but the first working piece of a new technology really comes as BAM.

Except that's not true at all.  The big scientists didn't really DO anything, it was all just words.  Count the number of medeival technical advances that were the product of scientific method versus those that came about by some shmo in the trenches saying 'this isn't the best way to do this anymore'.  For every lightbulb and nuclear weapon there are a thousand compasses or cotton gins or movable type printing presses.

[bjlong]

Really, even now we have research that goes more like "This might be interesting somewhere down the line, but as for now? It's a ways off." Things get funded, yes, but these things are small improvements on previous theories or innovations. The small improvements can lead to interesting findings, which can then lead to unexpected synergies. For example, Maxwell unified E&M, but the real interesting thing was when he discovered EM waves have exactly the speed of light. This is the cornerstone of most optics, now, where as before geometrical optics was the tool that people had for understanding how light works.

Now, I'm not saying that researching x tech shouldn't be possible. (In fact, with some civs, it would be necessary as magic becomes a part of the game.) Instead, I'm saying it should be a game of chance, where you hope you've put your eggs into the right basket to get that new spell for killing things. But it might turn out to be a dead end, or open up a new branch of healing stuffs instead.

[Dellnak]

Except that's not true at all.  The big scientists didn't really DO anything, it was all just words.  Count the number of medeival technical advances that were the product of scientific method versus those that came about by some shmo in the trenches saying 'this isn't the best way to do this anymore'.  For every lightbulb and nuclear weapon there are a thousand compasses or cotton gins or movable type printing presses.

First, a new piece of technology (be it completely new or an improvement to a previous one) needs to be imagined by someone. In this process it is helpful to be smart and imaginative (they could probably be under the same attribute) and know as much as possible about the field related to the technology. It is also useful to have a systematic approach to experimenting (either physical or mental) so that you work faster because you don't solve a same problem twice, don't repeat your mistakes, make good notes, etc.

Working as a team in helpful because you can usually divide the workload somehow, but also because a colleague can sometimes see problems from a different angle and thus solve it faster than you could by yourself.

So, the research "speed" of a dwarf is some kind of (possibly weighted) sum or product of his intelligence attribute, his researcher skill and his skill of the field of research, if that is applicable. So a miner would no doubt be better at researching for more effective pick tips than a brewer, if they were equally smart and researchy. I guess the shmo science you mentioned was because they had superior practical skills and so could actually see what needed improving. If there is no applicable field (as could be the case with completely new technologies), no skill level is used in calculations and the speed is a function of only intelligence and research skill.

However I do claim that researching should be an active job, though not necessarily done in a specialized workshop. I have always found it hard to think creatively when working hard, and I can't believe my dwarves are not working hard. I have nothing against smart dwarves spontaneously thinking about a problem while on breaks or nojobbing, which could even increase their research skill. However, that shouldn't contribute to a project unless the dwarf was the main researcher of it or reported his results to a researcher of the project.

I think it would be best (even if not completely realistic) if only the dwarves with research labor enabled and a specific research project designated would contribute the research, because if the researcher had to listen to the stupid suggestions of hundreds of stupid dwarves every day, the project would probably go backwards more than forward. Of course the mechanic behind the researching can be made arbitrarily complex, but I think overcomplexity should be avoided. Research laborers with no project could still train the skill by brainstorming something and in the process perhaps getting a good idea and starting a new project. However, the in addition to the innovation a working piece of technology requires a certain amount of designing and engineering, which must be done with a clear goal in mind. It might be good idea to really separate the innovation and engineering parts, with former producing only new technologies and latter making existing technologies better. This has a potential problem that researching is given too much weight and so I think it should be kept as simple as possible, which would mean a single workshop requiring a single labour.

Of course the research goals can be completely random, but I think that would be too frustrating to the player and it is not IMO realistic. While luck does have a major part in innovations, specific research goals can be viewed as almost compulsory prerequisites. You concentrate your thinking to certain aspects of the world, and which is more important, pay attention to things you're researching. I think a fair compromise would be to make research times non-deterministic. They could be drawn from a distribution (say, Gaussian) with relatively large mean, at the time of someone first spending any time on it.

[Granite26]

Yeah... you obviously aren't reading our replies.  I'm done with this

[bjlong]

Yeah... you obviously aren't reading our replies. 

