Print

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

I would like mechanisms to break down completly random (with init)

I would like mechanisms to break down with maintaince(with init)

I will stop playing df if mechanisms can break down even with an init

I think mechanisms should break down and there should not be an init to chance it

What is a mechanism and why should I care I just want to burn elves!!!

[forsaken1111]

And Urist shouted down the hole to the new engineering recruit, "What can you see?"

The recruit shouted back, "It's all jammed up!"

Urist said to him, "Just whack it a few times with a hammer, should get it loose."

"Okay!" There came the sound of two solid blows, a snap, and then the crash of rushing water.

Urist nodded, satisfied at a job well done, and closed the floor hatch. Another mechanism repaired, another cheese maker drowned. Life was good as chief engineer.

Whistling to himself, he set off to order a new hammer and put in a request for a new recruit.

[thijser]

I think it would also make sense if the speed at which a mechanisms wears down would be related to the object it's used in and the conditions it's build in.

I think that things that should have an affect on this are:
1 Ground on which it is build if it's sand or mud it will increase the speed at which is wears down.
2 Weather: if an object is exposed to the weather it will wear down faster, if it's build outside it will dirrectly increase the speed at which it wears down (wind and small particals getting inside the mechanism) aswell as rain which does a large amout of damage.
3 Blood and vomit, it somewhat increases the speed at which items wear down.

[NW_Kohaku]

...because traps are so underpowered.

Anyway, it seems reasonable to me that a fortress with heavy reliance on mechanical contraptions should need a bit of support staff. I think it would be neat to have a bunch of mechanics scurrying about keeping things ship-shape. Dwarfy. That said, I agree that mechanical failure should be essentially preventable, and that the player should feel "oops" rather than "what the q***!" when it happens. Degree of shoddiness should be displayed.

Other considerations aside, I also agree that heavy and continuous use should be more demanding. Failures should be a Poisson process, i.e. as time passes without maintenance, more and more things start to fail, but there's no upper bound on the amount of time to failure. (Ancient traps/mechanisms discovered in working condition = fun)

The problem with that is that your basic traps are going to take up about 30-50 mechanisms.

A Four-function calculator or a dwarven Turing machine take tens if not  hundreds of thousands of mechanisms.

That's several orders of magnitude more mechanisms, which in turn means that even if you only needed one dwarf just maintaining 40 mechanisms, then to maintain 40,000 mechanisms, you need 1,000 dwarves on nothing but maintainance duty.  Plus, you know, all the support staff to feed and clothe and protect them, plus the dwarves to build the project.

Plus what happens when all your doors get jammed stuck, you don't see the message, and all of a sudden, half your fort starves to death because they were locked out of the food warehouse, or the drain to the waterfall got clogged, and the fortress flooded.

I only see this as a way to randomly spring punishment on players for not weeding out the one important message in all the message spam, and make the truly great megaprojects impossible.

[thijser]

The problem with that is that your basic traps are going to take up about 30-50 mechanisms.

A Four-function calculator or a dwarven Turing machine take tens if not  hundreds of thousands of mechanisms.

Because of this you could make it into an init option. Also how many of these things were really made I believe it's only like 1 or 2. This doesn't seem like a good argument "we can't have it because those few who have set it opon themself to do the most difficult thing passble in this game would have much more trouble".

[NW_Kohaku]

Yes, there are only a few "most incredible" megaprojects, but that is a function only of your definition of "most incredible".

It isn't a matter of how many people use it, but of whether we want to encourage or punish those people who go beyond the pale to show what Dwarf Fortress can truly do.  I'd rather encourage people to understand the mechanics of the game and build greater things than shut them down, even if you give people a "cheater option" to still make it technically possible.

[thijser]

I think they should think about it as a greater challange. They produce 4 times as many mechanisms to make shure they only use masterworks. Or just cheat a little. I don't think this should stop them. What they are doing is already impressive and this could be seen as the speed 0 trick. If they can build this then it's only a matter of time before they can make 4 times as many mechanisms, so it's "fair" that they are allowed to cheat so that they don't have to maintain everything. Cheating in df isn't something to worry about it can be considered "fair" if the related megaproject is large enough.

[Antsan]

You can make computers fail proof to a degree, so this would make the challenge harder but not impossible. Computers weren't invented in the middle ages for a reason and Babbage failed with his computer certainly not because he wasn't able to do the math. Real computers have wear, too, and they're not even mechanical.

Consider a dwarfen computer consisting of 500,000 mechanisms.
I now will try to estimate how many dwarfes you will need, when the half life of mechanisms is at one year.
One year has 1,200*336=403,200 time units in a year (according to the wiki). A delayed action takes 100 steps, so I assume an average dwarf will take 100 steps to control a mechanism and 100 steps to repair it in case of wear. Those are 500,000*100 steps for controlling and 250,000*100 steps for repair per year, together 75,000,000 time steps. So we need [75,000,000/403,200]=187 dwarves. This is a rather high number, but I guess a masterful mechanisms half life should be 10, 20 or maybe even 100 years and not one.
You can tweak the half life of a mechanism and the time a dwarf takes for maintenance so that this number shrinks with masterful or maybe exceptional mechanisms to a reasonable number, so I don't see any problem with that. Statistics are your friend and it's all just a question of balance (even if my calculations should be total rubbish).

[Agamemnon]

I support this only if masterwork mechanisms remain failproof.

Then anyone doing megaprojects with mechanisms just has a small delay and thousands of mechanisms of inferior quality get added to the "bottom of the magmapipe" stockpile.

Using mechanism failure from inferior quality to nerf cagetraps a bit sounds like a reasonable idea. But essential things like bridges and floodgates should remain operational in any case. Making a dwarf pull the right lever in time is already hard enough. Maybe we could build multiple redundant mechanisms into those. When one jams, the next takes over.

[Antsan]

But essential things like bridges and floodgates should remain operational in any case. Making a dwarf pull the right lever in time is already hard enough. Maybe we could build multiple redundant mechanisms into those. When one jams, the next takes over.

That is what's done to make fail proof machines and computers in real life, iirc. Being able to put multiple mechanisms into everything using mechanisms seems good to me.

[thijser]

Well I think that what they are beeing used for should also influance how easy they fail.
A lever is a simple mechanism and won't wear down quite as easily. On the other hand a cage trap is a complex device that even in real live is difficult so it should fail quite quickly.

I think that this order makes sense:
cage trap
weapon trap
support
bridge
stone trap
pressure plate
floodgate/hatch
pump
lever
gear assambly

[Untelligent]

I kind of like this idea.

Obviously if it went in, there'd be an init option to toggle it. There's no point in discussing that bit.

[Threlicus]

I support mechanical failure, as long as it is also possible to schedule and perform maintenance which prevents it (perhaps an appointed noble position that schedules engineers' jobs for maintenance). The constraint on how many traps and such you have should be how much time your engineers have to keep your traps in working order, not how often you can press a->t and b->T->(trap of choice). Agreed that we should have masterwork mechanisms be failure-proof.

The other advantage of adding maintenance is it means that players will need to either use masterwork mechanisms in inaccessible places, or build in maintenance access shafts -- which of course are themselves a means of entry to your fortress. Yay!

[Mel_Vixen]

Depending on the material and a average strain i would bet most mechanisms will work atleast for years if not decades before they fail. Ok stuff made from sandstone or wood might break apart earlier if the exposed say to a harsh environments.

The average dwarven fort has not to many mechanisms so one or two mechanics might be sufficient. Computers (and control devises) are rather rare and might need more maintaince but as said if a cog does not break for a decade or two a minor percentage of dwarves to maintain them should be enough.

I furthermore suggest that gears and other mechnical parts wear more down when they are moving and interacting with other stuff so if a mechanism stands still it should only experience damage from the environment if any.

[sweitx]

...because traps are so underpowered.

Anyway, it seems reasonable to me that a fortress with heavy reliance on mechanical contraptions should need a bit of support staff. I think it would be neat to have a bunch of mechanics scurrying about keeping things ship-shape. Dwarfy. That said, I agree that mechanical failure should be essentially preventable, and that the player should feel "oops" rather than "what the q***!" when it happens. Degree of shoddiness should be displayed.

Other considerations aside, I also agree that heavy and continuous use should be more demanding. Failures should be a Poisson process, i.e. as time passes without maintenance, more and more things start to fail, but there's no upper bound on the amount of time to failure. (Ancient traps/mechanisms discovered in working condition = fun)

The problem with that is that your basic traps are going to take up about 30-50 mechanisms.

A Four-function calculator or a dwarven Turing machine take tens if not  hundreds of thousands of mechanisms.

That's several orders of magnitude more mechanisms, which in turn means that even if you only needed one dwarf just maintaining 40 mechanisms, then to maintain 40,000 mechanisms, you need 1,000 dwarves on nothing but maintainance duty.  Plus, you know, all the support staff to feed and clothe and protect them, plus the dwarves to build the project.

Plus what happens when all your doors get jammed stuck, you don't see the message, and all of a sudden, half your fort starves to death because they were locked out of the food warehouse, or the drain to the waterfall got clogged, and the fortress flooded.

I only see this as a way to randomly spring punishment on players for not weeding out the one important message in all the message spam, and make the truly great megaprojects impossible.