[Dellnak]

Yeah... you obviously aren't reading our replies.  I'm done with this

That is your right as a fellow poster, but I can assure you I have read the replies multiple times. I am not sure where our disagreement is, but I guess it's either the existence of innovations or the role of full-time researchers. Either we have a different view of how the game should be or either one did not understand what the other said. Whatever the reason, I cannot solve the disagreement by myself.

I hope another poster joins the discussion and offers a new opinion.

[Mephansteras]

The disconnect lies in your attempt to apply the scientific method to the time period DF is in. But really, with the exception of a few rich alchemists or philosophers, there was no such thing as a dedicated researcher. You didn't have people going off to labs or whatnot, nor was there any discipline in making hypothesis and testing them. You also didn't have societies directing research for the most part. The vast majority of the time you had improvements come about because someone with practical experience in an area was messing around with something or accidentally found a better way.

Getting a better pick head, for example, is not going to be driven by the fortress going "Research better mining techniques". It's going to be one of your miners and blacksmiths going "You know...if we shaped it this way you'd probably get a better strike out of the pick". It'd be completely random and not something that the player should plan for or really direct.

The only real research you're likely to get is a philosopher or alchemist noble (maybe the Dungeon Master) taking ores that don't have any real use yet and trying to figure them out. But that's not something that should be directed or necessarily useful for the fortress. You'd just get the noble walking around and grabbing random chunks of cinnabar or whatever and messing with it in the smelter. Maybe he'd figure out how to do something with it, maybe not, but it'd be very undirected research.

The other thing to keep in mind is that research in this time period is slow. You might get an improvement once every hundred years or so, unless you go lucky and some tech from somewhere else was brought in. Really not something the player is going to get involved in. Even going from Iron to Steel is a process that took centuries, in some places thousands of years (anyone who figured out the secret guarded it closely). You'd have to run a civilization for a very, very long time to make that change.

Basically...I don't think it's very practical to worry about technological advances from a Fortress Mode perspective. Maybe from a high level Civ perspective. But even there you don't have a solid tech tree to go off of. India apparently had a reliable damascus steel process for about a thousand years that they kept from their neighbors. And the Chinese had advances like gunpowder and repeating crossbows long before Europeans had anything of the sort. There was also a civilization in Africa that happened upon iron working very early on, and never even bothered with bronze for weapons.

Games like to make technology into a nice static tree that is followed neatly, but it's really not the case in real life. Technology progression is random and very heavily dependent on necessity and luck.

[neek]

The issue at hand is that most games demand a think-tank to garner technologies, as opposed to a stroke of genius. What you propose is the same as most games that deal with technology: You tell them what you want researched, there's a definable tree so you can metagame the tree and figure out which discoveries yield different discoveries. Sort of like telling your scientists in Civilization to think until they get you gunpowder, and it's only 4000 BC. You know what they want, and there's a specific plot they have to take. Even with a tree, this makes researching rather linear, which doesn't fit in with Dwarf Fortress: This game is absurdly not linear.

In reality, most of the modern inventions that we use today weren't the product of think tanks or research labs that have a definitive goal; these are modern concepts. The President didn't ask Henry Ford to research Assembly Line Production so we can get Automobile; rather, he took an innovative approach to the industry. Investors didn't tell the original computer companies to research integrated chips and microprocessors; rather, someone had the idea and proposed it to the investors. Da Vinci was part of a think tank. Neither were the Greeks who built the Antikythera Mechanism.

There's a BBC production that describes how technologies came around. I wish I could tell you its name, because it is invaluable in this sort of discussion. A few major ones: The pocket watch, pendulum clock, water clock, and mechanical clocks all rely on the same physical properties. It took one person to realize that the energy released of a suspended spinning stone can also be expressed as a pendulum; that it can also be stored in a spring made it even more portable. The smelting of certain metals became even more prominent when a man realized that the clay vessels used in glass making heat up (by glowing white hot and reflecting thermal radiation back into the center) to far greater temperatures than thought possible. The movable type printing press was simply a realization that the Greek olive press can be useful for more than olives.

But that not being the point, the main disagreement here is that suggesting a think-tank like structure to propel the technology of your Dwarves doesn't fit the scheme of the game. We're looking at a pre-Renaissance, no scientific method using culture. Asking them to research something is about as futile as asking a dwarf to take from the nearest stone stockpile (rather than the more preferable Dwarvenly way, walking across the map and getting mauled by a Sasquatch.)

... or, what Meph said.