There's some talk about making this into an init option.
And have broken mechanism automatically generate a "fix me" option.
Or have mechanism always goes through the following 3 state (check on a seasonal basis?).
1. Working - Mechanism work as intended.  May advance to 2A or 2B.
2A. Dirty - Mechanism still working but may jam or break in a season or more (will work for at least a season).  Will trigger a maintainance request for mechanics to come and clean/maintain it.  Can only advnce to 3A when it comes up.
2B. Worn - Mechanism still working but may jam or break in a season or more (will work for at least a season).  Will trigger a maintainance request for mechanics to come and replace it (takes a new mechanism).
3A. Jammed - Mechanism no longer work require mechanic to go and unjam it.  Can advance to either 3A or 3B.
3B. Broken - Mechanism no longer work, the structure is still intact, but now requier mechanics hauling a new mechanism to replace it (structures depending on mechanism such as gears will not deconstruct, but simply have the mechanism component labelled as broken).

The path (A or B) taken depends on the quality, material, and the time in use of said mechanism.

Higher quality mechanism will advance from 1 ~ 3 much slower and will more likely take the A route versus the B route.  No quality mechanism may need replacement almost every year.  Masterwork should probably take about 2 years to go from working to worn and several years to need replacement.  Artifact mechanism would never break, but will still need cleaning (only take A path).

This mean that early on, you must rely on less mechanism (or hand-pick for masterwork to reduce maintenance) else
Mechanism material should also have an effect, for example, steel mechanism should triple the time needed between maintainence/replacement.

[thijser]

...because traps are so underpowered.

Anyway, it seems reasonable to me that a fortress with heavy reliance on mechanical contraptions should need a bit of support staff. I think it would be neat to have a bunch of mechanics scurrying about keeping things ship-shape. Dwarfy. That said, I agree that mechanical failure should be essentially preventable, and that the player should feel "oops" rather than "what the q***!" when it happens. Degree of shoddiness should be displayed.

Other considerations aside, I also agree that heavy and continuous use should be more demanding. Failures should be a Poisson process, i.e. as time passes without maintenance, more and more things start to fail, but there's no upper bound on the amount of time to failure. (Ancient traps/mechanisms discovered in working condition = fun)

The problem with that is that your basic traps are going to take up about 30-50 mechanisms.

A Four-function calculator or a dwarven Turing machine take tens if not  hundreds of thousands of mechanisms.

That's several orders of magnitude more mechanisms, which in turn means that even if you only needed one dwarf just maintaining 40 mechanisms, then to maintain 40,000 mechanisms, you need 1,000 dwarves on nothing but maintainance duty.  Plus, you know, all the support staff to feed and clothe and protect them, plus the dwarves to build the project.

Plus what happens when all your doors get jammed stuck, you don't see the message, and all of a sudden, half your fort starves to death because they were locked out of the food warehouse, or the drain to the waterfall got clogged, and the fortress flooded.

I only see this as a way to randomly spring punishment on players for not weeding out the one important message in all the message spam, and make the truly great megaprojects impossible.

There's some talk about making this into an init option.
And have broken mechanism automatically generate a "fix me" option.
Or have mechanism always goes through the following 3 state (check on a seasonal basis?).
1. Working - Mechanism work as intended.  May advance to 2A or 2B.
2A. Dirty - Mechanism still working but may jam or break in a season or more (will work for at least a season).  Will trigger a maintainance request for mechanics to come and clean/maintain it.  Can only advnce to 3A when it comes up.
2B. Worn - Mechanism still working but may jam or break in a season or more (will work for at least a season).  Will trigger a maintainance request for mechanics to come and replace it (takes a new mechanism).
3A. Jammed - Mechanism no longer work require mechanic to go and unjam it.  Can advance to either 3A or 3B.
3B. Broken - Mechanism no longer work, the structure is still intact, but now requier mechanics hauling a new mechanism to replace it (structures depending on mechanism such as gears will not deconstruct, but simply have the mechanism component labelled as broken).

The path (A or B) taken depends on the quality, material, and the time in use of said mechanism.

Higher quality mechanism will advance from 1 ~ 3 much slower and will more likely take the A route versus the B route.  No quality mechanism may need replacement almost every year.  Masterwork should probably take about 2 years to go from working to worn and several years to need replacement.  Artifact mechanism would never break, but will still need cleaning (only take A path).

This mean that early on, you must rely on less mechanism (or hand-pick for masterwork to reduce maintenance) else
Mechanism material should also have an effect, for example, steel mechanism should triple the time needed between maintainence/replacement.

Well part of the idea is that you need to sent someone to the mechanism to evalutate the situation. Of course showing/not showing the current state of a mechamism could be an init option.


But perhaps we should look at a good formula to decide how big the chance is a sytem gets into trouble.

something like
chance to break is
if random (400 000,0) <Time since last check^0.5/(quality bonus+material bonus)+other damage
trough this would make repeaters rather prone to break perhaps we somehow could mix that in.
Another option is to instead of adding a chance that it breaks each use make a chance that it "break" each frame and then only breaks when it's used something like
each frame
if random (400 000 0,0) < time since last check/(quilitybonus+material bonus)+other damage